Modification of muscle antagonistic relations and hand trajectory on the dynamic motion of Musculoskeletal Humanoid

TL;DR

This work tackles dynamic motion control in musculoskeletal humanoids by introducing two complementary strategies: antagonist modifier to suppress antagonist muscle tension via real-time tension sensing, and agonist modifier to correct hand trajectory using vision-based target tracking. The methods are validated through two experiments, first showing reduced internal forces with antagonist modulation, and second demonstrating successful badminton hitting when both modifiers are active. The key contribution is a tension-aware and vision-guided control framework that aligns physical motion with planned trajectories, enabling smoother, more accurate dynamic tasks in systems lacking full joint-angle sensing. This approach paves the way for broader motion capabilities and motivates future hardware and algorithmic extensions to whole-body musculoskeletal robots.

Abstract

In recent years, some research on musculoskeletal humanoids is in progress. However, there are some challenges such as unmeasurable transformation of body structure and muscle path, and difficulty in measuring own motion because of lack of joint angle sensor. In this study, we suggest two motion acquisition methods. One is a method to acquire antagonistic relations of muscles by tension sensing, and the other is a method to acquire correct hand trajectory by vision sensing. Finally, we realize badminton shuttlecock-hitting motion of Kengoro with these two acquisition methods.

Figures (11)

• Figure 1: System overview of this research.
• Figure 2: Musculoskeletal humanoids Kengorohumanoids2016:asano:kengoro.
• Figure 3: Figure of Agonist muscle and Antagonist muscle.
• Figure 4: System of antagonist modifier.
• Figure 5: Shuttle and racket.