The field of autonomous systems is rapidly advancing, with a focus on improving safety and efficiency. Recent developments have highlighted the importance of communication and cooperation in multi-agent systems, as well as the need for robust and reliable control methods. Advances in reinforcement learning and control barrier functions are enabling the development of safe and efficient autonomous systems, with applications in areas such as urban air mobility, cyber defense, and power systems. Notably, the use of ensemble defense approaches and hierarchical multi-agent reinforcement learning are showing promising results in enhancing the robustness and safety of autonomous systems. Noteworthy papers include: Learning to Communicate in Multi-Agent Reinforcement Learning for Autonomous Cyber Defence, which proposes a game design where defender agents learn to communicate and defend against imminent cyber threats. Hierarchical Multi-Agent Reinforcement Learning with Control Barrier Functions for Safety-Critical Autonomous Systems, which proposes a safe hierarchical approach based on control barrier functions to ensure safety in multi-agent systems.
Safety and Efficiency in Autonomous Systems
Sources
Real-Time Communication-Aware Ride-Sharing Route Planning for Urban Air Mobility: A Multi-Source Hybrid Attention Reinforcement Learning Approach
Learning to Communicate in Multi-Agent Reinforcement Learning for Autonomous Cyber Defence
Hierarchical Multi-Agent Reinforcement Learning with Control Barrier Functions for Safety-Critical Autonomous Systems
Red-Team Multi-Agent Reinforcement Learning for Emergency Braking Scenario
Advancing Robustness in Deep Reinforcement Learning with an Ensemble Defense Approach
Safe Reinforcement Learning-based Automatic Generation Control
Stability Constrained Voltage Control in Distribution Grids with Arbitrary Communication Infrastructure
Auto-SGCR: Automated Generation of Smart Grid Cyber Range Using IEC 61850 Standard Models