SRL: Combining SLIP Model and Reinforcement Learning for Agile Robotic Jumping

2026-06-16

AI Summary

This work addresses the limitations of traditional Spring-Loaded Inverted Pendulum (SLIP) models on irregular terrain, which suffer from overly idealized assumptions, as well as the poor sample efficiency and high data demands of pure reinforcement learning approaches. To overcome these challenges, the authors propose a hybrid control framework termed SRL that, for the first time, integrates the physical priors of the SLIP model—implemented as feedforward control—with the adaptive feedback capabilities of deep reinforcement learning into a unified feedforward–feedback architecture. This synergy substantially improves training efficiency and environmental adaptability. The approach enables stable hopping on both bipedal and quadrupedal robotic platforms, achieving position tracking errors below 0.1 meters and velocity errors within ±3%. Successful sim-to-real transfer further validates the framework’s robustness and practicality.

Abstract

Robotic jumping is pivotal in applications such as search and rescue and logistics, where crossing obstacles and enhancing mobility efficiency are critical. The Spring-Loaded Inverted Pendulum (SLIP) model leverages simplified spring-mass dynamics that naturally encode biologically plausible hopping motions, yet its performance degrades on irregular terrain due to idealized assumptions regarding contact and joint dynamics. Meanwhile, Reinforcement Learning (RL) can adapt to diverse and complex environments but often requires extensive data from unguided exploration. The complementary strengths of SLIP's physically grounded baseline and RL's adaptive capabilities motivate a hybrid framework that overcomes these individual limitations. We therefore propose Spring-loaded Reinforcement Learning (SRL), which integrates SLIP-based feedforward control signals with RL-driven real-time feedback, enabling continuous optimization of robotic jumping. Experimental results demonstrate that SRL can achieve more stable jumps with much less training time than the baseline method, maintaining an average position tracking error below 0.1 m and velocity tracking errors within +/-3% of the target values. Through bipedal and quadrupedal simulations of ground and stair jumping, as well as sim-to-sim and sim-to-real validations, SRL exhibits robust adaptability to various task requirements and environmental complexities, underscoring its potential for real-world deployment.

Problem

• robotic jumping
• SLIP model
• reinforcement learning
• terrain adaptability
• mobility efficiency

Innovation

• Spring-Loaded Inverted Pendulum
• Reinforcement Learning
• hybrid control
• robotic jumping
• sim-to-real transfer