mlsanityData to 14 Oct 2026

MORTAR: A Model-based Runtime Action Repair Framework for AI-enabled Cyber-Physical Systems

arXiv 2408.03892 ↗First posted 7 Aug 2024cs.SE cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Cyber-Physical Systems (CPSs) are increasingly prevalent across various industrial and daily-life domains, with applications ranging from robotic operations to autonomous driving. With recent advancements in artificial intelligence (AI), learning-based components, especially AI controllers, have become essential in enhancing the functionality and efficiency of CPSs. However, the lack of interpretability in these AI controllers presents challenges to the safety and quality assurance of AI-enabled CPSs (AI-CPSs). Existing methods for improving the safety of AI controllers often involve neural network repair, which requires retraining with additional adversarial examples or access to detailed internal information of the neural network. Hence, these approaches have limited applicability for black-box policies, where only the inputs and outputs are accessible during operation. To overcome this, we propose MORTAR, a runtime action repair framework designed for AI-CPSs in this work. MORTAR begins by constructing a prediction model that forecasts the quality of actions proposed by the AI controller. If an unsafe action is detected, MORTAR then initiates a repair process to correct it. The generation of repaired actions is achieved through an optimization process guided by the safety estimates from the prediction model. We evaluate the effectiveness of MORTAR across various CPS tasks and AI controllers. The results demonstrate that MORTAR can efficiently improve task completion rates of AI controllers under specified safety specifications. Meanwhile, it also maintains minimal computational overhead, ensuring real-time operation of the AI-CPSs.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 36 of its 69 references that are in the map. First posted on arXiv 7 Aug 2024.

Builds on

36 of 69 references.

References · 36

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2015Continuous control with deep reinforcement learningcited by 3,407
  4. 2015Trust Region Policy Optimizationcited by 2,580
  5. 2016Adversarial examples in the physical worldcited by 1,769
  6. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 998
  7. 2017Constrained Policy Optimizationcited by 983
  8. 2019Adversarial Training for Free!cited by 643
  9. 2017Safe Reinforcement Learning via Shieldingcited by 449
  10. 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 351
  11. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 279
  12. 2018Generating Adversarial Examples with Adversarial Networkscited by 258
24 more
  1. 2017DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Carscited by 251
  2. 2018Safe Exploration in Continuous Action Spacescited by 247
  3. 2020Projection-Based Constrained Policy Optimizationcited by 203
  4. 2017A General Safety Framework for Learning-Based Control in Uncertain Robotic Systemscited by 163
  5. 2019End-to-End Robotic Reinforcement Learning without Reward Engineeringcited by 122
  6. 2020Conservative Safety Critics for Explorationcited by 113
  7. 2022LAS-AT: Adversarial Training with Learnable Attack Strategycited by 95
  8. 2018Universal Adversarial Trainingcited by 87
  9. 2019Safe Reinforcement Learning Using Robust MPCcited by 85
  10. 2018Safe Reinforcement Learning with Model Uncertainty Estimatescited by 72
  11. 2020Neural Dynamic Policies for End-to-End Sensorimotor Learningcited by 58
  12. 2021Semantics for Robotic Mapping, Perception and Interaction: A Surveycited by 56
  13. 2019Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributionscited by 48
  14. 2015Shield Synthesis: Runtime Enforcement for Reactive Systemscited by 42
  15. 2018Deep Reinforcement Learning for Wireless Sensor Scheduling in Cyber-Physical Systemscited by 23
  16. 2019Comparison of Deep Reinforcement Learning and Model Predictive Control for Adaptive Cruise Controlcited by 22
  17. 2021Neural Network Repair with Reachability Analysiscited by 19
  18. 2023Towards Building AI-CPS with NVIDIA Isaac Sim: An Industrial Benchmark and Case Study for Robotics Manipulationcited by 11
  19. 2020A Study on the Challenges of Using Robotics Simulators for Testingcited by 10
  20. 2020Runtime-Safety-Guided Policy Repaircited by 6
  21. 2020Learning a Low-dimensional Representation of a Safe Region for Safe Reinforcement Learning on Dynamical Systemscited by 6
  22. 2023AutoRepair: Automated Repair for AI-Enabled Cyber-Physical Systems under Safety-Critical Conditionscited by 4
  23. 2013A Notion of Robustness for Cyber-Physical Systemscited by 2
  24. 2020Early Validation of Cyber-Physical Space Systems via Multi-Concerns Integrationcited by 1

Built on by

Not cited yet.