SLIM: Sim-to-Real Legged Instructive Manipulation via Long-Horizon Visuomotor Learning

TL;DR

This paper introduces SLIM, a low-cost, hierarchical visuomotor system trained in simulation that effectively performs long-horizon manipulation tasks in the real world with high success rates, using accessible hardware and techniques to bridge the sim-to-real gap.

Contribution

The paper presents a novel hierarchical reinforcement learning approach with a teacher-student training pipeline and sim-to-real transfer techniques for low-cost legged robots performing complex tasks.

Findings

Achieves nearly 80% success rate in real-world long-horizon tasks.

Operates at 1.5x the speed of expert teleoperation.

Demonstrates effective deployment in diverse environments.

Abstract

We present a low-cost legged mobile manipulation system that solves long-horizon real-world tasks, trained by reinforcement learning purely in simulation. This system is made possible by 1) a hierarchical design of a high-level policy for visual-mobile manipulation following task instructions, and a low-level quadruped locomotion policy, 2) a teacher and student training pipeline for the high level, which trains a teacher to tackle long-horizon tasks using privileged task decomposition and target object information, and further trains a student for visual-mobile manipulation via RL guided by the teacher's behavior, and 3) a suite of techniques for minimizing the sim-to-real gap. In contrast to many previous works that use high-end equipments, our system demonstrates effective performance with more accessible hardware -- specifically, a Unitree Go1 quadruped, a WidowX-250S arm, and a single wrist-mounted RGB camera -- despite the increased challenges of sim-to-real transfer. Trained fully in simulation, a single policy autonomously solves long-horizon tasks involving search, move to, grasp, transport, and drop into, achieving nearly 80% real-world success. This performance is comparable to that of expert human teleoperation on the same tasks while the robot is more efficient, operating at about 1.5x the speed of the teleoperation. Finally, we perform extensive ablations on key techniques for efficient RL training and effective sim-to-real transfer, and demonstrate effective deployment across diverse indoor and outdoor scenes under various lighting conditions.