The Iterative Learning Gain That Optimizes Real-Time Torque Tracking for Ankle Exoskeletons in Human Walking Under Gait Variations.

Juanjuan Zhang,Steven H Collins

doi:10.3389/fnbot.2021.653409

Abstract

Lower-limb exoskeletons often use torque control to manipulate energy flow and ensure human safety. The accuracy of the applied torque greatly affects how well the motion is assisted and therefore improving it is always of interest. Feed-forward iterative learning, which is similar to predictive stride-wise integral control, has proven an effective compensation to feedback control for torque tracking in exoskeletons with complicated dynamics during human walking. Although the intention of iterative learning was initially to benefit average tracking performance over multiple strides, we found that, after proper gain tuning, it can also help improve real-time torque tracking. We used theoretical analysis to predict an optimal iterative-learning gain as the inverse of the passive actuator stiffness. Walking experiments resulted in an optimum gain equal to 0.99 ± 0.38 times the predicted value, confirming our hypothesis. The results of this study provide guidance for the design of torque controllers in robotic legged locomotion systems and will help improve the performance of robots that assist gait.

Highlights

Being able to reduce interface impedance, increase the reactiveness of robotic systems and improve human safety and comfort (Haddadin et al, 2008; Lasota et al, 2014), torque control has been widely used in physical human-robot interactive systems
We investigated the effects of iterative learning gain on the torque tracking performance of lower-limb exoskeletons using an ankle exoskeleton driven by a uni-directional Bowden cable
The system we investigated was a tethered ankle exoskeleton made of an off-board real-time controller and geared motor, a uni-directional Bowden cable transmission with a series spring, and an exoskeleton frame that interfaced with the human body (Figure 1)

Summary

Introduction

Being able to reduce interface impedance, increase the reactiveness of robotic systems and improve human safety and comfort (Haddadin et al, 2008; Lasota et al, 2014), torque control has been widely used in physical human-robot interactive systems. This is especially true in lowerlimb systems, which help human bodies to locomote and were involved in high density of energy exchange. Improving torque control performance has always been an active interest in the field of lower-limb human-robot interactive systems

Objectives

Methods

Findings

Discussion

Conclusion