Xiao Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xiao Wang has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Matthias Althoff
Technical University of Munich
2 shared publications - Hanna Krasowski
University of California, Berkeley
1 shared publication - Jian Li
School of Materials Science and Engineering, Jiangsu University, Zhenjiang, PR China
1 shared publication - Niklas Kochdumper
IRIF institute, Universit Paris Cit, Paris, France
1 shared publication - Stanley Bak
Stony Brook University, Stony Brook, NY, USA
1 shared publication
Latest Publications
- Open Access
- 19 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 9 (9) , 5490-5505
Safe Reinforcement Learning for Automated Vehicles via Online Reachability Analysis
Ensuring safe and capable motion planning is paramount for automated vehicles. Traditional methods are limited in their ability to handle complex and unpredictable traffic situations. Model-free reinforcement learning (RL) addresses this challenge by generalizing across different traffic situations without requiring explicit knowledge of all possible outcomes. However, it also poses challenges due to its inherent lack of safety guarantees. To bridge this gap, we integrate online reachability analysis into model-free RL to provide real-time safety guarantees. Reachability analysis helps to identify unsafe states and actions, enabling provably safe decision-making in automated vehicles. We evaluate the effectiveness of our approach through extensive numerical experiments. Our results demonstrate that we can efficiently provide safety guarantees without impairing the performance of the learned agent.
- 19 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 9 (9) , 5490-5505
Open AccessSafe Reinforcement Learning for Automated Vehicles via Online Reachability Analysis
Ensuring safe and capable motion planning is paramount for automated vehicles. Traditional methods are limited in their ability to handle complex and unpredictable traffic situations. Model-free reinforcement learning (RL) addresses this challenge by generalizing across different traffic situations without requiring explicit knowledge of all possible outcomes. However, it also poses challenges due to its inherent lack of safety guarantees. To bridge this gap, we integrate online reachability analysis into model-free RL to provide real-time safety guarantees. Reachability analysis helps to identify unsafe states and actions, enabling provably safe decision-making in automated vehicles. We evaluate the effectiveness of our approach through extensive numerical experiments. Our results demonstrate that we can efficiently provide safety guarantees without impairing the performance of the learned agent.
- Top Cited PapersOpen Access
- 13 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Control Systems
- Vol. 2, 79-92
Provably Safe Reinforcement Learning via Action Projection Using Reachability Analysis and Polynomial Zonotopes
While reinforcement learning produces very promising results for many applications, its main disadvantage is the lack of safety guarantees, which prevents its use in safety-critical systems. In this work, we address this issue by a safety shield for nonlinear continuous systems that solve reach-avoid tasks. Our safety shield prevents applying potentially unsafe actions from a reinforcement learning agent by projecting the proposed action to the closest safe action. This approach is called action projection and is implemented via mixed-integer optimization. The safety constraints for action projection are obtained by applying parameterized reachability analysis using polynomial zonotopes, which enables to accurately capture the nonlinear effects of the actions on the system. In contrast to other state-of-the-art approaches for action projection, our safety shield can efficiently handle input constraints and dynamic obstacles, eases incorporation of the spatial robot dimensions into the safety constraints, guarantees robust safety despite process noise and measurement errors, and is well suited for high-dimensional systems, as we demonstrate on several challenging benchmark systems.
- 13 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Control Systems
- Vol. 2, 79-92
Top Cited PapersOpen AccessProvably Safe Reinforcement Learning via Action Projection Using Reachability Analysis and Polynomial Zonotopes
While reinforcement learning produces very promising results for many applications, its main disadvantage is the lack of safety guarantees, which prevents its use in safety-critical systems. In this work, we address this issue by a safety shield for nonlinear continuous systems that solve reach-avoid tasks. Our safety shield prevents applying potentially unsafe actions from a reinforcement learning agent by projecting the proposed action to the closest safe action. This approach is called action projection and is implemented via mixed-integer optimization. The safety constraints for action projection are obtained by applying parameterized reachability analysis using polynomial zonotopes, which enables to accurately capture the nonlinear effects of the actions on the system. In contrast to other state-of-the-art approaches for action projection, our safety shield can efficiently handle input constraints and dynamic obstacles, eases incorporation of the spatial robot dimensions into the safety constraints, guarantees robust safety despite process noise and measurement errors, and is well suited for high-dimensional systems, as we demonstrate on several challenging benchmark systems.
- 22 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (2) , 4773-4780
Ensuring Safety of Learning-Based Motion Planners Using Control Barrier Functions
Reinforcement learning (RL) has been successfully applied to sequential decision-making problems, e.g., playing computer games or solving robotic tasks in simulations. However, RL methods are not yet ready to be applied to real robotic systems if safety is a major concern. To address this issue, we propose a safety layer based on control barrier functions to ensure the safety for an RL-based motion planner for highway scenarios with a continuous action space. Our method ensures legal safety by following traffic rules. Moreover, we propose a relaxation mechanism so that safety is restored as soon as possible when other vehicles violate traffic rules and render our optimization problem infeasible. We evaluate our approach using a real-world highway dataset and a traffic simulator. Numerical experiments confirm that an agent equipped with our proposed safety layer does not cause any accidents during learning and yet reaches the goal as often as an agent without a safety layer.
- 22 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (2) , 4773-4780
Ensuring Safety of Learning-Based Motion Planners Using Control Barrier Functions
Reinforcement learning (RL) has been successfully applied to sequential decision-making problems, e.g., playing computer games or solving robotic tasks in simulations. However, RL methods are not yet ready to be applied to real robotic systems if safety is a major concern. To address this issue, we propose a safety layer based on control barrier functions to ensure the safety for an RL-based motion planner for highway scenarios with a continuous action space. Our method ensures legal safety by following traffic rules. Moreover, we propose a relaxation mechanism so that safety is restored as soon as possible when other vehicles violate traffic rules and render our optimization problem infeasible. We evaluate our approach using a real-world highway dataset and a traffic simulator. Numerical experiments confirm that an agent equipped with our proposed safety layer does not cause any accidents during learning and yet reaches the goal as often as an agent without a safety layer.
- 8 July 2013
- Published by Taylor & Francis in Tribology Transactions
- Vol. 56 (6) , 1011-1018
Numerical Simulation of the Influence of the Bulges around Laser Surface Textures on the Tribological Performance
To improve the tribological performance of mating surfaces such as journal bearings, dimples with different shapes were introduced through laser surface texturing. In laser surface texturing processes, bulges are always induced around the dimples and may influence the tribological performance of the surface. In this article, the influence of the bulges on the tribological performance of a journal bearing is numerically investigated. Variations in both the friction coefficient and the eccentricity ratio of a journal bearing with the width and height of the bulges are studied. The influence of the radius of the dimples on the tribological performance of a journal bearing is also investigated. Results show that the friction coefficient increases with both the width and the height of bulges in the case of journal bearing under light and moderate loading conditions, which is in accord with the experimental results in previous works. The linear variation in the friction coefficient with the height of the bulges indicates that the wear rate of the bulges may be investigated by monitoring the friction coefficient. The abnormality of friction coefficient variation with the width of the bulges in the case of heavy loading condition indicates that introducing the bulges is an effective approach for reasonable tribological performance. From the results for the influence of the dimple radius, it can be concluded that although the introduction of dimples can improve the tribological performance, it also has some disadvantages such as decreasing the load-carrying capacity.
- 8 July 2013
- Published by Taylor & Francis in Tribology Transactions
- Vol. 56 (6) , 1011-1018
Numerical Simulation of the Influence of the Bulges around Laser Surface Textures on the Tribological Performance
To improve the tribological performance of mating surfaces such as journal bearings, dimples with different shapes were introduced through laser surface texturing. In laser surface texturing processes, bulges are always induced around the dimples and may influence the tribological performance of the surface. In this article, the influence of the bulges on the tribological performance of a journal bearing is numerically investigated. Variations in both the friction coefficient and the eccentricity ratio of a journal bearing with the width and height of the bulges are studied. The influence of the radius of the dimples on the tribological performance of a journal bearing is also investigated. Results show that the friction coefficient increases with both the width and the height of bulges in the case of journal bearing under light and moderate loading conditions, which is in accord with the experimental results in previous works. The linear variation in the friction coefficient with the height of the bulges indicates that the wear rate of the bulges may be investigated by monitoring the friction coefficient. The abnormality of friction coefficient variation with the width of the bulges in the case of heavy loading condition indicates that introducing the bulges is an effective approach for reasonable tribological performance. From the results for the influence of the dimple radius, it can be concluded that although the introduction of dimples can improve the tribological performance, it also has some disadvantages such as decreasing the load-carrying capacity.