Co-jump: Cooperative Jumping with Quadrupedal Robots via Multi-Agent Reinforcement Learning

2026-02-11

AI Summary

This work addresses the limited jumping performance of individual quadrupedal robots due to actuation constraints by proposing a multi-agent reinforcement learning approach based on MAPPO. Two quadruped robots achieve coordinated jumping without explicit communication or predefined motion primitives, relying solely on proprioceptive feedback. A progressive curriculum strategy is employed to overcome the challenges of sparse rewards and high-impact contact dynamics. Experiments demonstrate successful multi-directional jumps onto platforms as high as 1.5 meters in both simulation and real-world settings, with individual leg tip elevation reaching 1.1 meters—representing a 144% improvement over independent jumping. This study presents the first demonstration of communication-free, proprioception-driven cooperative jumping, establishing a novel paradigm for collaborative locomotion in constrained environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Multiagent Systems: Multiagent Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

While single-agent legged locomotion has witnessed remarkable progress, individual robots remain fundamentally constrained by physical actuation limits. To transcend these boundaries, we introduce Co-jump, a cooperative task where two quadrupedal robots synchronize to execute jumps far beyond their solo capabilities. We tackle the high-impulse contact dynamics of this task under a decentralized setting, achieving synchronization without explicit communication or pre-specified motion primitives. Our framework leverages Multi-Agent Proximal Policy Optimization (MAPPO) enhanced by a progressive curriculum strategy, which effectively overcomes the sparse-reward exploration challenges inherent in mechanically coupled systems. We demonstrate robust performance in simulation and successful transfer to physical hardware, executing multi-directional jumps onto platforms up to 1.5 m in height. Specifically, one of the robots achieves a foot-end elevation of 1.1 m, which represents a 144% improvement over the 0.45 m jump height of a standalone quadrupedal robot, demonstrating superior vertical performance. Notably, this precise coordination is achieved solely through proprioceptive feedback, establishing a foundation for communication-free collaborative locomotion in constrained environments.

Problem

• cooperative jumping
• quadrupedal robots
• multi-agent reinforcement learning
• physical actuation limits
• synchronization

Innovation

• multi-agent reinforcement learning
• cooperative locomotion
• quadrupedal robots
• communication-free coordination
• curriculum learning