Learning Control from Raw Position Measurements

TL;DR

This paper introduces VF-MC-PILCO, a model-based reinforcement learning algorithm that effectively learns control policies from raw position data without needing velocity estimators, simplifying implementation on mechanical systems.

Contribution

The paper presents VF-MC-PILCO, a novel velocity-free MBRL algorithm that models system dynamics using past positions and inputs, eliminating the need for velocity estimators.

Findings

VF-MC-PILCO performs comparably to previous algorithms with velocity estimators.

The method is validated on simulated and real mechanical systems.

It simplifies control learning by removing the need for state estimators.

Abstract

We propose a Model-Based Reinforcement Learning (MBRL) algorithm named VF-MC-PILCO, specifically designed for application to mechanical systems where velocities cannot be directly measured. This circumstance, if not adequately considered, can compromise the success of MBRL approaches. To cope with this problem, we define a velocity-free state formulation which consists of the collection of past positions and inputs. Then, VF-MC-PILCO uses Gaussian Process Regression to model the dynamics of the velocity-free state and optimizes the control policy through a particle-based policy gradient approach. We compare VF-MC-PILCO with our previous MBRL algorithm, MC-PILCO4PMS, which handles the lack of direct velocity measurements by modeling the presence of velocity estimators. Results on both simulated (cart-pole and UR5 robot) and real mechanical systems (Furuta pendulum and a ball-and-plate rig) show that the two algorithms achieve similar results. Conveniently, VF-MC-PILCO does not require the design and implementation of state estimators, which can be a challenging and time-consuming activity to be performed by an expert user.