PUBLICATIONS

Books  |  Patents  | 2026 | 2025  |  2024  |  2023   |   2022   |   2021   |  2020   |   2019   |   2018   |   2017   |   2016   |   2015   |   2014   |   2013   |   2012   |   2011   |   2010   |   2009   |   2008   |   2007   |   before 2007

2025

Shuo Liu, Zhe Huang, Jun Zeng, Koushil Sreenath, and Calin Belta, Learning-Enabled Iterative Convex Optimization for Safety-Critical Model Predictive Control, IEEE Open Journal of Control Systems, vol. 4, {4}, pp. 482-500, doi=10.1109/OJCSYS.2025.3612245, 2025 (pdf)

@ARTICLE{11174009,
author={Liu, Shuo and Huang, Zhe and Zeng, Jun and Sreenath, Koushil and Belta, Calin A.},
journal={IEEE Open Journal of Control Systems},
title={Learning-Enabled Iterative Convex Optimization for Safety-Critical Model Predictive Control},
year={2025},
volume={4},
number={},
pages={482-500},
keywords={Safety;Predictive control;Optimization;Iterative methods;Robots;Convex functions;Real-time systems;Optimal control;Artificial neural networks;Numerical models;Constrained control;optimal control;control barrier function;nonlinear predictive control;machine learning},
doi={10.1109/OJCSYS.2025.3612245}}

Cristian-Ioan  Vasile, Jana Tumova, Sertac Karaman, Calin Belta and Daniela Rus, Optimal On-the-fly Route Planning with Rich Transportation Requests, IEEE Transactions on Robotics, pp. 1-16, doi=10.1109/TRO.2025.3577010, 2025 (pdf)

@ARTICLE{11024558,
author={Vasile, Cristian-Ioan and Tumova, Jana and Karaman, Sertac and Belta, Calin and Rus, Daniela},
journal={IEEE Transactions on Robotics}, 
title={Optimal On-the-fly Route Planning with Rich Transportation Requests}, 
year={2025},
volume={},
number={},
pages={1-16},
keywords={Planning;Logic;Transportation;Roads;Robots;Delays;Cost function;Vehicle routing;Vehicle dynamics;Stochastic processes;Route Planning;Mobility on Demand;Autonomous Agents;Temporal Logic;MILP},
doi={10.1109/TRO.2025.3577010}}

Shuo Liu, Calin Belta, Risk-Aware Adaptive Control Barrier Functions for Safe Control of Nonlinear Systems under Stochastic Uncertainty, 64th IEEE Conference on Decision and Control (CDC), Rio de Janeiro, Brazil, 2025 (pdf)

@INPROCEEDINGS{11312202,
author={Liu, Shuo and Belta, Calin A.},
booktitle={2025 IEEE 64th Conference on Decision and Control (CDC)}, 
title={Risk-Aware Adaptive Control Barrier Functions for Safe Control of Nonlinear Systems under Stochastic Uncertainty}, 
year={2025},
volume={},
number={},
pages={4875-4881},
keywords={Uncertainty;System dynamics;Optimal control;Safety;Nonlinear systems;Adaptive control;Standards},
doi={10.1109/CDC57313.2025.11312202}}

Clinton Enwerem, Aniruddh Puranic, John S. Baras, Calin Belta, Safety-Aware Reinforcement Learning for Control via Risk-Sensitive Action-Value Iteration and Quantile Regression, 64th IEEE Conference on Decision and Control (CDC), Rio de Janeiro, Brazil, 2025 (pdf)

@INPROCEEDINGS{11312575,
author={Enwerem, Clinton and Puranic, Aniruddh G. and Baras, John S. and Belta, Calin},
booktitle={2025 IEEE 64th Conference on Decision and Control (CDC)}, 
title={Safety-Aware Reinforcement Learning for Control via Risk-Sensitive Action-Value Iteration and Quantile Regression}, 
year={2025},
volume={},
number={},
pages={4890-4895},
keywords={Training;Costs;Sensitivity;Navigation;Heuristic algorithms;Stochastic processes;Reinforcement learning;Safety;Mobile robots;Convergence},
doi={10.1109/CDC57313.2025.11312575}}

Yi-Hsuan Chen, Shuo Liu, Wei Xiao, Calin Belta, Michael Otte, Control Barrier Functions via Minkowski Operations for Safe Navigation among Polytopic Sets, 64th IEEE Conference on Decision and Control (CDC), Rio de Janeiro, Brazil, 2025 (pdf)

@INPROCEEDINGS{11312188,
author={Chen, Yi-Hsuan and Liu, Shuo and Xiao, Wei and Belta, Calin and Otte, Michael},
booktitle={2025 IEEE 64th Conference on Decision and Control (CDC)}, 
title={Control Barrier Functions via Minkowski Operations for Safe Navigation among Polytopic Sets}, 
year={2025},
volume={},
number={},
pages={4481-4488},
keywords={Geometry;Translation;Navigation;System dynamics;Shape;Convex functions;Robots;Optimization;Ellipsoids;Videos},
doi={10.1109/CDC57313.2025.11312188}}

Eric Palanques Tost, Hanna Krasowski, Murat Arcak, Ron Weiss, Calin Belta, STL-based Optimization of Biomolecular Neural Networks for Regression and Control, 64th IEEE Conference on Decision and Control (CDC), Rio de Janeiro, Brazil, 2025 (pdf)

@INPROCEEDINGS{11312596,
author={Palanques-Tost, Eric and Krasowski, Hanna and Arcak, Murat and Weiss, Ron and Belta, Calin},
booktitle={2025 IEEE 64th Conference on Decision and Control (CDC)}, 
title={STL-based Optimization of Biomolecular Neural Networks for Regression and Control}, 
year={2025},
volume={},
number={},
pages={3276-3281},
keywords={Training;Semantics;Circuits;Artificial neural networks;Numerical models;Feedback control;Logic;Function approximation;Optimization;Diseases},
doi={10.1109/CDC57313.2025.11312596}}

Peter Crowley, Brendan Long, Andrew Schoer, Zachary Serlin, Makai Mann, Tyler Gonsalves, John Kliem, Calin Belta, Pyquaticus: A Sim-to-Real Pipeline for Learning in Multi-Agent Maritime Strategy Games, 19th International Symposium on Experimental Robotics (ISER), Santa Fe, NM, 2025 (pdf)

@INPROCEEDINGS{pyquaticus2005,
author={Peter Crowley, Brendan Long, Andrew Schoer, Zachary Serlin, Makai Mann, Tyler Gonsalves, John Kliem, Calin Belta},
booktitle={19th International Symposium on Experimental Robotics (ISER), Santa Fe, NM}, 
title={Pyquaticus: A Sim-to-Real Pipeline for Learning in Multi-Agent Maritime Strategy Games}, 
year={2025},
volume={},
number={},
pages={},
keywords={}}

Wenliang Liu, Danyang Li, Erfan Aasi, Daniela Rus, Roberto Tron and Calin Belta, Interpretable Imitation Learning via Generative Adversarial STL Inference and Control, 2nd International Conference on Neuro-symbolic Systems (NeuS), Philadelphia, PA, 2025 (pdf)

@InProceedings{pmlr-v288-liu25a,
title = {Interpretable Imitation Learning via Generative Adversarial STL Inference and Control},
author = {Liu, Wenliang and Li, Danyang and Aasi, Erfan and Rus, Daniela and Tron, Roberto and Belta, Calin},
booktitle = {Proceedings of the International Conference on Neuro-symbolic Systems},
pages = {472--489},
year = {2025},
editor = {Pappas, George and Ravikumar, Pradeep and Seshia, Sanjit A.},
volume = {288},
series = {Proceedings of Machine Learning Research},
month = {28--30 May},
publisher = {PMLR},
pdf = {https://raw.githubusercontent.com/mlresearch/v288/main/assets/liu25a/liu25a.pdf},
url = {https://proceedings.mlr.press/v288/liu25a.html},
abstract = {Imitation learning methods have demonstrated considerable success in teaching autonomous systems complex tasks through expert demonstrations. However, a limitation of these methods is their lack of interpretability, particularly in understanding the specific task the learning agent aims to accomplish. In this paper, we propose a novel imitation learning method that combines Signal Temporal Logic (STL) inference and control synthesis, enabling the explicit representation of the task as an STL formula. This approach not only provides a clear understanding of the task but also supports the integration of human knowledge and allows for adaptation to out-of-distribution scenarios by manually adjusting the STL formulas and fine-tuning the policy. We employ a Generative Adversarial Network (GAN)-inspired approach to train both the inference and policy networks, effectively narrowing the gap between expert and learned policies. The efficiency of our algorithm is demonstrated through simulations, showcasing its practical applicability and adaptability.}
}

Ryan Matheu, Aniruddh Gopinath Puranic, John S. Baras, and Calin Belta, OMTBT: Online Monitoring of Temporal Behavior Trees with Applications to Closed-Loop Learning, 23rd European Control Conference (ECC), 2025, Thessaloniki, Greece, 2025 (pdf)

@INPROCEEDINGS{11187275,
author={Matheu, Ryan and Puranic, Aniruddh G. and Baras, John S. and Belta, Calin},
booktitle={2025 European Control Conference (ECC)}, 
title={OMTBT: Online Monitoring of Temporal Behavior Trees with Applications to Closed-Loop Learning}, 
year={2025},
volume={},
number={},
pages={2129-2135},
keywords={Simulation;Semantics;Multivalued logic;Control systems;Real-time systems;Robustness;Safety;Trajectory;Robots;Monitoring},
doi={10.23919/ECC65951.2025.11187275}}

Hanna Krasowski, Eric Palanques-Tost, Calin Belta and Murat Arcak, Learning Biomolecular Models using Signal Temporal Logic, Learning for Dynamics and Control Conference (L4DC), University of Michigan, 2025 (pdf)

@InProceedings{pmlr-v283-krasowski25a,
title = {Learning Biomolecular Models using Signal Temporal Logic},
author = {Krasowski, Hanna and Palanques-Tost, Eric and Belta, Calin and Arcak, Murat},
booktitle = {Proceedings of the 7th Annual Learning for Dynamics \& Control Conference},
pages = {1365--1377},
year = {2025},
editor = {Ozay, Necmiye and Balzano, Laura and Panagou, Dimitra and Abate, Alessandro},
volume = {283},
series = {Proceedings of Machine Learning Research},
month = {04--06 Jun},
publisher = {PMLR},
pdf = {https://raw.githubusercontent.com/mlresearch/v283/main/assets/krasowski25a/krasowski25a.pdf},
url = {https://proceedings.mlr.press/v283/krasowski25a.html},
abstract = {Modeling dynamical biological systems is key for understanding, predicting, and controlling complex biological behaviors. Traditional methods for identifying governing equations, such as ordinary differential equations (ODEs), typically require extensive quantitative data, which is often scarce in biological systems due to experimental limitations. To address this challenge, we introduce an approach that determines biomolecular models from qualitative system behaviors expressed as Signal Temporal Logic (STL) statements, which are naturally suited to translate expert knowledge into computationally tractable specifications. Our method represents the biological network as a graph, where edges represent interactions between species, and uses a genetic algorithm to identify the graph. To infer the parameters of the ODEs modeling the interactions, we propose a gradient-based algorithm. On a numerical example, we evaluate two loss functions using STL robustness and analyze different initialization techniques to improve the convergence of the approach.}
}

Ahmad Ahmad, Mehdi Kermanshah, Kevin Leahy, Zachary Serlin, Ho Chit Siu, Makai Mann, Cristian-Ioan Vasile, Roberto Tron and Calin Belta , Accelerating Proximal Policy Optimization Learning Using Task Prediction for Solving Environments with Delayed Rewards,  Learning for Dynamics and Control Conference (L4DC), University of Michigan, 2025 (pdf)

@InProceedings{pmlr-v283-ahmad25a,
title = {Accelerating Proximal Policy Optimization Learning Using Task Prediction for Solving Environments with Delayed Rewards},
author = {Ahmad, Ahmad and Kermanshah, Mehdi and Leahy, Kevin and Serlin, Zachary and Siu, Ho Chit and Mann, Makai and Vasile, Cristian-Ioan and Tron, Roberto and Belta, Calin},
booktitle = {Proceedings of the 7th Annual Learning for Dynamics \& Control Conference},
pages = {566--578},
year = {2025},
editor = {Ozay, Necmiye and Balzano, Laura and Panagou, Dimitra and Abate, Alessandro},
volume = {283},
series = {Proceedings of Machine Learning Research},
month = {04--06 Jun},
publisher = {PMLR},
pdf = {https://raw.githubusercontent.com/mlresearch/v283/main/assets/ahmad25a/ahmad25a.pdf},
url = {https://proceedings.mlr.press/v283/ahmad25a.html},
abstract = {In this paper, we tackle the challenging problem of delayed rewards in reinforcement learning (RL). While Proximal Policy Optimization (PPO) has emerged as a leading Policy Gradient method, its performance can degrade under delayed rewards. We introduce two key enhancements to PPO: a hybrid policy architecture that combines an offline policy (trained on expert demonstrations) with an online PPO policy, and a reward shaping mechanism using Time Window Temporal Logic (TWTL). The hybrid architecture leverages offline data throughout training while maintaining PPO’s theoretical guarantees. Building on the monotonic improvement framework of Trust Region Policy Optimization (TRPO), we prove that our approach ensures improvement over both the offline policy and previous iterations, with a bounded performance gap of $(2\varsigma\gamma\alpha^2)/(1-\gamma)^2$, where $\alpha$ is the mixing parameter, $\gamma$ is the discount factor, and $\varsigma$ bounds the expected advantage. Additionally, we prove that our TWTL-based reward shaping preserves the optimal policy of the original problem. TWTL enables formal translation of temporal objectives into immediate feedback signals that guide learning. We demonstrate the effectiveness of our approach through extensive experiments on an inverted pendulum and a lunar lander environments, showing improvements in both learning speed and final performance compared to standard PPO and offline-only approaches.}
}

Ryan Matheu, Aniruddh G. Puranic, John S. Baras and Calin Belta, BT2Automata: Expressing Behavior Trees as Automata for Formal Control Synthesis, Hybrid Systems: Computation and Control Irvine, Ca, 2025 (pdf)

@inproceedings{10.1145/3716863.3718042,
author = {Matheu, Ryan and Puranic, Aniruddh G. and Baras, John S. and Belta, Calin},
title = {BT2Automata: Expressing Behavior Trees as Automata for Formal Control Synthesis},
year = {2025},
isbn = {9798400715044},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3716863.3718042},
doi = {10.1145/3716863.3718042},
abstract = {This research presents a novel approach to bridging the gap between the interpretable and flexible nature of Behavior Trees (BTs) and the rigorous formal verification and synthesis capabilities of temporal logics. Temporal logics, such as Linear Temporal Logic (LTL) and Metric Interval Temporal Logic (MITL), are widely used for task specification due to their intuitive syntax for expressing temporally evolving behaviors. However, encoding complex task dependencies and recovery actions in temporal logic can lead to intractability. BTs, known for their modular structure and dynamic adaptability, have gained popularity in robotics for task specification. Despite the advantages of BTs, their flexible structure complicates formal analysis for safety and performance guarantees, limiting their use in control synthesis. This work presents a novel approach by translating BTs into Timed Automata (TA), thus enabling falsification (counterexample generation) with Uppaal to identify inconsistencies and ensure language completeness, especially when defined with timing constraints. This integration allows for the detection of potential inconsistencies, the monitoring of temporal properties, and the synthesis of automaton and sampling based control strategies that guarantee satisfaction of task objectives.},
booktitle = {Proceedings of the 28th ACM International Conference on Hybrid Systems: Computation and Control},
articleno = {13},
numpages = {11},
keywords = {Behavior Trees, Control Synthesis, Falsification, Temporal Logic, Timed Automata},
location = {Irvine, CA, USA},
series = {HSCC '25}
}

Erfan Aasi, Mingyu  Cai, Cristian-Ioan Vasile, Calin Belta, A Two-Level Control Algorithm for Autonomous Driving in Urban Environments, Transactions on Intelligent Transportation Systems, vol. 26, issue 1, pp. 410-424, DOI: 10.1109/TITS.2024.3486557, 2025 (pdf)

@ARTICLE{10753297,
  author={Aasi, Erfan and Cai, Mingyu and Vasile, Cristian-Ioan and Belta, Calin},
  journal={IEEE Transactions on Intelligent Transportation Systems}, 
  title={A Two-Level Control Algorithm for Autonomous Driving in Urban Environments}, 
  year={2025},
  volume={26},
  number={1},
  pages={410-424},
  keywords={Autonomous vehicles;Vehicle dynamics;Uncertainty;Vectors;Runtime;Logic;Urban areas;Stochastic processes;Real-time systems;Predictive models;Autonomous vehicles;formal methods;predictive control},
  doi={10.1109/TITS.2024.3486557}}

Wenliang Liu, Suhail Alsalehi, Noushin Mehdipour, Ezio Bartocci, Calin Belta, Quantifying the Satisfaction of Spatio-Temporal Logic Specifications for Multi-Agent Control, IEEE Transactions on Automatic Control, pp. 1-16, doi: 10.1109/TAC.2025.3538747, 2025 (pdf)

@ARTICLE{10872804,
author={Liu, Wenliang and Alsalehi, Suhail and Mehdipour, Noushin and Bartocci, Ezio and Belta, Calin},
journal={IEEE Transactions on Automatic Control}, 
title={Quantifying the Satisfaction of Spatio-Temporal Logic Specifications for Multi-Agent Control}, 
year={2025},
volume={},
number={},
pages={1-16},
keywords={Robustness;Logic;Optimization;Measurement;Training;Imitation learning;Trajectory;Semantics;Robot kinematics;Recurrent neural networks},
doi={10.1109/TAC.2025.3538747}}

Shuo Liu, Yihui Mao, Calin Belta, Safety-Critical Planning and Control for Dynamic Obstacle Avoidance Using Control Barrier Functions, American Control Conference (ACC), 2025 (pdf)

@INPROCEEDINGS{11107805,
author={Liu, Shuo and Mao, Yihui and Belta, Calin A.},
booktitle={2025 American Control Conference (ACC)}, 
title={Safety-Critical Planning and Control for Dynamic Obstacle Avoidance Using Control Barrier Functions}, 
year={2025},
volume={},
number={},
pages={348-354},
keywords={Trajectory planning;Heuristic algorithms;Prediction algorithms;Mathematical models;Trajectory;Safety;Planning;Collision avoidance;Robots;Predictive control},
doi={10.23919/ACC63710.2025.11107805}}

Guang Yang, Mingyu Cai, Ahmad Ahmad, Amanda Prorok, Roberto Tron, Calin Belta, LQR-CBF-RRT*: Safe and Optimal Motion Planning, American Control Conference (ACC), 2025 (pdf)

@INPROCEEDINGS{11108073,
author={Yang, Guang and Cai, Mingyu and Ahmad, Ahmad and Prorok, Amanda and Tron, Roberto and Belta, Calin},
booktitle={2025 American Control Conference (ACC)}, 
title={LQR-CBF-RRT*: Safe and Optimal Motion Planning}, 
year={2025},
volume={},
number={},
pages={3700-3705},
keywords={Regulators;Optimal control;Benchmark testing;Planning;Safety;Trajectory;Computational efficiency;Nonlinear systems},
doi={10.23919/ACC63710.2025.11108073}}
Books  |  Patents  | 2026 | 2025  |  2024  |  2023   |   2022   |   2021   |  2020   |   2019   |   2018   |   2017   |   2016   |   2015   |   2014   |   2013   |   2012   |   2011   |   2010   |   2009   |   2008   |   2007   |   before 2007