DOOMED: Direct Online Optimization of Modeling Errors in Dynamics

TL;DR

This paper introduces DOOMED, an online learning algorithm that directly minimizes modeling errors in robot dynamics to improve control accuracy in real time, even without explicit knowledge of the true system dynamics.

Contribution

It proposes a novel gradient-based online optimization method for real-time correction of inverse dynamics errors during robot control.

Findings

Effective real-time correction of dynamics errors demonstrated

Improved tracking accuracy in robotic control tasks

Online learning adapts to changing system dynamics

Abstract

It has long been hoped that model-based control will improve tracking performance while maintaining or increasing compliance. This hope hinges on having or being able to estimate an accurate inverse dynamics model. As a result, substantial effort has gone into modeling and estimating dynamics (error) models. Most recent research has focused on learning the true inverse dynamics using data points mapping observed accelerations to the torques used to generate them. Unfortunately, if the initial tracking error is bad, such learning processes may train substantially off-distribution to predict well on actual observed acceleration rather then the desired accelerations. This work takes a different approach. We define a class of gradient-based online learning algorithms we term Direct Online Optimization for Modeling Errors in Dynamics (DOOMED) that directly minimize an objective measuring the divergence between actual and desired accelerations. Our objective is defined in terms of the true system's unknown dynamics and is therefore impossible to evaluate. However, we show that its gradient is measurable online from system data. We develop a novel adaptive control approach based on running online learning to directly correct (inverse) dynamics errors in real time using the data stream from the robot to accurately achieve desired accelerations during execution.