Safe Bayesian Optimization for the Control of High-Dimensional Embodied Systems

TL;DR

This paper introduces HdSafeBO, a novel high-dimensional safe Bayesian optimization algorithm that efficiently and safely optimizes control policies for complex, high-dimensional embodied systems, including human motion control, with safety guarantees.

Contribution

HdSafeBO is the first algorithm to enable safe, high-dimensional control optimization using local optimistic exploration and isometric embedding techniques.

Findings

Successfully optimized high-dimensional musculoskeletal control systems.

Achieved safety guarantees with probabilistic safety bounds.

Demonstrated real-world application in neural stimulation for human motion.

Abstract

Learning to move is a primary goal for animals and robots, where ensuring safety is often important when optimizing control policies on the embodied systems. For complex tasks such as the control of human or humanoid control, the high-dimensional parameter space adds complexity to the safe optimization effort. Current safe exploration algorithms exhibit inefficiency and may even become infeasible with large high-dimensional input spaces. Furthermore, existing high-dimensional constrained optimization methods neglect safety in the search process. In this paper, we propose High-dimensional Safe Bayesian Optimization with local optimistic exploration (HdSafeBO), a novel approach designed to handle high-dimensional sampling problems under probabilistic safety constraints. We introduce a local optimistic strategy to efficiently and safely optimize the objective function, providing a probabilistic safety guarantee and a cumulative safety violation bound. Through the use of isometric embedding, HdSafeBO addresses problems ranging from a few hundred to several thousand dimensions while maintaining safety guarantees. To our knowledge, HdSafeBO is the first algorithm capable of optimizing the control of high-dimensional musculoskeletal systems with high safety probability. We also demonstrate the real-world applicability of HdSafeBO through its use in the safe online optimization of neural stimulation induced human motion control.