Multi-robot system improves factory task assignment and navigation safety
Learning Multi-Agent Task Assignment and Navigation in the Factory: from Simulation to Real Robots
RoboticsMachine Learning
Summary
Coordinating multiple robots to handle tasks in a factory can be tricky and unsafe. The authors created a new method called FMAPPO that helps robots work together better by combining sensor data and task info. They tested this in simulations and on real robots, and found it made the robots faster and safer at moving parts around machines. Their method also keeps working well when commands update less often, which is useful for real-world robots.
What this means in practice
- •For factory automation engineers: Coordinate multiple robots to efficiently assign tasks and navigate factory floors while avoiding collisions using fused sensor and task data.
- •For autonomous vehicle developers: Improve decentralized control strategies for fleets of ground robots navigating complex environments with partial sensor inputs.
Authors
Abdalwhab Bakheet Mohamed Abdalwhab, Giovanni Beltrame, David St-Onge
Abstract
Reinforcement learning (RL) has shown considerable promise for robotic decision-making, yet deploying multi-agent RL (MARL) on physical multi-robot systems in industrial environments remains challenging. This paper investigates the real-world applicability of decentralized MARL for multi-robot multi-machine tending. We propose Feature-fusion Multi-Agent Proximal Policy Optimization (FMAPPO), which fuses 2D LiDAR measurements with task-specific state information to enable safe decentralized multi-robot task assignment and navigation. A complete simulation-to-reality pipeline was developed using high-fidelity robotic simulation and ROS2 and deployed on physical mobile-manipulator platforms operating under realistic real-world conditions, with the robotic arms disabled during the experiments. We further investigate the sensitivity of the learned policy to command update frequency, an important consideration for real-world deployment. Comparative evaluation in simulation demonstrated that FMAPPO significantly outperformed state-of-the-art baselines with a large effect size, achieving improvements of 106\% and 21\% in parts delivery and 48\% and 11\% in parts collection over MAPPO and SMAPPO, respectively. FMAPPO also increased machine utilization by 31 and 10 percentage points, respectively, while reducing collisions by 18\% and 15\% and increasing the safety score by 14 and 6 percentage points compared with MAPPO and SMAPPO, respectively. Furthermore, real-world experiments demonstrated that the learned decentralized policies can coordinate multiple robots to service multiple machines while maintaining safe operation under real-world sensing and control constraints. Videos of the real-world experiment are available online https://anonymouspapers123.github.io/FMAPPO/.