A Robust Task-Level Control Architecture for Learned Dynamical Systems

TL;DR

This paper introduces L1-DS, a robust control architecture combining adaptive control and dynamic time warping to improve task-space tracking in robot motion planning, addressing unmodeled dynamics and disturbances.

Contribution

The paper presents a novel L1-augmented control framework that enhances DS-based learning from demonstration with robustness to mismatches and temporal misalignments.

Findings

Effective handling of task-execution mismatch in experiments

Improved phase-consistent tracking demonstrated on datasets

Robustness to unmodeled dynamics and disturbances

Abstract

Dynamical system (DS)-based learning from demonstration (LfD) is a powerful tool for generating motion plans in the operation (`task') space of robotic systems. However, the realization of the generated motion plans is often compromised by a ''task-execution mismatch'', where unmodeled dynamics, persistent disturbances, and system latency cause the robot's actual task-space state to diverge from the desired motion trajectory. We propose a novel task-level robust control architecture, L1-augmented Dynamical Systems (L1-DS), that explicitly handles the task-execution mismatch in tracking a nominal motion plan generated by any DS-based LfD scheme. Our framework augments any DS-based LfD model with a nominal stabilizing controller and an L1 adaptive controller. Furthermore, we introduce a windowed Dynamic Time Warping (DTW)-based target selector, which enables the nominal stabilizing controller to handle temporal misalignment for improved phase-consistent tracking. We demonstrate the efficacy of our architecture on the LASA and IROS handwriting datasets.