Decentralized Navigation of a Cable-Towed Load using Quadrupedal Robot Team via MARL

2025-03-23

AI Summary

Coordinating multiple quadrupedal robots to collaboratively drag cable-suspended loads while avoiding obstacles in cluttered, unstructured environments remains challenging due to complex rope-coupled dynamics and scalability constraints. Method: This paper proposes the first scalable, decentralized multi-agent reinforcement learning (MARL) planning framework explicitly designed for rope-coupled dynamics. It integrates centralized training with decentralized execution (CTDE), local-observation-based decentralized decision-making, physics-informed simulation modeling, and real-world deployment, augmented by a team-size-adaptive curriculum learning mechanism. Contribution/Results: The framework enables real-time coordination of 1–12 quadrupeds with constant inference latency; exploits rope slack/tension dynamics for autonomous navigation through narrow passages; demonstrates 1–4-robot collaboration on physical platforms and scales to 12 agents in simulation; and exhibits strong robustness against environmental disturbances and load variations.

Technology Category

• Intelligent Robots: Multi-Robot Systems
• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Decentralized Web and Fediverse systems

Abstract

This work addresses the challenge of enabling a team of quadrupedal robots to collaboratively tow a cable-connected load through cluttered and unstructured environments while avoiding obstacles. Leveraging cables allows the multi-robot system to navigate narrow spaces by maintaining slack when necessary. However, this introduces hybrid physical interactions due to alternating taut and slack states, with computational complexity that scales exponentially as the number of agents increases. To tackle these challenges, we developed a scalable and decentralized system capable of dynamically coordinating a variable number of quadrupedal robots while managing the hybrid physical interactions inherent in the load-towing task. At the core of this system is a novel multi-agent reinforcement learning (MARL)-based planner, designed for decentralized coordination. The MARL-based planner is trained using a centralized training with decentralized execution (CTDE) framework, enabling each robot to make decisions autonomously using only local (ego) observations. To accelerate learning and ensure effective collaboration across varying team sizes, we introduce a tailored training curriculum for MARL. Experimental results highlight the flexibility and scalability of the framework, demonstrating successful deployment with one to four robots in real-world scenarios and up to twelve robots in simulation. The decentralized planner maintains consistent inference times, regardless of the team size. Additionally, the proposed system demonstrates robustness to environment perturbations and adaptability to varying load weights. This work represents a step forward in achieving flexible and efficient multi-legged robotic collaboration in complex and real-world environments.

Problem

• Enabling quadrupedal robots to collaboratively tow a cable-connected load
• Managing hybrid physical interactions in cluttered environments
• Scalable decentralized coordination via MARL for varying team sizes

Innovation

• Decentralized MARL planner for quadrupedal robot team
• CTDE framework enables autonomous local decision-making
• Scalable curriculum training for varying team sizes