Robot teams learn to navigate safely despite changing connections
Graph-Based Safe Reinforcement Learning for Multi-Agent Systems with Time-Varying Topology
Robotics
Summary
Keeping groups of robots safe while they work together is hard when their communication links keep changing and they only see part of their surroundings. The authors created a method that uses a special safety check to make sure robots do not break important safety rules, even as they learn from experience. They combined this safety check with advanced networks that help the robots understand each other’s positions and make good decisions, even when their communication changes over time. Their method was tested on real robots and showed it could keep them stable and safe while moving in tricky situations.
multi-agent reinforcement learningtime-varying topologyControl Barrier-Like FunctionLiDARGraph Attention Networkactor-critic architecturesafety constraintsdifferential-drive robotscollaborative navigationfield-of-view limitations
Authors
Xiao Sizhe, Dong Lijing, Bai Rui, Tan Xin
Abstract
This paper presents a graph-based safe multi-agent reinforcement learning (MARL) framework for cooperative navigation with time-varying topology. To address the critical challenge of ensuring safety in environments with sensing constraints, a safety-decoupled mechanism is introduced through a Control Barrier-Like Function (CBLF) action screening layer. This mechanism bridges the gap between discrete LiDAR perception and continuous safety constraints, ensuring that physical safety constraints are strictly satisfied regardless of the learning progress. Building upon this safety foundation, a unified structural architecture is proposed, integrating a attention-based actor and a Graph Attention Network (GAT) centralized critic. The actor utilizes a value vector reconstruction mechanism that explicitly encodes relative geometric relations through a collaborative tracking error matrix, enabling scale-insensitive policy learning under time-varying communication topologies. Meanwhile, the GAT-based critic models evolving interaction structures for accurate global value estimation. The proposed framework is validated on real differential-drive robot platforms, and experimental results demonstrate superior stability and safety in dynamic scenarios with limited fields-of-view.