# Learning walk and trot from the same objective using different types of exploration

## TL;DR

This paper introduces a novel method for quadruped gait learning that encodes gait symmetry properties into the exploration process, enabling efficient learning of walk and trot gaits with improved performance.

## Contribution

It proposes a new approach to incorporate gait symmetry into policy search, enhancing exploration and learning efficiency for quadruped locomotion.

## Key findings

- Learned walk and trot gaits outperform random gaits.
- Encoding symmetry improves exploration efficiency.
- Performance significantly enhanced compared to baseline.

## Abstract

In quadruped gait learning, policy search methods that scale high dimensional continuous action spaces are commonly used. In most approaches, it is necessary to introduce prior knowledge on the gaits to limit the highly non-convex search space of the policies. In this work, we propose a new approach to encode the symmetry properties of the desired gaits, on the initial covariance of the Gaussian search distribution, allowing for strategic exploration. Using episode-based likelihood ratio policy gradient and relative entropy policy search, we learned the gaits walk and trot on a simulated quadruped. Comparing these gaits to random gaits learned by initialized diagonal covariance matrix, we show that the performance can be significantly enhanced.