Combining model-predictive control and predictive reinforcement learning for stable quadrupedal robot locomotion

TL;DR

This paper presents a hybrid control approach combining model-predictive control and reinforcement learning to enable stable, real-time quadrupedal robot locomotion, overcoming limitations of each method individually.

Contribution

The work introduces a novel hybrid control framework that integrates MPC with RL using a neural network-based Q-function, improving stability and computational efficiency for robot gait generation.

Findings

Achieves stable quadrupedal locomotion at short horizons where MPC alone fails.

Operates in real-time without prior training, demonstrating live adaptability.

Balances control performance with computational complexity effectively.

Abstract

Stable gait generation is a crucial problem for legged robot locomotion as this impacts other critical performance factors such as, e.g. mobility over an uneven terrain and power consumption. Gait generation stability results from the efficient control of the interaction between the legged robot's body and the environment where it moves. Here, we study how this can be achieved by a combination of model-predictive and predictive reinforcement learning controllers. Model-predictive control (MPC) is a well-established method that does not utilize any online learning (except for some adaptive variations) as it provides a convenient interface for state constraints management. Reinforcement learning (RL), in contrast, relies on adaptation based on pure experience. In its bare-bone variants, RL is not always suitable for robots due to their high complexity and expensive simulation/experimentation. In this work, we combine both control methods to address the quadrupedal robot stable gate generation problem. The hybrid approach that we develop and apply uses a cost roll-out algorithm with a tail cost in the form of a Q-function modeled by a neural network; this allows to alleviate the computational complexity, which grows exponentially with the prediction horizon in a purely MPC approach. We demonstrate that our RL gait controller achieves stable locomotion at short horizons, where a nominal MP controller fails. Further, our controller is capable of live operation, meaning that it does not require previous training. Our results suggest that the hybridization of MPC with RL, as presented here, is beneficial to achieve a good balance between online control capabilities and computational complexity.