Adaptive robot movement improves precision in critical task areas

Interaction-Stiffness-Guided Basis Allocation in Dynamic Movement Primitives for Efficient Skill Transfer

Robotics

Summary

Robots learn movements using a method called Dynamic Movement Primitives (DMPs), but traditional ways give equal attention to all parts of a movement, even if some parts need more precision. The authors propose a new approach called Stage-Criticality-Guided DMPs that helps robots focus more on important parts of a movement by using clues from how hard it is for a human to interact with the robot and how consistent the shared examples are. This method redistributes how the robot represents movements so it’s more detailed where precision matters most and simpler elsewhere. Tests on handwriting and real robot tasks show this method improves accuracy without making the model more complex.

What this means in practice

Authors

Chan Xu, Silu Chen, Dehao Wang, Xiyu Chen, Dexin Jiang, Chi Zhang, Guilin Yang, Chenguang Yang, Zaojun Fang

Abstract

Dynamic Movement Primitives (DMPs) provide a compact and stable formulation for trajectory representation and generalization in robot skill learning. However, their predefined basis layout limits the allocation of approximation capacity according to stage-dependent precision requirements. To address this issue, this article proposes Stage-Criticality-Guided Dynamic Movement Primitives (SC-DMPs) with adaptive basis allocation for precision-critical skill learning. Operator-robot interaction stiffness and a trajectory-consistency cue derived from cross-demonstration task-space variability are integrated to construct a stage-criticality index. Guided by this index, basis centers are redistributed in normalized time through inverse cumulative criticality and mapped to the canonical phase domain, while their bandwidths are refined to adjust local approximation support. This enables denser and more flexible representation at high-criticality stages while retaining sparser allocation elsewhere. Experiments on handwriting trajectories and three real-robot tasks show that the inferred criticality is concentrated in geometrically demanding and task-constrained regions. Comparisons with DMPs, ProMPs, ProDMP, GP-MP, and KMP demonstrate improved trajectory reproduction, endpoint generalization, and task-critical accuracy while retaining a compact model and the stable structure of classical DMPs.
View PDFOpen arXiv