Robust and Safe Multi-Agent Reinforcement Learning with Communication for Autonomous Vehicles: From Simulation to Hardware
On the map: End-to-end autonomous driving → Cooperative and V2X driving
Abstract (arXiv)
Deep multi-agent reinforcement learning (MARL) has been demonstrated effectively in simulations for multi-robot problems. For autonomous vehicles, the development of vehicle-to-vehicle (V2V) communication technologies provide opportunities to further enhance system safety. However, zero-shot transfer of simulator-trained MARL policies to dynamic hardware systems remains challenging, and how to leverage communication and shared information for MARL has limited demonstrations on hardware. This problem is challenged by discrepancies between simulated and physical states, system state and model uncertainties, practical shared information design, and the need for safety guarantees in both simulation and hardware. This paper designs RSR-RSMARL, a novel Robust and Safe MARL framework that supports Real-Sim-Real (RSR) policy adaptation for multi-agent systems with communication among agents, with both simulation and hardware demonstrations. RSR-RSMARL leverages state (includes shared state information among agents) and action representations considering real system complexities for MARL formulation. The MARL policy is trained with robust MARL algorithm to enable zero-shot transfer to hardware considering the sim-to-real gap. A safety shield module using Control Barrier Functions (CBFs) provides safety guarantee for each individual agent. Experimental results on 1/10th-scale autonomous vehicles with V2V communication demonstrate the ability of RSR-RSMARL framework to enhance driving safety and coordination across multiple configurations. These findings emphasize the importance of jointly designing robust policy representations and modular safety architectures to enable scalable, generalizable RSR transfer in multi-agent autonomy.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Cooperative and V2X driving, one of 9 directions in End-to-end autonomous driving. Builds on 25 of its 34 references that are in the map. First posted on arXiv 1 Jun 2025.
Builds on
25 of 34 references.
Closest ancestors · 1
Same direction, newest first
Other references · 24
- 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,401
- 2020Sim-to-Real Transfer in Deep Reinforcement Learning for Robotics: a Surveycited by 388
- 2021Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learningcited by 337
- 2019Active Domain Randomizationcited by 161
- 2022V2X-Sim: Multi-Agent Collaborative Perception Dataset and Benchmark for Autonomous Drivingcited by 157
- 2024Deep Reinforcement Learning for Robotics: A Survey of Real-World Successescited by 153
- 2024Reconciling Reality through Simulation: A Real-to-Sim-to-Real Approach for Robust Manipulationcited by 135
- 2020Ultra Fast Structure-aware Deep Lane Detectioncited by 121
- 2021Safe Multi-Agent Reinforcement Learning via Shieldingcited by 60
- 2024TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correctioncited by 50
- 2022Efficient Adversarial Training without Attacking: Worst-Case-Aware Robust Reinforcement Learningcited by 42
- 2023Robust Multi-Agent Reinforcement Learning with State Uncertaintycited by 35
12 more
- 2022What is the Solution for State-Adversarial Multi-Agent Reinforcement Learning?cited by 21
- 2020A Multi-Agent Reinforcement Learning Approach For Safe and Efficient Behavior Planning Of Connected Autonomous Vehiclescited by 12
- 2022Stable and Efficient Shapley Value-Based Reward Reallocation for Multi-Agent Reinforcement Learning of Autonomous Vehiclescited by 11
- 2019A Fleet of Miniature Cars for Experiments in Cooperative Drivingcited by 9
- 2021Safe Multi-Agent Reinforcement Learning through Decentralized Multiple Control Barrier Functionscited by 9
- 2024Learning Multi-Agent Loco-Manipulation for Long-Horizon Quadrupedal Pushingcited by 9
- 2024The Cambridge RoboMaster: An Agile Multi-Robot Research Platformcited by 8
- 2023Multi-Agent Reinforcement Learning Guided by Signal Temporal Logic Specificationscited by 6
- 2023Safety Guaranteed Robust Multi-Agent Reinforcement Learning with Hierarchical Control for Connected and Automated Vehiclescited by 4
- 2024A Survey on Small-Scale Testbeds for Connected and Automated Vehicles and Robot Swarmscited by 4
- 2023Learning Adaptive Safety for Multi-Agent Systemscited by 3
- 2024Reinforcement Learning-based Receding Horizon Control using Adaptive Control Barrier Functions for Safety-Critical Systemscited by 3
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2024Multi-Agent Reinforcement Learning for Autonomous Driving: A Survey
- 2025MARL-OT: Multi-Agent Reinforcement Learning Guided Online Fuzzing to Detect Safety Violation in Autonomous Driving Systems
- 2025Multi-agent Uncertainty-Aware Pessimistic Model-Based Reinforcement Learning for Connected Autonomous Vehicles
- 2023Shared Information-Based Safe And Efficient Behavior Planning For Connected Autonomous Vehicles
- 2026Meta-Multi-Agent Reinforcement Learning for Fast Adaptation of Interactive Policies with Applications to Autonomous Driving
- 2024Safe Multi-agent Reinforcement Learning with Natural Language Constraints
- 2025Evo-MARL: Co-Evolutionary Multi-Agent Reinforcement Learning for Internalized Safety
- 2021Minimizing Safety Interference for Safe and Comfortable Automated Driving with Distributional Reinforcement Learning