Learning Multi-Agent Local Collision-Avoidance for Collaborative Carrying tasks with Coupled Quadrupedal Robots

2026-03-24

AI Summary

This work addresses the challenge of achieving collision-free coordinated motion for multi-robot cooperative transport in unknown, obstacle-rich environments. The authors propose a reinforcement learning–based hierarchical control strategy that enables two quadruped robots, connected via a spherical joint, to follow a commanded velocity direction and avoid obstacles in real time using only onboard perception—without relying on precomputed trajectories or a global map. A novel object-centric high-level perception mechanism, combined with a game-inspired curriculum training strategy, enhances the system’s adaptive coordination capabilities in dynamic and complex settings. Hardware experiments demonstrate that the proposed approach significantly outperforms baseline methods based on optimization and decentralized reinforcement learning in previously unseen environments.

Technology Category

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

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Robotic collaborative carrying could greatly benefit human activities like warehouse and construction site management. However, coordinating the simultaneous motion of multiple robots represents a significant challenge. Existing works primarily focus on obstacle-free environments, making them unsuitable for most real-world applications. Works that account for obstacles, either overfit to a specific terrain configuration or rely on pre-recorded maps combined with path planners to compute collision-free trajectories. This work focuses on two quadrupedal robots mechanically connected to a carried object. We propose a Reinforcement Learning (RL)-based policy that enables tracking a commanded velocity direction while avoiding collisions with nearby obstacles using only onboard sensing, eliminating the need for precomputed trajectories and complete map knowledge. Our work presents a hierarchical architecture, where a perceptive high-level object-centric policy commands two pretrained locomotion policies. Additionally, we employ a game-inspired curriculum to increase the complexity of obstacles in the terrain progressively. We validate our approach on two quadrupedal robots connected to a bar via spherical joints, benchmarking it against optimization-based and decentralized RL baselines. Our hardware experiments demonstrate the ability of our system to locomote in unknown environments without the need for a map or a path planner. The video of our work is available in the multimedia material.

Problem

• multi-agent coordination
• collision avoidance
• quadrupedal robots
• collaborative carrying
• obstacle-rich environments

Innovation

• Reinforcement Learning
• Multi-Agent Coordination
• Local Collision Avoidance
• Quadrupedal Robots
• Onboard Sensing