publications

Accepted papers

2026

1. 

   Top-down Traffic Scenario Generation via Joint Initial-Goal Diffusion and Trajectory Infilling

   Lee, Da Saem, Pant, Yash Vardhan, and Fischmeister, Sebastian

   In IEEE International Conference on Intelligent Transportation Systems (ITSC) 2026

   Abs Bib arXiv

   Robust traffic simulators are crucial for developing and testing autonomous vehicles to reduce the costly, labor-intensive real-world data collection process and the need for physical presence on the road. However, existing simulators require agents’ initial states to generate trajectories, which limits scalability and diversity due to restrictions on the given initial states. While data-driven agent initialization has been widely studied, the generated initial states are not interpretable in terms of why the agents are initialized at those specific locations. Given known initial states, trajectory generation is also a challenging problem, as the model must learn the variability of the destination and how agents should reach it over time. In this paper, we propose TrafficDiffuser, a top-down traffic scenario generation framework that generates high-level traffic scenarios, defined by initial and goal state pairs, by jointly modeling them. The high-level scenario generation makes initial states better interpretable and reduces trajectory generation into as simple as an infilling problem. We demonstrate how the generated high-level traffic scenarios can be used, including constraining based on different trajectory modes and integrating them with existing trajectory generation models. We conduct extensive experiments on the Argoverse 2 motion prediction dataset to evaluate how well the generated outputs capture real-world distributions. In addition to generating goal states, TrafficDiffuser outperforms the next-best approach for agent initialization, reducing speed distribution distance by 55.3% and the off-road rate by 2.8%.

   @inproceedings{lee2026topdown, title = {{Top-down Traffic Scenario Generation via Joint Initial-Goal Diffusion and Trajectory Infilling}}, author = {Lee, Da Saem and Pant, Yash Vardhan and Fischmeister, Sebastian}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (ITSC)}, year = {2026}, arxiv = {2608.11407}, img = {ITSC26_trafficdiffuser.png} }

2. 

   Efficient Multi-Objective Planning with Weighted Maximization using Large Neighbourhood Search

   Kalavadia, Krishna, Dutta, Shamak, Pant, Yash Vardhan, and Smith, Stephen L.

   In IEEE International Conference on Robotics and Automation (ICRA) 2026

   Abs Bib arXiv Code

   Autonomous navigation often requires the simultaneous optimization of multiple objectives. The most common approach scalarizes these into a single cost function using a weighted sum, but this method is unable to find all possible trade-offs and can therefore miss critical solutions. An alternative, the weighted maximum of objectives, can find all Pareto-optimal solutions, including those in non-convex regions of the trade-off space that weighted sum methods cannot find. However, the increased computational complexity of finding weighted maximum solutions in the discrete domain has limited its practical use. To address this challenge, we propose a novel search algorithm based on the Large Neighbourhood Search framework that efficiently solves the weighted maximum planning problem. Through extensive simulations, we demonstrate that our algorithm achieves comparable solution quality to existing weighted maximum planners with a runtime improvement of 1-2 orders of magnitude, making it a viable option for autonomous navigation.

   @inproceedings{kalavadia2026wmlns, title = {{Efficient Multi-Objective Planning with Weighted Maximization using Large Neighbourhood Search}}, author = {Kalavadia, Krishna and Dutta, Shamak and Pant, Yash Vardhan and Smith, Stephen L.}, booktitle = {IEEE International Conference on Robotics and Automation (ICRA)}, year = {2026}, arxiv = {2604.04826}, code = {https://github.com/CL2-UWaterloo/WM-LNS}, img = {ICRA26_wmlns.jpg} }

3. 

   Hierarchical Informative Path Planning via Graph Guidance and Trajectory Optimization

   Iskander, Avraiem, Dutta, Shamak, Murrant, Kevin, Pant, Yash Vardhan, and Smith, Stephen L.

   In American Control Conference (ACC) 2026

   Abs Bib arXiv

   We study informative path planning (IPP) with travel budgets in cluttered environments, where an agent collects measurements of a latent field modeled as a Gaussian process (GP) to reduce uncertainty at target locations. Graph-based solvers provide global guarantees but assume pre-selected measurement locations, while continuous trajectory optimization supports path-based sensing but is computationally intensive and sensitive to initialization in obstacle-dense settings. We propose a hierarchical framework with three stages: (i) graph-based global planning, (ii) segment-wise budget allocation using geometric and kernel bounds, and (iii) spline-based refinement of each segment with hard constraints and obstacle pruning. By combining global guidance with local refinement, our method achieves lower posterior uncertainty than graph-only and continuous baselines, while running faster than continuous-space solvers (up to 9x faster than gradient-based methods and 20x faster than black-box optimizers) across synthetic cluttered environments and Arctic datasets.

   @inproceedings{iskander2026ipp, title = {{Hierarchical Informative Path Planning via Graph Guidance and Trajectory Optimization}}, author = {Iskander, Avraiem and Dutta, Shamak and Murrant, Kevin and Pant, Yash Vardhan and Smith, Stephen L.}, booktitle = {American Control Conference (ACC)}, year = {2026}, arxiv = {2601.17227}, img = {ACC26_ipp.png} }

2025

1. 

   Adaptformer: Sequence Models as Adaptive Iterative Planners

   Karthikeyan, Akash, and Pant, Yash Vardhan

   In Conference on Robots and Vision (CRV) 2025

   Abs Bib arXiv

   Despite recent advances in learning-based behavioral planning for autonomous systems, decision-making in multi-task missions remains a challenging problem. For instance, a mission might require a robot to explore an unknown environment, locate the goals, and navigate to them, even if there are obstacles along the way. Such problems are difficult to solve due to: a) sparse rewards, meaning a reward signal is available only once all the tasks in a mission have been satisfied, and b) the agent having to perform tasks at run-time that are not covered in the training data, e.g., demonstrations only from an environment where all doors were unlocked. Consequently, state-of-the-art decision-making methods in such settings are limited to missions where the required tasks are well-represented in the training demonstrations and can be solved within a short planning horizon. To overcome these limitations, we propose Adaptformer, a stochastic and adaptive planner that utilizes sequence models for sample-efficient exploration and exploitation. This framework relies on learning an energy-based heuristic, which needs to be minimized over a sequence of high-level decisions. To generate successful action sequences for long-horizon missions, Adaptformer aims to achieve shorter sub-goals, which are proposed through an intrinsic sub-goal curriculum. Through these two key components, Adaptformer allows for generalization to out-of-distribution tasks and environments, i.e., missions that were not a part of the training data. Empirical results in multiple simulation environments demonstrate the effectiveness of our method. Notably, Adaptformer not only outperforms the state-of-the-art method by up to 25% in multi-goal maze reachability tasks but also successfully adapts to multi-task missions that the state-of-the-art method could not complete, leveraging demonstrations from single-goal-reaching tasks.

   @inproceedings{karthikeyan2025adaptformer, title = {{Adaptformer: Sequence Models as Adaptive Iterative Planners}}, author = {Karthikeyan, Akash and Pant, Yash Vardhan}, booktitle = {Conference on Robots and Vision (CRV)}, year = {2025}, arxiv = {2412.00293}, img = {CRV25_adaptformer.png} }

2. 

   GenPlan: Generative Sequence Models as Adaptive Planners

   Karthikeyan, Akash, and Pant, Yash Vardhan

   In AAAI Conference on Artificial Intelligence 2025

   Abs Bib arXiv Code

   Sequence models have demonstrated remarkable success in behavioral planning by leveraging previously collected demonstrations. However, solving multi-task missions remains a significant challenge, particularly when the planner must adapt to unseen constraints and tasks, such as discovering goals and unlocking doors. Such behavioral planning problems are challenging to solve due to: a) agents failing to adapt beyond the single task learned through their reward function, and b) inability to generalize to new environments, e.g., those with walls and locked doors, when trained only in planar environments. Consequently, state-of-the-art decision-making methods are limited to missions where the required tasks are well-represented in the training demonstrations and can be solved within a short (temporal) planning horizon. To address this, we propose GenPlan: a stochastic and adaptive planner that leverages discrete-flow models for generative sequence modeling, enabling sample-efficient exploration and exploitation. This framework relies on an iterative denoising procedure to generate a sequence of goals and actions. This approach captures multi-modal action distributions and facilitates goal and task discovery, thereby generalizing to out-of-distribution tasks and environments, i.e., missions not part of the training data. We demonstrate the effectiveness of our method through multiple simulation environments. Notably, GenPlan outperforms state-of-the-art methods by over 10% on adaptive planning tasks, where the agent adapts to multi-task missions while leveraging demonstrations from single-goal-reaching tasks. Our code is available at https://github.com/CL2-UWaterloo/GenPlan.

   @inproceedings{karthikeyan2025genplan, title = {{GenPlan: Generative Sequence Models as Adaptive Planners}}, author = {Karthikeyan, Akash and Pant, Yash Vardhan}, booktitle = {AAAI Conference on Artificial Intelligence}, year = {2025}, arxiv = {2412.08565}, code = {https://github.com/CL2-UWaterloo/GenPlan}, img = {AAAI25_genplan.png} }

3. 

   Drive-by-Logic: Trajectory Generation for Nonholonomic Ground Robots with Signal Temporal Logic Objectives

   Kolapalli, Praneeth, Pant, Yash Vardhan, and Smith, Stephen L.

   In IEEE Conference on Control Technology and Applications (CCTA) 2025

   Abs Bib HTML

   Autonomous mobile robots are actively applied to execute complex tasks, such as package delivery, autonomous taxiing, and search-and-rescue. Signal Temporal Logic (STL) offers a powerful formalism for such complex tasks. However, designing plans (trajectories) that satisfy tasks formalized by STL grammar, particularly for nonholonomic systems such as a car-like robot or a fixed-wing aircraft, is a challenging problem. This paper proposes a method to generate trajectories for a multi-robot system with car-like robots to perform complex tasks specified with STL grammar. The proposed method solves a nonlinear program (NLP) to construct trajectories with several constant curvature curves that satisfy the specification. In doing so, it also guarantees the kinematic feasibility of the solution trajectories. Extensive simulation studies show that the proposed method finds satisfying solutions4 \timesfaster than a model predictive control baseline. Additionally, it is able to construct trajectories for complex STL specifications that the baseline fails to satisfy.

   @inproceedings{kolapalli2025drivebylogic, title = {{Drive-by-Logic: Trajectory Generation for Nonholonomic Ground Robots with Signal Temporal Logic Objectives}}, author = {Kolapalli, Praneeth and Pant, Yash Vardhan and Smith, Stephen L.}, booktitle = {IEEE Conference on Control Technology and Applications (CCTA)}, year = {2025}, html = {https://ieeexplore.ieee.org/document/11151389}, img = {placeholder_black.png} }

4. 

   Path Diffuser: Diffusion Model for Data-Driven Traffic Simulator

   Lee, Da Saem, Karthikeyan, Akash, Pant, Yash Vardhan, and Fischmeister, Sebastian

   In IEEE International Conference on Intelligent Transportation Systems (ITSC) 2025

   Abs Bib arXiv Code

   Simulating diverse and realistic traffic scenarios is critical for developing and testing autonomous planning. Traditional rule-based planners lack diversity and realism, while learning-based simulators often replay, forecast, or edit scenarios using historical agent trajectories. However, they struggle to generate new scenarios, limiting scalability and diversity due to their reliance on fully annotated logs and historical data. Thus, a key challenge for a learning-based simulator’s performance is that it requires agents’ past trajectories and pose information in addition to map data, which might not be available for all agents on the road.Without which, generated scenarios often produce unrealistic trajectories that deviate from drivable areas, particularly under out-of-distribution (OOD) map scenes (e.g., curved roads). To address this, we propose Path Diffuser (PD): a two-stage, diffusion model for generating agent pose initializations and their corresponding trajectories conditioned on the map, free of any historical context of agents’ trajectories. Furthermore, PD incorporates a motion primitive-based prior, leveraging Frenet frame candidate trajectories to enhance diversity while ensuring road-compliant trajectory generation. We also explore various design choices for modeling complex multi-agent interactions. We demonstrate the effectiveness of our method through extensive experiments on the Argoverse2 Dataset and additionally evaluate the generalizability of the approach on OOD map variants. Notably, Path Diffuser outperforms the baseline methods by 1.92x on distribution metrics, 1.14x on common-sense metrics, and 1.62x on road compliance from adversarial benchmarks.

   @inproceedings{lee2025pathdiffuser, title = {{Path Diffuser: Diffusion Model for Data-Driven Traffic Simulator}}, author = {Lee, Da Saem and Karthikeyan, Akash and Pant, Yash Vardhan and Fischmeister, Sebastian}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (ITSC)}, year = {2025}, arxiv = {2509.24995}, code = {https://github.com/CL2-UWaterloo/PathDiffuser}, img = {ITSC25_pathdiffuser.png} }

5. 

   Fair-CoPlan: Negotiated Flight Planning with Fair Deconfliction for Urban Air Mobility

   Fronda, Nicole, Smith, Phil, Hoxha, Bardh, Pant, Yash Vardhan, and Abbas, Houssam

   In IEEE International Conference on Intelligent Transportation Systems (ITSC) 2025

   Abs Bib arXiv

   Urban Air Mobility (UAM) is an emerging transportation paradigm in which Uncrewed Aerial Systems (UAS) autonomously transport passengers and goods in cities. The UAS have different operators with different, sometimes competing goals, yet must share the airspace. We propose a negotiated, semi-distributed flight planner that optimizes UAS’ flight lengths \em in a fair manner. Current flight planners might result in some UAS being given disproportionately shorter flight paths at the expense of others. We introduce Fair-CoPlan, a planner in which operators and a Provider of Service to the UAM (PSU) together compute fair flight paths. Fair-CoPlan has three steps: First, the PSU constrains take-off and landing choices for flights based on capacity at and around vertiports. Then, operators plan independently under these constraints. Finally, the PSU resolves any conflicting paths, optimizing for path length fairness. By fairly spreading the cost of deconfliction Fair-CoPlan encourages wider participation in UAM, ensures safety of the airspace and the areas below it, and promotes greater operator flexibility. We demonstrate Fair-CoPlan through simulation experiments and find fairer outcomes than a non-fair planner with minor delays as a trade-off.

   @inproceedings{fronda2025faircoplan, title = {{Fair-CoPlan: Negotiated Flight Planning with Fair Deconfliction for Urban Air Mobility}}, author = {Fronda, Nicole and Smith, Phil and Hoxha, Bardh and Pant, Yash Vardhan and Abbas, Houssam}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (ITSC)}, year = {2025}, arxiv = {2508.14380}, img = {ITSC25_faircoplan.png} }

6. 

   Tractable Stochastic Hybrid Model Predictive Control Using Gaussian Processes for Repetitive Tasks in Unseen Environments

   D’Souza, Leroy, Pant, Yash Vardhan, and Fischmeister, Sebastian

   In European Control Conference (ECC) 2025

   Abs Bib arXiv

   Improving the predictive accuracy of a dynamics model is crucial to obtaining good control performance and safety from Model Predictive Controllers (MPC). One approach involves learning unmodelled (residual) dynamics, in addition to nominal models derived from first principles. Varying residual models across an environment manifest as modes of a piecewise residual (PWR) model that requires a) identifying how modes are distributed across the environment and b) solving a computationally intensive Mixed Integer Nonlinear Program (MINLP) problem for control. We develop an iterative mapping algorithm capable of predicting time-varying mode distributions. We then develop and solve two tractable approximations of the MINLP to combine with the predictor in closed-loop to solve the overall control problem. In simulation, we first demonstrate how the approximations improve performance by 4-18% in comparison to the MINLP while achieving significantly lower computation times (upto 250x faster). We then demonstrate how the proposed mapping algorithm incrementally improves controller performance (upto 3x) over multiple iterations of a trajectory tracking control task even when the mode distributions change over time.

   @inproceedings{dsouza2025hybridmpc, title = {{Tractable Stochastic Hybrid Model Predictive Control Using Gaussian Processes for Repetitive Tasks in Unseen Environments}}, author = {D'Souza, Leroy and Pant, Yash Vardhan and Fischmeister, Sebastian}, booktitle = {European Control Conference (ECC)}, year = {2025}, arxiv = {2508.18203}, img = {ECC25_hybridmpc.png} }

7. 

   Generating Malicious Demonstration Policies to Exploit Vulnerabilities in Inverse Reinforcement Learning

   Alipanah, Arezoo, and Pant, Yash Vardhan

   In Canadian AI Conference 2025

   Abs Bib HTML

   Reinforcement Learning (RL) algorithms depend on well-defined reward functions for policy optimization. Designing such functions is a complex task, even for domain experts. However, valid task demonstrations can still be collected from moderately skilled individuals. This motivates the use of methods such as Imitation Learning (IL) and Inverse Reinforcement Learning (IRL), where an expert provides demonstrations, allowing the algorithm to infer a policy or reward function that aligns with observed behavior. A common assumption in IRL is that demonstrations come from highly skilled experts. While some studies have explored the impact of suboptimal demonstrators, the influence of intentionally malicious demonstrations remains underexplored. This study introduces an adversarial demonstrator framework that systematically perturbs a subset of demonstrations to manipulate the reward function inferred by an IRL algorithm. Additionally, it quantifies the impact of such adversarial manipulations on the learned policy. Our results show that simply altering 10% of the demonstrations can lead the IRL algorithm to learn a faulty reward function, ultimately degrading the performance of the trained policy by up to three times the effect of adding 10% random trajectories to the result.

   @inproceedings{alipanah2025irl, title = {{Generating Malicious Demonstration Policies to Exploit Vulnerabilities in Inverse Reinforcement Learning}}, author = {Alipanah, Arezoo and Pant, Yash Vardhan}, booktitle = {Canadian AI Conference}, year = {2025}, html = {https://caiac.pubpub.org/ai2025}, img = {placeholder_black.png} }

8. 

   Multi-Objective Agent-Based Model Predictive Controller for Plug-and-Play Vehicle Control

   Zhong, Jiaming, Khoshnevisan, Ladan, Huang, Shucheng, Pirani, Mohammad, Pant, Yash Vardhan, and Khajepour, Amir

   IEEE Transactions on Intelligent Transportation Systems 2025

   Abs Bib arXiv

   Functional integration is a growing trend in vehicle control, often involving the coordination of multiple controllers to achieve various objectives simultaneously. The need for flexibility and reliability has led to a "plug-and-play" approach in control system design, which presents challenges for traditional integrated model predictive control (MPC). Agent-based model predictive control (AMPC) has recently emerged as a distributed solution that treats controllers as agents, creating a collaborative framework among them to reach a common goal. However, this approach struggles to manage distributed conflicting objectives when agents are coupled or interdependent. To address this, we propose a novel, practical distributed control scheme called multi-objective AMPC, which adapts the alternating direction method of multipliers (ADMM) into a general control strategy that approximates global optimization while decoupling objectives. We systematically develop three formulations that maintain convergence while addressing control regularization and inequality constraints, applying them to complex vehicle control systems for the first time. The proposed method has been tested on two vehicle control scenarios with a multi-objective topology. Different formulations are compared through simulations, and the most computationally efficient one was implemented on an electric vehicle for real-world evaluations. The results demonstrate that the proposed multi-objective AMPC can converge approximately to the same global optimum as integrated MPC with greater flexibility and the potential to reduce computational costs.

   @article{zhong2025plugplay, title = {{Multi-Objective Agent-Based Model Predictive Controller for Plug-and-Play Vehicle Control}}, author = {Zhong, Jiaming and Khoshnevisan, Ladan and Huang, Shucheng and Pirani, Mohammad and Pant, Yash Vardhan and Khajepour, Amir}, journal = {IEEE Transactions on Intelligent Transportation Systems}, volume = {26}, number = {12}, pages = {22818--22829}, year = {2025}, arxiv = {2609.12108}, img = {ITS25_plugplay.png} }

2024

1. 

   Model Predictive Control for Systems With Partially Unknown Dynamics Under Signal Temporal Logic Specifications

   Dai, Zhao-Feng, Pant, Yash Vardhan, and Smith, Stephen L.

   IEEE Control Systems Letters 2024

   Abs Bib HTML

   In this letter, we design a model predictive controller (MPC) for systems to satisfy Signal Temporal Logic (STL) specifications when the system dynamics are partially unknown, and only a nominal model and past runtime data are available. Our approach uses Gaussian process regression to learn a stochastic, data-driven model of the unknown dynamics, and manages uncertainty in the STL specification resulting from the stochastic model using Probabilistic Signal Temporal Logic (PrSTL). The learned model and PrSTL specification are then used to formulate a chance-constrained MPC. For systems with high control rates, we discuss a modification for improving the solution speed of the control optimization. In simulation case studies, our controller increases the frequency of satisfying the STL specification compared to controllers that use only the nominal dynamics model.

   @article{dai2024lcsl, title = {{Model Predictive Control for Systems With Partially Unknown Dynamics Under Signal Temporal Logic Specifications}}, author = {Dai, Zhao-Feng and Pant, Yash Vardhan and Smith, Stephen L.}, journal = {IEEE Control Systems Letters}, volume = {8}, pages = {2931--2936}, year = {2024}, html = {https://ieeexplore.ieee.org/document/10804141}, img = {placeholder_black.png} }

2. 

   Learning Agent-based Model Predictive Control for Holistic Vehicle Performance

   Zhong, Jiaming, Valiollahi Mehrizi, Reza, Pirani, Mohammad, Chao, Y., Kasaiezadeh, Alireza, Pant, Yash Vardhan, and Khajepour, Amir

   IEEE Transactions on Intelligent Transportation Systems 2024

   Abs Bib arXiv

   Agent-based model predictive control (AMPC) has recently been proposed as a distributed scheme that collaborates with all agents to achieve optimal holistic performance. However, its optimality highly depends on the prediction accuracy that requires all agents or their contributions to be known, which is too idealistic for actual implementation. This research proposes a novel practical hybrid control scheme - learning agent-based MPC (LAMPC), combining the model-based AMPC approach and data-based learning methods to improve the holistic vehicle performance for multi-agent systems. The Gaussian process regression (GPR) enhanced by an online data management strategy serves as the learning core to predict unknown contributions. A novel multi-step prediction mechanism leverages the GPR learning potential along the horizon. The predicted mean, representing the learned unknown contributions, completes the system model in the MPC for more accurate control. Meanwhile, a stochastic framework is formulated to guarantee control safety and feasibility using soft chance constraints based on the prediction variance. Both simulations and experiments show that, with the learning capability, LAMPC outperforms the traditional AMPC. LAMPC can achieve higher tracking performance in well-learned scenarios and always guarantee constraint satisfaction even in less-learned scenarios. Moreover, the proposed hybrid control scheme is efficient for real-time implementation and is flexible to any control agent topology.

   @article{zhong2024holistic, title = {{Learning Agent-based Model Predictive Control for Holistic Vehicle Performance}}, author = {Zhong, Jiaming and Valiollahi Mehrizi, Reza and Pirani, Mohammad and Chao, Y. and Kasaiezadeh, Alireza and Pant, Yash Vardhan and Khajepour, Amir}, journal = {IEEE Transactions on Intelligent Transportation Systems}, volume = {25}, number = {11}, pages = {17482--17492}, year = {2024}, arxiv = {2609.11871}, img = {ITS24_holistic.png} }

3. 

   Enhancing Safety in Mixed Traffic: Learning-Based Modeling and Efficient Control of Autonomous and Human-Driven Vehicles

   Wang, Jie, Pant, Yash Vardhan, Zhao, Lei, Antkiewicz, Michał, and Czarnecki, Krzysztof

   IEEE Transactions on Intelligent Transportation Systems 2024

   Abs Bib arXiv

   With the increasing presence of autonomous vehicles (AVs) on public roads, developing robust control strategies to navigate the uncertainty of human-driven vehicles (HVs) is crucial. This paper introduces an advanced method for modeling HV behavior, combining a first-principles model with Gaussian process (GP) learning to enhance velocity prediction accuracy and provide a measurable uncertainty. We validated this innovative HV model using real-world data from field experiments and applied it to develop a GP-enhanced model predictive control (GP-MPC) strategy. This strategy aims to improve safety in mixed vehicle platoons by integrating uncertainty assessment into distance constraints. Comparative simulation studies with a conventional model predictive control (MPC) approach demonstrated that our GP-MPC strategy ensures more reliable safe distancing and fosters efficient vehicular dynamics, achieving notably higher speeds within the platoon. By incorporating a sparse GP technique in HV modeling and adopting a dynamic GP prediction within the MPC framework, we significantly reduced the computation time of GP-MPC, marking it only 4.6% higher than that of the conventional MPC. This represents a substantial improvement, making the process about 100 times faster than our preliminary work without these approximations. Our findings underscore the effectiveness of learning-based HV modeling in enhancing both safety and operational efficiency in mixed-traffic environments, paving the way for more harmonious AV-HV interactions.

   @article{wang2024enhancing, title = {{Enhancing Safety in Mixed Traffic: Learning-Based Modeling and Efficient Control of Autonomous and Human-Driven Vehicles}}, author = {Wang, Jie and Pant, Yash Vardhan and Zhao, Lei and Antkiewicz, Micha{\l} and Czarnecki, Krzysztof}, journal = {IEEE Transactions on Intelligent Transportation Systems}, volume = {25}, number = {9}, pages = {12784--12799}, year = {2024}, arxiv = {2404.06732}, img = {ITS24_mixedtraffic.jpg} }

4. 

   Learning-Based Modeling of Human-Autonomous Vehicle Interaction for Improved Safety in Mixed-Vehicle Platooning Control

   Wang, Jie, Jiang, Zhihao, and Pant, Yash Vardhan

   Transportation Research Part C: Emerging Technologies 2024

   Abs Bib arXiv

   The rising presence of autonomous vehicles (AVs) on public roads necessitates the development of advanced control strategies that account for the unpredictable nature of human-driven vehicles (HVs). This study introduces a learning-based method for modeling HV behavior, combining a traditional first-principles approach with a Gaussian process (GP) learning component. This hybrid model enhances the accuracy of velocity predictions and provides measurable uncertainty estimates. We leverage this model to develop a GP-based model predictive control (GP-MPC) strategy to improve safety in mixed vehicle platoons by integrating uncertainty assessments into distance constraints. Comparative simulations between our GP-MPC approach and a conventional model predictive control (MPC) strategy reveal that the GP-MPC ensures safer distancing and more efficient travel within the mixed platoon. By incorporating sparse GP modeling for HVs and a dynamic GP prediction in MPC, we significantly reduce the computation time of GP-MPC, making it only marginally longer than standard MPC and approximately 100 times faster than previous models not employing these techniques. Our findings underscore the effectiveness of learning-based HV modeling in enhancing safety and efficiency in mixed-traffic environments involving AV and HV interactions.

   @article{wang2024trc, title = {{Learning-Based Modeling of Human-Autonomous Vehicle Interaction for Improved Safety in Mixed-Vehicle Platooning Control}}, author = {Wang, Jie and Jiang, Zhihao and Pant, Yash Vardhan}, journal = {Transportation Research Part C: Emerging Technologies}, volume = {162}, note = {Art. no. 104600}, year = {2024}, arxiv = {2303.09452}, img = {TRC24_wang.jpg} }

5. 

   Improving Safety in Mixed Traffic: A Learning-based Model Predictive Control for Autonomous and Human-Driven Vehicle Platooning

   Wang, Jie, Jiang, Zhihao, and Pant, Yash Vardhan

   Knowledge-Based Systems 2024

   Abs Bib arXiv

   As autonomous vehicles (AVs) become more common on public roads, their interaction with human-driven vehicles (HVs) in mixed traffic is inevitable. This requires new control strategies for AVs to handle the unpredictable nature of HVs. This study focused on safe control in mixed-vehicle platoons consisting of both AVs and HVs, particularly during longitudinal car-following scenarios. We introduce a novel model that combines a conventional first-principles model with a Gaussian process (GP) machine learning-based model to better predict HV behavior. Our results showed a significant improvement in predicting HV speed, with a 35.64% reduction in the root mean square error compared with the use of the first-principles model alone. We developed a new control strategy called GP-MPC, which uses the proposed HV model for safer distance management between vehicles in the mixed platoon. The GP-MPC strategy effectively utilizes the capacity of the GP model to assess uncertainties, thereby significantly enhancing safety in challenging traffic scenarios, such as emergency braking scenarios. In simulations, the GP-MPC strategy outperformed the baseline MPC method, offering better safety and more efficient vehicle movement in mixed traffic.

   @article{wang2024kbs, title = {{Improving Safety in Mixed Traffic: A Learning-based Model Predictive Control for Autonomous and Human-Driven Vehicle Platooning}}, author = {Wang, Jie and Jiang, Zhihao and Pant, Yash Vardhan}, journal = {Knowledge-Based Systems}, volume = {293}, note = {Art. no. 111673}, year = {2024}, arxiv = {2211.04665}, img = {KBS24_wang.png} }

6. 

   A Data-Driven Distributed Control Scheme: Learning Multi-Objective Agent-Based MPC for Path-Tracking

   Zhong, Jiaming, Valiollahi Mehrizi, Reza, Pant, Yash Vardhan, and Khajepour, Amir

   In IEEE International Conference on Intelligent Transportation Systems (ITSC) 2024

   Abs Bib arXiv

   Agent-based model predictive control (AMPC) has recently been proposed for vehicle systems with various controllers, such as differential braking and torque vectoring, where controllers are regarded as distributed agents contributing to the same objective. However, this scheme is challenging in handling multiple conflicting objectives with coupled agents. A common approach for such tasks is the integrated MPC, where all objectives and agents are stacked together in one optimization. Nevertheless, as more agents and objectives are involved, the integrated MPC will face challenges like computational burdens and maintenance difficulties in practice. To this end, this paper proposes a learning multi-objective AMPC that can improve design flexibility and computing efficiency. First, under the assumption of information exchange, a multi-objective AMPC tailored from the alternating direction method of multipliers (ADMM) is proposed to decouple the system and achieve the same performance as the integrated scheme iteratively. Second, a learning-based method for initializing iterations is proposed to accelerate convergence. In addition, a data management method is proposed for real-time efficiency, and an authentication module is designed for learning reliability. We compare the proposed scheme against the integrated scheme via a combined path-tracking simulation for autonomous vehicles with various controllers. The proposed scheme achieves the same control performance as the integrated one while reducing the computational time by 43.5%. Furthermore, the learning-based method saves 88.6% more computational time than without learning, making it suitable for real-time implementation.

   @inproceedings{zhong2024pathtracking, title = {{A Data-Driven Distributed Control Scheme: Learning Multi-Objective Agent-Based MPC for Path-Tracking}}, author = {Zhong, Jiaming and Valiollahi Mehrizi, Reza and Pant, Yash Vardhan and Khajepour, Amir}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (ITSC)}, year = {2024}, arxiv = {2609.12142}, img = {ITSC24_pathtracking.png} }

7. 

   🏆 Best Paper Award

   Robust Reinforcement Learning for Linear Temporal Logic Specifications with Finite Trajectory Duration

   Mortazavi Moghaddam, Soroush, Pant, Yash Vardhan, and Fischmeister, Sebastian

   In Canadian AI Conference 2024

   Abs Bib HTML

   Linear Temporal Logic (LTL), a formal behavioral specification language, offers a mathematically unambiguous and succinct way to represent operating requirements for a wide variety of Artificial Intelligence (AI) systems, including autonomous and robotic systems. Despite progress, learning policies that reliably satisfy complex LTL specifications in challenging environments remains an open problem. While LTL specifications are evaluated over infinite sequences, this work focuses on solving objectives within a given finite number of steps, as is to be expected in most real-world applications involving robotic or autonomous systems. We study the problem of generating trajectories of a system that satisfy a given LTLf specification in an environment with a priori unknown transition probabilities. Our proposed approach builds upon the popular AlphaGo Zero Reinforcement Learning (RL) framework, which has found great success in the two-player game of Go, to learn policies that can satisfy an LTLf specification given a limit on the trajectory duration. Extensive simulations on complex robot motion planning problems demonstrate that our approach achieves higher success rates in satisfying studied specifications with time constraints compared to state-of-the-art methods. Importantly, our approach succeeds in cases where the baseline method fails to find any satisfying policies.

   @inproceedings{moghaddam2024robustrl, title = {{Robust Reinforcement Learning for Linear Temporal Logic Specifications with Finite Trajectory Duration}}, author = {Mortazavi Moghaddam, Soroush and Pant, Yash Vardhan and Fischmeister, Sebastian}, booktitle = {Canadian AI Conference}, year = {2024}, html = {https://caiac.pubpub.org/pub/xgjwp6n6}, award = {Best Paper Award}, img = {CAIAC24_robustrl.png} }

2023

1. 

   Stochastic Hybrid Model Predictive Control using Gaussian Processes for Systems with Piecewise Residual Dynamics

   D’Souza, Leroy, and Pant, Yash Vardhan

   In American Control Conference (ACC) 2023

   Abs Bib HTML

   Due to their ability to model complex functions and their suitability for control design, Gaussian Processes (GPs) have recently found widespread use to learn residual dynamics that account for the mismatch between the nominal system model and the true underlying system dynamics. However, in cases where the residual dynamics differ drastically over regions of the state/input space, a single GP-based residual model could be inaccurate. When used to design controllers, this could result in controllers with poor performance that could violate state/input constraints. We propose the use of a GP-based hybrid residual dynamics model, which switches between different residual models across regions of the state and input space. We also design a Model Predictive Controller (MPC) that can leverage this hybrid residual dynamics model to ensure probabilistic constraint satisfaction. Through numerical studies, we demonstrate how the proposed controller outperforms a baseline single GP-based MPC baseline. Simulations show a 45% improvement in control performance in the best-case and probability of constraint violations within the desired threshold in contrast to the baseline approach.

   @inproceedings{dsouza2023pwr, title = {{Stochastic Hybrid Model Predictive Control using Gaussian Processes for Systems with Piecewise Residual Dynamics}}, author = {D'Souza, Leroy and Pant, Yash Vardhan}, booktitle = {American Control Conference (ACC)}, year = {2023}, html = {https://ieeexplore.ieee.org/document/10156547}, img = {placeholder_black.png} }

2022

1. 

   Distributed Trajectory Planning for Multi-Rotor UAVs with Signal Temporal Logic Objectives

   Pant, Yash Vardhan, Abbas, Houssam, and Mangharam, Rahul

   In IEEE Conference on Control Technology and Applications (CCTA) 2022

   Abs Bib HTML

   We develop a distributed motion planner for multi-robot systems with Signal Temporal Logic (STL) objectives. Existing approaches to STL-based motion planning are limited to fragments of STL that might not capture all desired behaviors or safety requirements. Focusing on the case of a fleet of multi-rotor Unmanned Aerial Vehicles (UAVs) jointly tasked with satisfying a given STL mission, we develop a distributed method, Fly-by-Distributed-Logic (FBDL), for trajectory planning. The proposed method generates trajectories that maximize the smooth robustness of STL specifications, where the associated optimization is solved in a distributed manner. This, to the best of our knowledge, is the first distributed method that can handle the full grammar of STL (bounded time). Simulation studies show the applicability of our approach to a variety of STL missions, including those with nested STL operators and the Until operator. We also compare its performance to a centralized approach and show that our method is computationally faster, but generally computes trajectories with lower robustness.

   @inproceedings{pant2022ccta, title = {{Distributed Trajectory Planning for Multi-Rotor UAVs with Signal Temporal Logic Objectives}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mangharam, Rahul}, booktitle = {IEEE Conference on Control Technology and Applications (CCTA)}, year = {2022}, html = {https://doi.org/10.1109/CCTA49430.2022.9966096}, img = {placeholder_black.png} }

2. 

   Modeling and Influencing Human Attentiveness in Autonomy-To-Human Perception Hand-offs

   Pant, Yash Vardhan, Thoravi Kumaravel, Balasaravanan, Shah, Ameesh, Kraemer, Erin, Vazquez-Chanlatte, Marcell, Kulkarni, Kshitij, Hartmann, Bjoern, and Seshia, Sanjit A.

   In IEEE International Conference on Intelligent Transportation Systems (ITSC) 2022

   Abs Bib PDF

   It is not uncommon for autonomous systems (e.g., self-driving cars) to require the timely intervention of a human operator to ensure safe operation. It is important to design these systems such that the human is brought into the decision-making loop in a manner that enables them to make a timely and correct decision. In this paper, we consider one such ap-plication, which we refer to as the perception hand-off problem, which brings the driver into the loop when the perception module of an Autonomous Vehicle (AV) is uncertain about the environment. We formalize the perception hand-off problem by designing a Partially Observable Markov Decision Process (POMDP) model. This model captures the latent cognitive state (attention) of the driver which can be influenced through a proposed query-based active information gathering (AIG) system for Human-Machine Interface (HMI). We design a web-based human study to identify the model parameters, and demonstrate the impact of the proposed HMI system. Results from this study show that the state of attentiveness does indeed impact the human performance, and our proposed active information gathering (AIG) actions, i.e., queries to the human driver, result in 7% faster responses from the human. Simulations with the identified POMDP model show that a learnt policy for deploying the AIG actions improves the percentage of correct responses from the human in the perception hand-off by around 5.4%, outperforming other baselines while also using fewer of these actions.

   @inproceedings{pant2022handoff, title = {{Modeling and Influencing Human Attentiveness in Autonomy-To-Human Perception Hand-offs}}, author = {Pant, Yash Vardhan and Thoravi Kumaravel, Balasaravanan and Shah, Ameesh and Kraemer, Erin and Vazquez-Chanlatte, Marcell and Kulkarni, Kshitij and Hartmann, Bjoern and Seshia, Sanjit A.}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (ITSC)}, year = {2022}, pdf = {https://escholarship.org/uc/item/6sd26898}, img = {ITSC22_handoff.png} }

3. 

   Distributed Traffic Flow Management for Uncrewed Aircraft Systems

   Chin, C., Li, Max Z., and Pant, Yash Vardhan

   In IEEE International Conference on Intelligent Transportation Systems (ITSC) 2022

   Abs Bib HTML

   Advanced Air Mobility (AAM) envisions multiple classes of uncrewed aircraft systems (UAS) performing a wide variety of tasks ranging from package delivery to infrastructure inspections. The scale and complexity of future AAM operations require new frameworks for UAS traffic management (UTM), with autonomous and federated air traffic management (ATM) services governing different regions of the AAM airspace. Current UTM solutions are often centralized, and assume that one entity optimizes UAS trajectories and schedules. However, current AAM concepts-of-operations depict competing UAS service suppliers (USS) in charge of ATM services for their own airspace region. We propose a distributed optimization-based approach to UTM, wherein each individual region solves a local traffic flow management problem, and communicates with other regions when necessary. To do so, we design a heuristic that enables distributed UAS traffic flow management, requiring no centralized coordination. We implement this heuristic and test it on a synthetic set of UAS trajectories, operating in an airspace divided into multiple regions. We show that, at a cost of increased delays relative to the centralized solution, our heuristic obtains a feasible solution, even though no singular region has access to trajectory information of other regions.

   @inproceedings{chin2022utm, title = {{Distributed Traffic Flow Management for Uncrewed Aircraft Systems}}, author = {Chin, C. and Li, Max Z. and Pant, Yash Vardhan}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (ITSC)}, year = {2022}, html = {https://doi.org/10.1109/ITSC55140.2022.9922227}, img = {placeholder_black.png} }

4. 

   Stable Interaction of Autonomous Vehicle Platoons with Human-Driven Vehicles

   Pirani, Mohammad, She, Yining, Tang, Renzhi, Jiang, Zhihao, and Pant, Yash Vardhan

   In American Control Conference (ACC) 2022

   Abs Bib

   A necessary prerequisite for the safe interaction of autonomous systems with a human-driven vehicle is for the overall closed-loop system (autonomous systems plus human-driven vehicle) to be stable. This paper studies the safe and stable interaction between a platoon of autonomous vehicles and a set of human-driven vehicles. Considering the longitudinal motion of the vehicles in the platoon, the problem is to ensure a safe emergency braking by the autonomous platoon considering the actions of human-driven vehicles, which may vary based on the driver type. We consider two types of platoon topologies, namely unidirectional and bidirectional. Safe emergency braking is characterized by a specific type of platoon stability, called head-to-tail stability (HTS). We present system-theoretic necessary and sufficient conditions for the combination of the autonomous platoon and human-driven vehicles to be HTS for two platoon control laws, namely the velocity tracking and the platoon formation. Modeling the input-output behavior of each vehicle via a transfer function, the HTS conditions restrict the human-driven vehicles’ transfer functions to have H_∞ norms below certain thresholds. A safe interaction algorithm first identifies the transfer functions of the human-driven vehicles. Then, it tunes the platoon control gains such that the overall system meets HTS conditions. Theoretical results are validated with both experimental data with human subject studies and simulation studies.

   @inproceedings{pirani2022acc, author = {Pirani, Mohammad and She, Yining and Tang, Renzhi and Jiang, Zhihao and Pant, Yash Vardhan}, title = {{Stable Interaction of Autonomous Vehicle Platoons with Human-Driven Vehicles}}, booktitle = {American Control Conference (ACC)}, year = {2022}, img = {ACC2022.png} }

2021

1. 

   Addressing the IEEE AV test challenge with Scenic and VerifAI

   Viswanadha, Kesav, Indaheng, Francis, Wong, Justin, Kim, Edward, Kalvan, Ellen, Pant, Yash, Fremont, Daniel J, and Seshia, Sanjit A

   In 2021 IEEE International Conference on Artificial Intelligence Testing (AITest) 2021

   Abs Bib

   This paper summarizes our formal approach to testing autonomous vehicles (AVs) in simulation for the IEEE AV Test Challenge. We demonstrate a systematic testing framework leveraging our previous work on formally-driven simulation for intelligent cyber-physical systems. First, to model and generate interactive scenarios involving multiple agents, we used SCENIC, a proba bilistic programming language for specifying scenarios. A SCENIC program defines an abstract scenario as a distribution over configurations of physical objects and their behaviors over time. Sampling from an abstract scenario yields many different concrete scenarios which can be run as test cases for the AV. Starting from a SCENIC program encoding an abstract driving scenario, we can use the VERIFAI toolkit to search within the scenario for failure cases with respect to multiple AV evaluation metrics. We demonstrate the effectiveness of our testing framework by identifying concrete failure scenarios for an open-source autopilot, Apollo, starting from a variety of realistic traffic scenarios.

   @inproceedings{viswanadha2021addressing, title = {Addressing the IEEE AV test challenge with Scenic and VerifAI}, author = {Viswanadha, Kesav and Indaheng, Francis and Wong, Justin and Kim, Edward and Kalvan, Ellen and Pant, Yash and Fremont, Daniel J and Seshia, Sanjit A}, booktitle = {2021 IEEE International Conference on Artificial Intelligence Testing (AITest)}, pages = {136--142}, year = {2021}, organization = {IEEE}, img = {AITest2021.png} }

2. 

   DEC-LOS-RRT: Decentralized path planning for multi-robot systems with Line-of-sight constrained communication

   Tuck, Victoria, Pant, Yash Vardhan, Seshia, Sanjit A., and Sastry, S Shankar

   In 2021 IEEE Conference on Control Technology and Applications (CCTA) 2021

   Abs Bib

   Decentralized planning for multi-agent systems,such as fleets of robots in a search-and-rescue operation, is oftenconstrained by limitations on how agents can communicate witheach other. One such limitation is the case when agents cancommunicate with each other only when they are in line-of-sight (LOS). Developing decentralized planning methods thatguarantee safety is difficult in this case, as agents that areoccluded from each other might not be able to communicateuntil it’s too late to avoid a safety violation. In this paper, wedevelop a decentralized planning method that explicitly avoidssituations where lack of visibility of other agents would leadto an unsafe situation. bxuilding on top of an existing Rapidly-exploring Random Tree (RRT)-based approach, our methodguarantees safety at each iteration. Simulation studies showthe effectiveness of our method and compare the degradationin performance with respect to a clairvoyant decentralizedplanning algorithm where agents can communicate despite notbeing in LOS of each other.

   @inproceedings{tuck2021ccta, title = {{DEC-LOS-RRT:} Decentralized path planning for multi-robot systems with Line-of-sight constrained communication}, author = {Tuck, Victoria and Pant, Yash Vardhan and Seshia, Sanjit A. and Sastry, S Shankar}, booktitle = {2021 IEEE Conference on Control Technology and Applications (CCTA)}, year = {2021}, organization = {IEEE}, img = {ccta21.png} }

3. 

   FADS: A Framework for Autonomous Drone Safety Using Temporal Logic-Based Trajectory Planning

   Pant, Yash Vardhan, Li, Max Z., Rodionova, Alena, Quaye, Rhudii A., Abbas, Houssam, Ryerson, Megan, and Mangharam, Rahul

   In Transportation Research, Part C: Emerging Technologies, special issue on Embracing Urban Air Mobility 2021

   Bib

   @inproceedings{pant2019fads, title = {{FADS: A Framework for Autonomous Drone Safety Using Temporal Logic-Based Trajectory Planning}}, author = {Pant, Yash Vardhan and Li, Max Z. and Rodionova, Alena and Quaye, Rhudii A. and Abbas, Houssam and Ryerson, Megan and Mangharam, Rahul}, booktitle = {Transportation Research, Part C: Emerging Technologies, special issue on Embracing Urban Air Mobility}, year = {2021}, img = {trc.png} }

4. 

   Co-design of Control and Planning for Multi-rotor UAVs with Signal Temporal Logic Specifications

   Pant, Yash Vardhan, Yin, He, Arcak, Murat, and Seshia, Sanjit A.

   In American Control Conference (ACC) 2021

   Abs Bib PDF

   Urban Air Mobility (UAM), or the scenario where multiple manned and Unmanned Aerial Vehicles (UAVs) carry out various tasks over urban airspaces, is a transportation concept of the future that is gaining prominence. UAM missions with complex spatial, temporal and reactive requirements can be succinctly represented using Signal Temporal Logic (STL), a behavioral specification language. However, planning and control of systems with STL specifications is computationally intensive, usually resulting in planning approaches that do not guarantee dynamical feasibility, or control approaches that cannot handle complex STL specifications. Here, we present an approach to co-design the planner and control such that a given STL specification (possibly over multiple UAVs) is satisfied with trajectories that are dynamically feasible and our controller can track them with a bounded tracking-error that the planner accounts for. The tracking controller is formulated for the non-linear dynamics of the individual UAVs, and the tracking error bound is computed for this controller when the trajectories satisfy some kinematic constraints. We also augment an existing multi-UAV STL-based trajectory generator in order to generate trajectories that satisfy such constraints. We show that this co-design allows for trajectories that satisfy a given STL specification, and are also dynamically feasible in the sense that they can be tracked with bounded error. The applicability of this approach is demonstrated through simulations of multi-UAV missions.

   @inproceedings{pant2020co, title = {Co-design of Control and Planning for Multi-rotor UAVs with Signal Temporal Logic Specifications}, author = {Pant, Yash Vardhan and Yin, He and Arcak, Murat and Seshia, Sanjit A.}, booktitle = {American Control Conference (ACC)}, year = {2021}, pdf = {https://arxiv.org/pdf/2009.14363.pdf}, img = {galaxy2.png} }

5. 

   Counterexample-Guided Synthesis of Perception Models and Control

   Ghosh, Shromona, Pant, Yash Vardhan, Ravanbakhsh, Hadi, and Seshia, Sanjit A.

   In American Control Conference (ACC) 2021

   Abs Bib

   Recent advances in learning-based perception systems have led to drastic improvements in the performance of robotic systems like autonomous vehicles and surgical robots. These perception systems, however, are hard to analyze and errors in them can propagate to cause catastrophic failures. In this paper, we consider the problem of synthesizing safe and robust controllers for robotic systems which rely on complex perception modules for feedback. We propose a counterexample-guided synthesis framework that iteratively builds simple surrogate models of the complex perception module and enables us to find safe control policies. The framework uses a falsifier to find counterexamples, or traces of the systems that violate a safety property, to extract information that enables efficient modeling of the perception modules and errors in it. These models are then used to synthesize controllers that are robust to errors in perception. If the resulting policy is not safe, we gather new counterexamples. By repeating the process, we eventually find a controller which can keep the system safe even when there is a perception failure. We demonstrate our framework on two scenarios in simulation, namely lane keeping and automatic braking, and show that it generates controllers that are safe, as well as a simpler model of a deep neural network-based perception system that can provide meaningful insight into operations of the perception system.

   @inproceedings{ghosh2021ceg, title = {{Counterexample-Guided Synthesis of Perception Models and Control}}, author = {Ghosh, Shromona and Pant, Yash Vardhan and Ravanbakhsh, Hadi and Seshia, Sanjit A.}, booktitle = {American Control Conference (ACC)}, year = {2021}, img = {CEG.png} }

6. 

   Learning-’N-Flying: A Learning-based, Decentralized Mission Aware UAS Collision Avoidance Scheme

   Rodionova, Alëna, Pant, Yash Vardhan, Kurtz, Connor, Jang, Kuk, Abbas, Houssam, and Mangharam, Rahul

   ACM Transactions on Cyber-Physical Systems 2021

   Abs Bib PDF

   Urban Air Mobility, the scenario where hundreds of manned and Unmanned Aircraft System (UAS) carry out a wide variety of missions (e.g. moving humans and goods within the city), is gaining acceptance as a transportation solution of the future. One of the key requirements for this to happen is safely managing the air traffic in these urban airspaces. Due to the expected density of the airspace, this requires fast autonomous solutions that can be deployed online. We propose Learning-’N-Flying (LNF) a multi-UAS Collision Avoidance (CA) framework. It is decentralized, works on-the-fly and allows autonomous UAS managed by different operators to safely carry out complex missions, represented using Signal Temporal Logic, in a shared airspace. We initially formulate the problem of predictive collision avoidance for two UAS as a mixed-integer linear program, and show that it is intractable to solve online. Instead, we first develop Learning-to-Fly (L2F) by combining: a) learning-based decision-making, and b) decentralized convex optimization-based control. LNF extends L2F to cases where there are more than two UAS on a collision path. Through extensive simulations, we show that our method can run online (computation time in the order of milliseconds), and under certain assumptions has failure rates of less than 1% in the worst-case, improving to near 0% in more relaxed operations. We show the applicability of our scheme to a wide variety of settings through multiple case studies.

   @article{rodionova2021learning, title = {Learning-'N-Flying: A Learning-based, Decentralized Mission Aware UAS Collision Avoidance Scheme}, author = {Rodionova, Al{\"e}na and Pant, Yash Vardhan and Kurtz, Connor and Jang, Kuk and Abbas, Houssam and Mangharam, Rahul}, journal = {ACM Transactions on Cyber-Physical Systems}, year = {2021}, pdf = {https://arxiv.org/abs/2101.10404}, selected = {true}, img = {TCPS21.png} }

2020

1. 

   Learning-to-Fly: Learning-based Collision Avoidance for Scalable Urban Air Mobility

   Rodionova, Alëna*, Pant, Yash Vardhan*, Jang, Kuk J., Abbas, Houssam, and Mangharam, Rahul

   In IEEE International Conference on Intelligent Transportation Systems 2020

   Abs Bib PDF

   With increasing urban population, there is global interest in Urban Air Mobility (UAM), where hundreds of autonomous Unmanned Aircraft Systems (UAS) execute missions in the airspace above cities. Unlike traditional human-in-the-loop air traffic management, UAM requires decentralized autonomous approaches that scale for an order of magnitude higher aircraft densities and are applicable to urban settings. We present Learning-to-Fly (L2F), a decentralized on-demand airborne collision avoidance framework for multiple UAS that allows them to independently plan and safely execute missions with spatial, temporal and reactive objectives expressed using Signal Temporal Logic. We formulate the problem of predictively avoiding collisions between two UAS without violating mission objectives as a Mixed Integer Linear Program (MILP). This however is intractable to solve online. Instead we develop L2F, a two-stage collision avoidance method that consists of: 1)a learning-based decision making scheme and 2) a distributed, linear programming-based UAS control algorithm. Through extensive simulations, we show the real-time applicability of our method which is ≈ 6000× faster than the MILP approach and can resolve 99.4% of collisions when there is ample room to maneuver, and shows graceful degradation in performance otherwise. We also compare L2F to two other methods and demonstrate an implementation on quad-rotor robots.

   @inproceedings{alena2020l2f, title = {{Learning-to-Fly: Learning-based Collision Avoidance for Scalable Urban Air Mobility}}, author = {Rodionova, Alëna* and Pant, Yash Vardhan* and Jang, Kuk J. and Abbas, Houssam and Mangharam, Rahul}, booktitle = {IEEE International Conference on Intelligent Transportation Systems}, year = {2020}, pdf = {https://arxiv.org/abs/2006.13267}, img = {l2f.png} }

2. 

   Learning-to-Fly RL: Reinforcement Learning-based Collision Avoidance for Scalable Urban Air Mobility

   Jang, Kuk J., Pant, Yash Vardhan, Rodionova, Alena, and Mangharam, Rahul

   In AIAA/IEEE Digital Avionics Systems Conference (DASC) 2020

   Abs Bib

   As hundreds of Unmanned Aircraft System (UAS) operate within urban airspaces, automated and decentralized UAS traffic management (UTM) will be critical to maintain safe and efficient operations. In this work, we present Learning-to-Fly with Reinforcement Learning (L2F-RL), a decentralized, on-demand Collision Avoidance (CA) framework that systematically combines machine learning with cooperative model predictive control for UAS collision avoidance while retaining satisfaction of higher-level mission objectives. L2F-RL consists of: 1) RL-based policy for conflict resolution (CR) with discrete-decision making, 2) decentralized, cooperative model predictive control for CA. To accelerate training with RL, we utilize reward shaping and curriculum learning. Our approach outperforms baseline approaches with a 99.01% separation rate (ratio of success to total test cases) in the worst case, improving to 100% in the best case with a 1000X improvement in computation time compared to centralized methods. Our results demonstrate the potential of combining learning approaches with optimization-based control, making it a significant contribution towardsscalable, decentralized UTM.

   @inproceedings{jang2020dasc, title = {{Learning-to-Fly RL: Reinforcement Learning-based Collision Avoidance for Scalable Urban Air Mobility}}, author = {Jang, Kuk J. and Pant, Yash Vardhan and Rodionova, Alena and Mangharam, Rahul}, booktitle = {AIAA/IEEE Digital Avionics Systems Conference (DASC)}, year = {2020}, img = {l2f_kj1.png} }

3. 

   Formal Scenario-Based Testing of Autonomous Vehicles: From Simulation to the Real World

   Fremont, Daniel J, Kim, Edward, Pant, Yash Vardhan, Seshia, Sanjit A, Acharya, Atul, Bruso, Xantha, Wells, Paul, Lemke, Steve, Yu, Qiang, and Mehta, Shalin

   In IEEE International Conference on Intelligent Transportation Systems (to appear) 2020

   Abs Bib PDF

   We present a new approach to automated scenario-based testing of the safety of autonomous vehicles, especially those using advanced artificial intelligence-based components, spanning both simulation-based evaluation as well as testing in the real world. Our approach is based on formal methods, combining formal specification of scenarios and safety properties, algorithmic test case generation using formal simulation, test case selection for track testing, executing test cases on the track, and analyzing the resulting data. Experiments with a real autonomous vehicle at an industrial testing ground support our hypotheses that (i) formal simulation can be effective at identifying test cases to run on the track, and (ii) the gap between simulated and real worlds can be systematically evaluated and bridged.

   @inproceedings{fremont2020formal, title = {{Formal Scenario-Based Testing of Autonomous Vehicles: From Simulation to the Real World}}, author = {Fremont, Daniel J and Kim, Edward and Pant, Yash Vardhan and Seshia, Sanjit A and Acharya, Atul and Bruso, Xantha and Wells, Paul and Lemke, Steve and Yu, Qiang and Mehta, Shalin}, booktitle = {IEEE International Conference on Intelligent Transportation Systems (to appear)}, year = {2020}, pdf = {https://arxiv.org/pdf/2003.07739.pdf}, img = {itsc20.png} }

4. 

   Anytime Computation and Control for Autonomous Systems

   Pant, Yash Vardhan, Abbas, Houssam, Mohta, Kartik, Quaye, Rhudii A., Nghiem, Truong X., Devietti, Joseph, and Mangharam, Rahul

   IEEE Transactions on Control Systems Technology 2020

   Abs Bib PDF

   The correct and timely completion of the sensing and action loop is of utmost importance in safety critical autonomous systems. Crucial to the performance of this feedback control loop are the computation time and accuracy of the estimator which produces state estimates used by the controller. These state estimators often use computationally expensive perception algorithms like visual feature tracking. With on-board computers on autonomous robots being computationally limited, the computation time of such an estimation algorithm can at times be high enough to result in poor control performance. We develop a framework for codesign of anytime estimation and robust control algorithms, taking into account computation delays and estimation inaccuracies. This is achieved by constructing an anytime estimator from an off-the-shelf perception-based estimation algorithm and obtaining a trade-off curve for its computation time versus estimation error. This is used in the design of a robust predictive control algorithm that at run-time decides a contract, or operation mode, for the estimator in addition to controlling the dynamical system to meet its control objectives at a reduced computation energy cost. This codesign provides a mechanism through which the controller can use the tradeoff curve to reduce estimation delay at the cost of higher inaccuracy, while guaranteeing satisfaction of control objectives. Experiments on a hexrotor platform running a visual-based algorithm for state estimation show how our method results in up to a 10% improvement in control performance while simultaneously saving 5%-6% in computation energy as compared to a method that does not leverage the codesign.

   @article{pant2020tcst, title = {{Anytime Computation and Control for Autonomous Systems}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mohta, Kartik and Quaye, Rhudii A. and Nghiem, Truong X. and Devietti, Joseph and Mangharam, Rahul}, journal = {IEEE Transactions on Control Systems Technology}, year = {2020}, pdf = {TCST20.pdf}, selected = {true}, img = {TCST20.png} }

2019

1. 

   Robustness Evaluation of Computer-aided Clinical trials for Medical Devices

   Jang, Kuk Jin, Pant, Yash Vardhan, Zhang, Bo, Weimer, James, and Mangharam, Rahul

   In Proceedings of the ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) 2019

   Abs Bib PDF

   Medical cyber-physical systems, such as the implantable cardioverter defibrillator (ICD), require evaluation of safety and efficacy in the context of a patient population in a clinical trial. Advances in computer modeling and simulation allow for generation of a simulated cohort or virtual cohort which mimics a patient population and can be used as a source of prior information. A major obstacle to acceptance of simulation results as a source of prior information is the lack of a framework for explicitly modeling sources of uncertainty in simulation results and quantifying the effect on trial outcomes. In this work, we formulate the Computer-Aided Clinical Trial (CACT) within a Bayesian statistical framework allowing explicit modeling of assumptions and utilization of simulation results at all stages of a clinical trial. To quantify the robustness of the CACT outcome with respect to a simulation assumption, we define δ-robustness as the minimum perturbation of the base prior distribution resulting in a change of the CACT outcome and provide a method to estimate the δ-robustness. We demonstrate the utility of the framework and how the results of δ-robustness evaluation can be utilized at various stages of a clinical trial through an application to the Rhythm ID Goes Head-to-head Trial (RIGHT), which was a comparative evaluation of the safety and efficacy of specific software algorithms across different implantable cardiac devices. Finally, we introduce a hardware interface that allows for direct interaction with the physical device in order to validate and confirm the results of a CACT for implantable cardiac devices.

   @inproceedings{jang2019iccps, title = {{Robustness Evaluation of Computer-aided Clinical trials for Medical Devices}}, author = {Jang, Kuk Jin and Pant, Yash Vardhan and Zhang, Bo and Weimer, James and Mangharam, Rahul}, booktitle = {Proceedings of the ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS)}, year = {2019}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1137&context=mlab_papers}, img = {cact.png} }

2. 

   Temporal Logic Robustness for General Signal Classes

   Abbas, Houssam, Pant, Yash Vardhan, and Mangharam, Rahul

   In Proceedings of the ACM International Conference on Hybrid Systems: Computation and Control (HSCC) 2019

   Abs Bib PDF

   In multi-agent systems, robots transmit their planned trajectories to each other or to a central controller, and each receiver plans its own actions by maximizing a measure of mission satisfaction. For missions expressed in temporal logic, the robustness function plays the role of satisfaction measure. Currently, a Piece-Wise Linear (PWL) or piece-wise constant reconstruction is used at the receiver. This allows an efficient robustness computation algorithm - a.k.a. monitoring - but is not adaptive to the signal class of interest, and does not leverage the compression properties of more general representations. When communication capacity is at a premium, this is a serious bottleneck. In this paper we first show that the robustness computation is significantly affected by how the continuous-time signal is reconstructed from the received samples, which can mean the difference between a successful control and a crash. We show that monitoring general spline-based reconstructions yields a smaller robustness error, and that it can be done with the same time complexity as monitoring the simpler PWL reconstructions. Thus robustness computation can now be adapted to the signal class of interest. We further show that the monitoring error is tightly upper-bounded by the L ∞ signal reconstruction error. We present a (non-linear) L ∞ -based scheme which yields even lower monitoring error than the spline-based schemes (which have the advantage of being faster to compute), and illustrate all results on two case studies. As an application of these results, we show how time-frequency specifications can be efficiently monitored online.

   @inproceedings{abbas2019hscc, title = {{Temporal Logic Robustness for General Signal Classes}}, author = {Abbas, Houssam and Pant, Yash Vardhan and Mangharam, Rahul}, booktitle = {{Proceedings of the ACM International Conference on Hybrid Systems: Computation and Control (HSCC)}}, year = {2019}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1138&context=mlab_papers}, img = {hscc19.png} }

3. 

   Fly-by-Logic: A Tool for Unmanned Aircraft System Fleet Planning using Temporal Logic

   Pant, Yash Vardhan, Quaye, Rhudii A., Abbas, Houssam, Varre, Akarsh, and Mangharam, Rahul

   In NASA Formal Methods Symposium 2019

   Abs Bib PDF Code

   Safe planning for fleets of Unmaned Aircraft Systems (UAS) performing complex missions in urban environments has typically been a challenging problem. In the United States of America, the National Aeronautics and Space Administration (NASA) and the Federal Aviation Administration (FAA) have been studying the regulation of the airspace when multiple such fleets of autonomous UAS share the same airspace, outlined in the Concept of Operations document (ConOps). While the focus is on the infrastructure and management of the airspace, the Unmanned Aircraft System (UAS) Traffic Management (UTM) ConOps also outline a potential airspace reservation based system for operation where operators reserve a volume of the airspace for a given time inter- val to operate in, but it makes clear that the safety (separation from other aircraft, terrain, and other hazards) is a responsibility of the drone fleet operators. In this work, we present a tool that allows an operator to plan out missions for fleets of multi-rotor UAS, performing complex time- bound missions. The tool builds upon a correct-by-construction planning method by translating missions to Signal Temporal Logic (STL). Along with a simple user interface, it also has fast and scalable mission planning abilities. We demonstrate our tool for one such mission.

   @inproceedings{pant2019tool, title = {{Fly-by-Logic: A Tool for Unmanned Aircraft System Fleet Planning using Temporal Logic}}, author = {Pant, Yash Vardhan and Quaye, Rhudii A. and Abbas, Houssam and Varre, Akarsh and Mangharam, Rahul}, booktitle = {NASA Formal Methods Symposium}, year = {2019}, organization = {NASA}, code = {https://github.com/yashpant/FlyByLogic}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1139&context=mlab_papers}, img = {nasafm.png} }

2018

1. 

   Design and implementation of a centralized system for autonomous unmanned aerial vehicle trajectory conflictresolution

   Li, Max Z., Tan, William R., Prakash, Sahithya M., Kennedy, John F., Ryerson, Megan S., Lee, Daewon, and Pant, Yash Vardhan

   In IEEE National Aerospace and Electronics Conference (NAECON) 2018

   Abs Bib PDF

   The ability of autonomous unmanned aerial vehicles(aUAVs) to carry out a diverse range of missions presentsenormous societal and economic benefits. However, one criticalsafety issue that must be addressed prior to large-scale aUAVadoption is the potential for aUAV-on-aUAV trajectory conflicts.Current decentralized methods for trajectory conflict resolutionare energy and resource intensive, rendering them difficult toscale and implement. We propose a centralized, infrastructure-to-vehicle (I2V) approach that leverages modular embedded systems(“control modules”). A voxel-based airspace reservation systemis maintained by the control module, which interrogates andretrieves the stored trajectories from aUAVs approaching thereservation airspace. Trajectory conflict search is performed by the control module, and conflict resolution protocols aretransmitted back to the aUAVs. Our communication protocolis Wi-Fi-based and leverages a lightweight on-board ODROIDmicrocontroller; the setup is agnostic to the specific aUAVplatform and firmware. We simulate various airspace situationsand test the resiliency of our reservation system in MATLAB. Wealso leverage two aUAV testbeds to demonstrate the performanceof our centralized air traffic control regime.

   @inproceedings{li2018naecon, title = {{Design and implementation of a centralized system for autonomous unmanned aerial vehicle trajectory conflictresolution}}, author = {Li, Max Z. and Tan, William R. and Prakash, Sahithya M. and Kennedy, John F. and Ryerson, Megan S. and Lee, Daewon and Pant, Yash Vardhan}, booktitle = {IEEE National Aerospace and Electronics Conference (NAECON)}, year = {2018}, pdf = {https://ieeexplore.ieee.org/document/8556719}, img = {max18.png} }

2. 

   Fly-by-Logic: Control of Multi-Drone Fleets with Temporal Logic Objectives

   Pant, Yash Vardhan, Abbas, Houssam, Quaye, Rhudii A, and Mangharam, Rahul

   In Proceedings of the ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) 2018

   Abs Bib PDF Code

   The problem of safe planning and control for multi- drone systems across a variety of missions is of critical impor- tance, as the scope of tasks assigned to such systems increases. In this paper, we present an approach to solve this problem for multi-quadrotor missions. Given a mission expressed in Signal Temporal Logic (STL), our controller maximizes robustness to generate trajectories for the quadrotors that satisfy the STL spec- ification in continuous-time. We also show that the constraints on our optimization guarantees that these trajectories can be tracked nearly perfectly by lower level off-the-shelf position and attitude controllers. Our approach avoids the oversimplifying abstractions found in many planning methods, while retaining the expressiveness of missions encoded in STL allowing us to handle complex spatial, temporal and reactive requirements. Through experiments, both in simulation and on actual quadrotors, we show the performance, scalability and real-time applicability of our method.

   @inproceedings{pant2018fly, title = {{Fly-by-Logic: Control of Multi-Drone Fleets with Temporal Logic Objectives}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Quaye, Rhudii A and Mangharam, Rahul}, booktitle = {Proceedings of the ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS)}, pages = {186--197}, year = {2018}, organization = {IEEE Press}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1127&context=mlab_papers}, code = {https://github.com/yashpant/FlyByLogic}, selected = {true}, img = {FbL_iccps.png} }

2017

1. 

   Smooth Operator: Control using the Smooth Robustness of Temporal Logic

   Pant, Yash Vardhan*, Abbas, Houssam*, and Mangharam, Rahul

   In IEEE Conference on Control Technology and Applications (CCTA) 2017

   Abs Bib PDF Code

   Modern control systems, like controllers for swarms of quadrotors, must satisfy complex control objectives while withstanding a wide range of disturbances, from bugs in their software to attacks on their sensors and changes in their environments. These requirements go beyond stability and tracking, and involve temporal and sequencing constraints on system response to various events. This work formalizes the requirements as formulas in Metric Temporal Logic (MTL), and designs a controller that maximizes the robustness of the MTL formula. Formally, if the system satisfies the formula with robustness r, then any disturbance of size less than r cannot cause it to violate the formula. Because robustness is not differentiable, this work provides arbitrarily precise, infinitely differentiable, approximations of it, thus enabling the use of powerful gradient descent optimizers. Experiments on a temperature control example and a two-quadrotor system demonstrate that this approach to controller design outperforms existing approaches to robustness maximization based on Mixed Integer Linear Programming and stochastic heuristics. Moreover, it is not constrained to linear systems.

   @inproceedings{pant2017smooth, title = {{Smooth Operator: Control using the Smooth Robustness of Temporal Logic}}, author = {Pant, Yash Vardhan* and Abbas, Houssam* and Mangharam, Rahul}, booktitle = {IEEE Conference on Control Technology and Applications (CCTA)}, pages = {1235--1240}, year = {2017}, organization = {IEEE}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1119&context=mlab_papers}, code = {https://github.com/yashpant/SmoothOperator0}, img = {sop.png} }

2016

1. 

   Robust Model Predictive Control for Non-Linear Systems with Input and State Constraints via Feedback Linearization

   Pant, Yash Vardhan, Abbas, Houssam, and Mangharam, Rahul

   In IEEE Conference on Decision and Control (CDC) 2016

   Abs Bib PDF

   Robust predictive control of non-linear systems under state estimation errors and input and state constraints is a challenging problem, and solutions to it have generally involved solving computationally hard non-linear optimizations. Feedback linearization has reduced the computational burden, but has not yet been solved for robust model predictive control under estimation errors and constraints. In this paper, we solve this problem of robust control of a non-linear system under bounded state estimation errors and input and state constraints using feedback linearization. We do so by developing robust constraints on the feedback linearized system such that the non-linear system respects its constraints. These constraints are computed at run-time using online reachability, and are linear in the optimization variables, resulting in a Quadratic Program with linear constraints. We also provide robust feasibility, recursive feasibility and stability results for our control algorithm. We evaluate our approach on two systems to show its applicability and performance.

   @inproceedings{pant2016robust, title = {{Robust Model Predictive Control for Non-Linear Systems with Input and State Constraints via Feedback Linearization}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mangharam, Rahul}, booktitle = {IEEE Conference on Decision and Control (CDC)}, pages = {5694--5699}, year = {2016}, organization = {IEEE}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1113&context=mlab_papers}, img = {cdc16.png} }

2015

1. 

   Co-design of Anytime Computation and Robust Control

   Pant, Yash Vardhan, Abbas, Houssam, Mohta, Kartik, Nghiem, Truong X., Devietti, Joseph, and Mangharam, Rahul

   In IEEE Real-Time Systems Symposium (RTSS) 2015

   Abs Bib PDF

   Control software of autonomous robots has stringent real-time requirements that must be met to achieve the control objectives. One source of variability in the performanceof a control system is the execution time and accuracy ofthe state estimator that provides the controller with stateinformation. This estimator is typically perception-based (e.g.,Computer Vision-based) and is computationally expensive. When the computational resources of the hardware platformbecome overloaded, the estimation delay can compromise con-trol performance and even stability. In this paper, we define a framework for co-designing anytime estimation and controlalgorithms, in a manner that accounts for implementation issues like delays and inaccuracies. We construct an anytimeperception-based estimator from standard off-the-shelf Com-puter Vision algorithms, and show how to obtain a trade-offcurve for its delay vs estimate error behavior. We use this anytime estimator in a controller that can use this trade-off curve at runtime to achieve its control objectives at areduced energy cost. When the estimation delay is too largefor correct operation, we provide an optimal manner in which the controller can use this curve to reduce estimation delay at the cost of higher inaccuracy, all the while guaranteeing basic objectives are met. We illustrate our approach on anautonomous hexrotor and demonstrate its advantage over a system that does not exploit co-design.

   @inproceedings{pant2015co, title = {{Co-design of Anytime Computation and Robust Control}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mohta, Kartik and Nghiem, Truong X. and Devietti, Joseph and Mangharam, Rahul}, booktitle = {IEEE Real-Time Systems Symposium (RTSS)}, pages = {43--52}, year = {2015}, organization = {IEEE}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1110&context=mlab_papers}, img = {rtss15_2.png} }

2. 

   Power-efficient Algorithms for Autonomous Navigation

   Pant, Yash Vardhan, Abbas, Houssam, Nischal, KN, Kelkar, Paritosh, Kumar, Dhruva, Devietti, Joseph, and Mangharam, Rahul

   In International Conference on Complex Systems Engineering (ICCSE) 2015

   Abs Bib PDF

   Real-time navigation of autonomous vehicles requires the processing of a large amount of sensor data by the perception algorithms onboard the vehicle, like object detection and localization. To meet the driving performance and safety requirements, these algorithms require the hardware to be over-engineered to always operate for the worst-case. This leads to excessive power consumption by the computation platform. In this paper, we study how platform-level optimizations affect the computation throughput and power, and how to use this trade-off to save computation power without overly degrading throughput and control performance. The approach uses an offline profiling stage of the perception algorithm, which gives us Throughput versus Power curves for various processor frequencies and various scheduling of the perception code on CPU and GPU. At runtime, we combine power and throughput into one objective function, and design a supervisor what will determine the frequency and CPU/GPU allocation to maximize the objective. We illustrate our approach on a scaled-down autonomous car which uses Vanishing Point navigation. Experimental results demonstrate that we can achieve an energy savings of upto 20% while degrading control performance by less than 1%.

   @inproceedings{pant2015power, title = {{Power-efficient Algorithms for Autonomous Navigation}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Nischal, KN and Kelkar, Paritosh and Kumar, Dhruva and Devietti, Joseph and Mangharam, Rahul}, booktitle = {International Conference on Complex Systems Engineering (ICCSE)}, pages = {1--6}, year = {2015}, organization = {IEEE}, pdf = {ICCSE15.pdf}, img = {iccse15.png} }

2014

1. 

   Peak Power Reduction in Hybrid Energy Systems with Limited Load Forecasts

   Pant, Yash Vardhan, Nghiem, Truong X, and Mangharam, Rahul

   In American Control Conference (ACC) 2014

   Abs Bib PDF

   Hybrid energy systems, which consist of a load powered by a source and a form of energy storage, find applications in many systems, e.g., the electric grid and electric vehicles. A key problem for hybrid energy systems is the reduction of peak power consumption to ensure cost-efficient operation as peak power draws require additional resources and adversely affect the system reliability and lifetime. Furthermore, in some cases such as electric vehicles, the load dynamics are fast, not perfectly known in advance and the on-board computation power is often limited, making the implementation of traditional optimal control difficult. We aim to develop a control scheme to reduce the peak power drawn from the source for hybrid energy systems with limited computation power and limited load forecasts. We propose a scheme with two control levels and provide a sufficient condition for control of the different energy storage/generation components to meet the instantaneous load while satisfying a peak power threshold. The scheme provides performance comparable to Model Predictive Control, while requiring less computation power and only coarse-grained load predictions. For a case study, we implement the scheme for a battery-supercapacitor-powered electric vehicle with real world drive cycles to demonstrate the low execution time and effective reduction of the battery power (hence temperature), which is crucial to the lifetime of the battery.

   @inproceedings{pant2014peak, title = {{Peak Power Reduction in Hybrid Energy Systems with Limited Load Forecasts}}, author = {Pant, Yash Vardhan and Nghiem, Truong X and Mangharam, Rahul}, booktitle = {American Control Conference (ACC)}, pages = {4212--4217}, year = {2014}, organization = {IEEE}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1084&context=mlab_papers}, img = {acc14.png} }

2011

1. 

   Autoplug: An automotive test-bed for electronic controller unit testing and verification

   Drolia, Utsav*, Wang, Zhenyan*, Pant, Yash Vardhan*, and Mangharam, Rahul

   In IEEE International Conference on Intelligent Transportation Systems 2011

   Abs Bib PDF

   In 2010, over 20.3 million vehicles were recalled. Software issues related to automotive controls such as cruise control, anti-lock braking system, traction control and stability control, account for an increasingly large percentage of the overall vehicles recalled. There is a need for new and scalable methods to evaluate automotive controls in a realistic and open setting. We have developed AutoPlug, an automotive Electronic Controller Unit (ECU) test-bed to diagnose, test, update and verify controls software. AutoPlug consists of multiple ECUs interconnected by a CAN bus, a race car driving simulator which behaves as the plant model and a vehicle controls monitor in Matlab. As the ECUs drive the simulated vehicle, the physicsbased simulation provides feedback to the controllers in terms of acceleration, yaw, friction and vehicle stability. This closedloop platform is then used to evaluate multiple vehicle control software modules such as traction, stability and cruise control. With this test-bed we highlight approaches for runtime ECU software diagnosis and testing of the stability and performance of the vehicle. Code updates can be executed via a smart phone so drivers may remotely “patch” their vehicle. This closedloop automotive control test-bed allows the automotive research community to explore the capabilities and challenges of safe and secure remote code updates for vehicle recalls management.

   @inproceedings{drolia2011autoplug, title = {{Autoplug: An automotive test-bed for electronic controller unit testing and verification}}, author = {Drolia, Utsav* and Wang, Zhenyan* and Pant, Yash Vardhan* and Mangharam, Rahul}, booktitle = {IEEE International Conference on Intelligent Transportation Systems}, pages = {1187--1192}, year = {2011}, organization = {IEEE}, pdf = {https://repository.upenn.edu/cgi/viewcontent.cgi?article=1045&context=mlab_papers}, img = {itsc11.png} }

Demo abstracts, Workshop papers and Works-in-Progress

2025

1. 

   DiffFP: Learning Behaviors from Scratch via Diffusion-based Fictitious-Play

   Karthikeyan, Akash, and Pant, Yash Vardhan

   In IJCAI 2025 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems 2025

   Abs Bib

   Self-play reinforcement learning has demonstrated significant success in learning complex strategic and interactive behaviors in competitive multi-agent games. However, achieving such behaviors in continuous decision spaces remains challenging. Ensuring adaptability and generalization in self-play settings is critical for achieving competitive performance in dynamic multi-agent environments. These challenges often cause methods to converge slowly or fail to converge at all to a Nash equilibrium, making agents vulnerable to strategic exploitation by unseen opponents. To address these challenges, we propose DiffFP, a fictitious play (FP) framework that estimates the best response to unseen opponents while learning a robust and multimodal behavioral policy. Specifically, we approximate the best response using a diffusion policy that leverages generative modeling to learn adaptive and diverse strategies. Through empirical evaluation, we demonstrate that the proposed FP framework converges towards ε-Nash equilibria in continuous- space zero-sum games. We validate our method on complex multi-agent environments, including racing and multi-particle zero-sum games. Simulation results show that the learned policies are robust against diverse opponents and outperform baseline reinforcement learning policies. Our approach achieves up to 3\times faster convergence and 30\times higher success rates on average against RL-based baselines, demonstrating its robustness to opponent strategies and stability across training iterations

   @inproceedings{karthikeyan2025difffp, title = {{DiffFP: Learning Behaviors from Scratch via Diffusion-based Fictitious-Play}}, author = {Karthikeyan, Akash and Pant, Yash Vardhan}, booktitle = {{IJCAI 2025 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems}}, year = {2025}, img = {placeholder_black.png} }

2. 

   Learning Adaptive Diffusion Policies for Hybrid Dynamical Systems

   D’Souza, Leroy, Karthikeyan, Akash, Pant, Yash Vardhan, and Fischmeister, Sebastian

   In IJCAI 2025 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems 2025

   Bib

   @inproceedings{dsouza2025adaptivediffusion, title = {{Learning Adaptive Diffusion Policies for Hybrid Dynamical Systems}}, author = {D'Souza, Leroy and Karthikeyan, Akash and Pant, Yash Vardhan and Fischmeister, Sebastian}, booktitle = {{IJCAI 2025 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems}}, year = {2025}, img = {placeholder_black.png} }

3. 

   Exploiting Demonstration Vulnerabilities: Malicious Agents in Inverse Reinforcement Learning

   Alipanah, Arezoo, and Pant, Yash Vardhan

   In Cybersecurity and Privacy Institute (CPI) Graduate Student Conference 2025

   Bib

   @inproceedings{alipanah2025exploiting, title = {{Exploiting Demonstration Vulnerabilities: Malicious Agents in Inverse Reinforcement Learning}}, author = {Alipanah, Arezoo and Pant, Yash Vardhan}, booktitle = {{Cybersecurity and Privacy Institute (CPI) Graduate Student Conference}}, year = {2025}, img = {placeholder_black.png} }

2024

1. 

   Learning-Based Hybrid Model Predictive Control for Systems With Partially Known and Evolving Dynamics

   D’Souza, Leroy, Pant, Yash Vardhan, and Fischmeister, Sebastian

   In AAAI 2024 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems 2024

   Bib

   @inproceedings{dsouza2024evolving, title = {{Learning-Based Hybrid Model Predictive Control for Systems With Partially Known and Evolving Dynamics}}, author = {D'Souza, Leroy and Pant, Yash Vardhan and Fischmeister, Sebastian}, booktitle = {{AAAI 2024 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems}}, year = {2024}, img = {placeholder_black.png} }

2. 

   AdaptFormer: Sequence Models as Adaptive Iterative Planners

   Karthikeyan, Akash, and Pant, Yash Vardhan

   In AAAI 2024 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems 2024

   Abs Bib

   Despite recent advances in learning-based behavioral planning for autonomous systems, decision-making in multi-task missions remains a challenging problem. For instance, a mission might require a robot to explore an unknown environment, locate the goals, and navigate to them, even if there are obstacles along the way. Such problems are difficult to solve due to: a) sparse rewards, meaning a reward signal is available only once all the tasks in a mission have been satisfied, and b) the agent having to perform tasks at run-time that are not covered in the training data, e.g., demonstrations only from an environment where all doors were unlocked. Consequently, state-of-the-art decision-making methods in such settings are limited to missions where the required tasks are well-represented in the training demonstrations and can be solved within a short planning horizon. To overcome these limitations, we propose Adaptformer, a stochastic and adaptive planner that utilizes sequence models for sample-efficient exploration and exploitation. This framework relies on learning an energy-based heuristic, which needs to be minimized over a sequence of high-level decisions. To generate successful action sequences for long-horizon missions, Adaptformer aims to achieve shorter sub-goals, which are proposed through an intrinsic sub-goal curriculum. Through these two key components, Adaptformer allows for generalization to out-of-distribution tasks and environments, i.e., missions that were not a part of the training data. Empirical results in multiple simulation environments demonstrate the effectiveness of our method. Notably, Adaptformer not only outperforms the state-of-the-art method by up to 25% in multi-goal maze reachability tasks but also successfully adapts to multi-task missions that the state-of-the-art method could not complete, leveraging demonstrations from single-goal-reaching tasks.

   @inproceedings{karthikeyan2024adaptformer, title = {{AdaptFormer: Sequence Models as Adaptive Iterative Planners}}, author = {Karthikeyan, Akash and Pant, Yash Vardhan}, booktitle = {{AAAI 2024 Spring Symposium on User-Aligned Assessment of Adaptive AI Systems}}, year = {2024}, img = {placeholder_black.png} }

2020

1. 

   Poster: Formal Scenario-Based Testing of Autonomous Vehicles

   Fremont, Daniel J, Kim, Edward, Pant, Yash Vardhan, Seshia, Sanjit A, Acharya, Atul, Bruso, Xantha, Wells, Paul, Lemke, Steve, Yu, Qiang, and Mehta, Shalin

   In Automated Vehicles Symposium (AVS) 2020

   Bib

   @inproceedings{fremont2020poster, title = {{Poster: Formal Scenario-Based Testing of Autonomous Vehicles}}, author = {Fremont, Daniel J and Kim, Edward and Pant, Yash Vardhan and Seshia, Sanjit A and Acharya, Atul and Bruso, Xantha and Wells, Paul and Lemke, Steve and Yu, Qiang and Mehta, Shalin}, booktitle = {{Automated Vehicles Symposium (AVS)}}, year = {2020}, img = {placeholder_black.png} }

2019

1. 

   Distributed planning of Multi-rotor drone fleets using the Smooth Robustness of Signal Temporal Logic

   Pant, Yash Vardhan, Abbas, Houssam, and Mangharam, Rahul

   In Monitoring and Testing of CPS Workshop (MTCPS) at CPS Week 2019

   Bib

   @inproceedings{pantMTCPS19, title = {{Distributed planning of Multi-rotor drone fleets using the Smooth Robustness of Signal Temporal Logic}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mangharam, Rahul}, year = {2019}, booktitle = {{Monitoring and Testing of CPS Workshop (MTCPS) at CPS Week}}, img = {placeholder_black.png} }

2017

1. 

   Control with Temporal Logic Requirements (poster)

   Pant, Yash Vardhan, Abbas, Houssam, and Mangharam, Rahul

   In SRC TECHCON 2017

   Bib

   @inproceedings{pantSRC17, title = {{Control with Temporal Logic Requirements (poster)}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mangharam, Rahul}, year = {2017}, booktitle = {{SRC TECHCON}}, note = {tech}, img = {placeholder_black.png} }

2. 

   Control using the Smooth Robustness of Temporal Logic

   Pant, Yash Vardhan, Abbas, Houssam, and Mangharam, Rahul

   In Monitoring and Testing of CPS Workshop (MTCPS) at CPS Week 2017

   Bib

   @inproceedings{pantMTCPS17, title = {{Control using the Smooth Robustness of Temporal Logic}}, author = {Pant, Yash Vardhan and Abbas, Houssam and Mangharam, Rahul}, year = {2017}, booktitle = {{Monitoring and Testing of CPS Workshop (MTCPS) at CPS Week}}, img = {placeholder_black.png} }

2015

1. 

   Hardware Optimizations for Anytime Perception and Control

   Nischal, K. N., Kelkar, Paritosh, Kumar, Dhruva, Pant, Yash Vardhan, Abbas, Houssam, Devietti, Joseph, and Mangharam, Rahul

   In Work-in-progress, Real-Time Systems Symposium (RTSS) 2015

   Bib

   @inproceedings{KNRTSS15, title = {{Hardware Optimizations for Anytime Perception and Control}}, author = {Nischal, K. N. and Kelkar, Paritosh and Kumar, Dhruva and Pant, Yash Vardhan and Abbas, Houssam and Devietti, Joseph and Mangharam, Rahul}, year = {2015}, booktitle = {{Work-in-progress, Real-Time Systems Symposium (RTSS)}}, img = {placeholder_black.png} }

2014

1. 

   Demo: Low-cost Autonomous Navigation with Anytime Control and Computation

   Gurniak, Paul, and Pant, Yash Vardhan

   In University Transportation Center (UPenn-CMU) Annual Meeting 2014

   Bib

   @inproceedings{Gurniak14, title = {{Demo: Low-cost Autonomous Navigation with Anytime Control and Computation}}, author = {Gurniak, Paul and Pant, Yash Vardhan}, year = {2014}, booktitle = {{University Transportation Center (UPenn-CMU) Annual Meeting}}, img = {placeholder_black.png} }

2013

1. 

   Knock NOx: Model-based Remote Diagnostics of a Diesel Exhaust Control System

   Pant, Yash Vardhan, Nghiem, Truong X, and Mangharam, Rahul

   In Work-in-Progress IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS) 2013

   Bib

   @inproceedings{pant2013knock, title = {Knock NOx: Model-based Remote Diagnostics of a Diesel Exhaust Control System}, author = {Pant, Yash Vardhan and Nghiem, Truong X and Mangharam, Rahul}, booktitle = {Work-in-Progress IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS)}, pages = {17--20}, year = {2013}, img = {placeholder_black.png} }

2. 

   Final Report: Protodrive: Rapid Prototyping and Simulation of Electric Vehicle Powertrains

   Pant, Yash Vardhan, Jain, Harsh, Mulay, Abhijeet, and Dutta, Rajib

   In Intel Cornell Cup 2013

   Bib

   @inproceedings{CornellCup13, title = {{Final Report: Protodrive: Rapid Prototyping and Simulation of Electric Vehicle Powertrains}}, author = {Pant, Yash Vardhan and Jain, Harsh and Mulay, Abhijeet and Dutta, Rajib}, year = {2013}, booktitle = {{Intel Cornell Cup}}, img = {placeholder_black.png} }

3. 

   Final Report: Protodrive: Simulation of Electric Vehicle Powertrains

   Price, William, Jain, Harsh, Pant, Yash Vardhan, and Mangharam, Rahul

   In World Embedded Software Competition 2013

   Bib

   @inproceedings{WESC13, title = {{Final Report: Protodrive: Simulation of Electric Vehicle Powertrains}}, author = {Price, William and Jain, Harsh and Pant, Yash Vardhan and Mangharam, Rahul}, year = {2013}, booktitle = {{World Embedded Software Competition}}, img = {placeholder_black.png} }

2012

1. 

   Protodrive: An experimental platform for electric vehicle energy scheduling and control

   Diaz, Stephanie, Jain, Harsh, Pant, Yash Vardhan, and Mangharam, Rahul

   In Demo Abstract Real-Time Systems Symposium (RTSS) 2012

   Bib

   @inproceedings{Diaz12, title = {{Protodrive: An experimental platform for electric vehicle energy scheduling and control}}, author = {Diaz, Stephanie and Jain, Harsh and Pant, Yash Vardhan and Mangharam, Rahul}, year = {2012}, booktitle = {{Demo Abstract Real-Time Systems Symposium (RTSS)}}, img = {placeholder_black.png} }

2. 

   AutoPlug 2.0

   Pant, Yash Vardhan

   In Real-Time and Embedded Technology and Applications Symposium (RTAS): Demo session at CPS Week 2012

   Bib

   @inproceedings{plug12, title = {{AutoPlug 2.0}}, author = {Pant, Yash Vardhan}, year = {2012}, booktitle = {{Real-Time and Embedded Technology and Applications Symposium (RTAS): Demo session at CPS Week}}, img = {placeholder_black.png} }

3. 

   Observer-based Sensor Fault Detection and Isolation

   Pant, Yash Vardhan, and Mangharam, Rahul

   In Work-in-progress, ACM International Conference on High Confidence Networked Systems (HiCoNS) at CPS Week 2012

   Bib

   @inproceedings{hicons12, title = {{Observer-based Sensor Fault Detection and Isolation}}, author = {Pant, Yash Vardhan and Mangharam, Rahul}, year = {2012}, booktitle = {{Work-in-progress, ACM International Conference on High Confidence Networked Systems (HiCoNS) at CPS Week}}, img = {placeholder_black.png} }