Goal-Driven Dynamics Learning via Bayesian Optimization

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.

This is the extended version of the CDC'17 paper titled "Goal-Driven Dynamics Learning via Bayesian Optimization."

References in corpus (3)

• Gaussian Processes for Data-Efficient Learning in Robotics and Control
• Safe Controller Optimization for Quadrotors with Gaussian Processes
• BayesOpt: A Bayesian Optimization Library for Nonlinear Optimization, Experimental Design and Bandits

Cited by in corpus (13)

• Temporal Difference Models: Model-Free Deep RL for Model-Based Control
• MPC Controller Tuning using Bayesian Optimization Techniques
• Objective Mismatch in Model-based Reinforcement Learning
• MBMF: Model-Based Priors for Model-Free Reinforcement Learning
• A Survey on Autonomous Vehicle Control in the Era of Mixed-Autonomy: From Physics-Based to AI-Guided Driving Policy Learning
• Gradient-Aware Model-based Policy Search
• Meta-Learning Acquisition Functions for Transfer Learning in Bayesian Optimization
• Mitigating Covariate Shift in Imitation Learning via Offline Data Without Great Coverage
• The Differentiable Cross-Entropy Method
• Preference-based MPC calibration
• Learning to Slide Unknown Objects with Differentiable Physics Simulations
• Context-Specific Validation of Data-Driven Models
• Accelerating Goal-Directed Reinforcement Learning by Model Characterization