Multi-Robot Multi-Goal Motion Planning with Stochastic Skills

2026-10-06

AI Summary

This study addresses the challenge of dynamic coordination and planning for multiple robots executing stochastic manipulation skills within shared workspaces. The proposed method integrates stochastic skills into sampling-based motion planning by formulating a Markov Decision Process (MDP) over a multimodal composite roadmap, thereby explicitly resolving skill uncertainty during the planning phase to synthesize reactive policies. This approach overcomes the pessimistic limitations inherent in conservative baselines, enabling robust dynamic coordination. Consequently, robots can adjust their motions in real time based on the execution states of other agents, significantly enhancing multi-robot collaboration efficiency.

Abstract

As robots are increasingly deployed in groups and share workspaces to execute real-world tasks, planning their concurrent motions around complex manipulation skills becomes essential. These skills involve continuous physical execution and may exhibit stochastic behavior, resulting in variable execution times and uncertain continuous trajectories. Existing planners either limit execution to single-robot scenarios, rely on open-loop paths, or use post-hoc scheduling that prevents dynamic coordination. In this paper, we address this gap by integrating stochastic skills into sampling-based multi-robot planning by formulating the problem as a Markov Decision Process (MDP) over a multi-modal composite roadmap. For stochastic skills, solving the MDP yields a reactive policy that allows controllable robots to dynamically adapt their motions in response to other robots' execution of manipulation skills. By resolving skill uncertainty directly at planning time, this approach avoids the pessimism of conservative baselines and unlocks robust, dynamic multi-robot coordination. Code for the planners is available at https://www.vhartmann.com/stochastic-skills.

Problem

• Multi-Robot Motion Planning
• Stochastic Skills
• Manipulation Uncertainty
• Dynamic Coordination

Innovation

• Multi-Robot Motion Planning
• Stochastic Skills
• Markov Decision Process
• Reactive Policy
• Multi-Modal Roadmap