Inferring Smooth Control: Monte Carlo Posterior Policy Iteration with Gaussian Processes

TL;DR

This paper introduces a novel Monte Carlo posterior policy iteration method using Gaussian process priors to achieve smooth control in high-dimensional robot tasks, enhancing sample efficiency and control smoothness.

Contribution

It proposes a new inference-based control approach with Gaussian process priors and an optimized likelihood temperature for smoother, more effective robot control.

Findings

Achieves smooth control with high sample efficiency

Matches prior heuristic methods in performance

Demonstrates effectiveness on high-dimensional robot tasks

Abstract

Monte Carlo methods have become increasingly relevant for control of non-differentiable systems, approximate dynamics models and learning from data. These methods scale to high-dimensional spaces and are effective at the non-convex optimizations often seen in robot learning. We look at sample-based methods from the perspective of inference-based control, specifically posterior policy iteration. From this perspective, we highlight how Gaussian noise priors produce rough control actions that are unsuitable for physical robot deployment. Considering smoother Gaussian process priors, as used in episodic reinforcement learning and motion planning, we demonstrate how smoother model predictive control can be achieved using online sequential inference. This inference is realized through an efficient factorization of the action distribution and a novel means of optimizing the likelihood temperature to improve importance sampling accuracy. We evaluate this approach on several high-dimensional robot control tasks, matching the sample efficiency of prior heuristic methods while also ensuring smoothness. Simulation results can be seen at https://monte-carlo-ppi.github.io/.