Phase Portraits as Movement Primitives for Fast Humanoid Robot Control

TL;DR

This paper introduces Phase Portrait Movement Primitives (PPMP), a low-dimensional, fast, and autonomous control method for humanoid robots that predicts dynamics using coupled oscillators, reducing computational complexity compared to traditional optimal control approaches.

Contribution

The paper presents PPMP, a novel primitive that models dynamics in phase space with coupled oscillators, enabling fast, efficient, and autonomous robot control without heavy reliance on model-based estimators.

Findings

PPMP enables real-time control on a 20-DOF humanoid robot.

The method achieves fast reaction times and anticipative pose adaptation.

PPMP reduces computational load compared to traditional control methods.

Abstract

Currently, usual approaches for fast robot control are largely reliant on solving online optimal control problems. Such methods are known to be computationally intensive and sensitive to model accuracy. On the other hand, animals plan complex motor actions not only fast but seemingly with little effort even on unseen tasks. This natural sense to infer temporal dynamics and coordination motivates us to approach robot control from a motor skill learning perspective to design fast and computationally light controllers that can be learned autonomously by the robot under mild modeling assumptions. This article introduces Phase Portrait Movement Primitives (PPMP), a primitive that predicts dynamics on a low dimensional phase space which in turn is used to govern the high dimensional kinematics of the task. The stark difference with other primitive formulations is a built-in mechanism for phase prediction in the form of coupled oscillators that replaces model-based state estimators such as Kalman filters. The policy is trained by optimizing the parameters of the oscillators whose output is connected to a kinematic distribution in the form of a phase portrait. The drastic reduction in dimensionality allows us to efficiently train and execute PPMPs on a real human-sized, dual-arm humanoid upper body on a task involving 20 degrees-of-freedom. We demonstrate PPMPs in interactions requiring fast reactions times while generating anticipative pose adaptation in both discrete and cyclic tasks.