Goal-Driven Dynamics Learning via Bayesian Optimization

TL;DR

This paper introduces a task-specific, Bayesian optimization-based method for learning local linear dynamics models to improve control performance in complex robots, demonstrated through simulations and real quadrotor experiments.

Contribution

It proposes a novel active learning framework that directly optimizes dynamics models for control performance rather than true dynamics, using Bayesian optimization.

Findings

Effective in improving control performance on a quadrotor

Successful in both simulation and real-world experiments

Demonstrates rapid convergence to desired performance

Abstract

Real-world robots are becoming increasingly complex and commonly act in poorly understood environments where it is extremely challenging to model or learn their true dynamics. Therefore, it might be desirable to take a task-specific approach, wherein the focus is on explicitly learning the dynamics model which achieves the best control performance for the task at hand, rather than learning the true dynamics. In this work, we use Bayesian optimization in an active learning framework where a locally linear dynamics model is learned with the intent of maximizing the control performance, and used in conjunction with optimal control schemes to efficiently design a controller for a given task. This model is updated directly based on the performance observed in experiments on the physical system in an iterative manner until a desired performance is achieved. We demonstrate the efficacy of the proposed approach through simulations and real experiments on a quadrotor testbed.