Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning

by   Xingyou Song, et al.

Learning adaptable policies is crucial for robots to operate autonomously in our complex and quickly changing world. In this work, we present a new meta-learning method that allows robots to quickly adapt to changes in dynamics. In contrast to gradient-based meta-learning algorithms that rely on second-order gradient estimation, we introduce a more noise-tolerant Batch Hill-Climbing adaptation operator and combine it with meta-learning based on evolutionary strategies. Our method significantly improves adaptation to changes in dynamics in high noise settings, which are common in robotics applications. We validate our approach on a quadruped robot that learns to walk while subject to changes in dynamics. We observe that our method significantly outperforms prior gradient-based approaches, enabling the robot to adapt its policy to changes based on less than 3 minutes of real data.



There are no comments yet.


page 1

page 5

page 6

page 7


NoRML: No-Reward Meta Learning

Efficiently adapting to new environments and changes in dynamics is crit...

Fast Online Adaptation in Robotics through Meta-Learning Embeddings of Simulated Priors

Meta-learning algorithms can accelerate the model-based reinforcement le...

Evolvability ES: Scalable and Direct Optimization of Evolvability

Designing evolutionary algorithms capable of uncovering highly evolvable...

Meta Reinforcement Learning Based Sensor Scanning in 3D Uncertain Environments for Heterogeneous Multi-Robot Systems

We study a novel problem that tackles learning based sensor scanning in ...

A Framework for On-line Learning of Underwater Vehicles Dynamic Models

Learning the dynamics of robots from data can help achieve more accurate...

Model-Based Meta-Reinforcement Learning for Flight with Suspended Payloads

Transporting suspended payloads is challenging for autonomous aerial veh...

Is Fast Adaptation All You Need?

Gradient-based meta-learning has proven to be highly effective at learni...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.