Hopping Control of a Monopod Robot: Reinforcement Learning vs Raibert Heuristic

As a classic control problem, monopod hopping was traditionally solved using a state machine-based approach known as the Raibert Heuristic. Recently, reinforcement learning (RL) has emerged as a powerful alternative for robotic control. In this paper, we apply RL to monopod hopping control and compare its performance with the traditional Raibert Heuristic. Simulations demonstrate that RL achieves faster hopping speeds and better robustness than the Raibert method. Both control strategies were implemented on a simulated monopod robot and evaluated based on hopping speed and robustness. The results highlight RL’s superiority in complex environments, while the Raibert approach remains a good solution for simpler scenarios. (The attached video can be found in https://www.bilibili.com/video/BV1eTxgeuEjV)