An approach for combining transparency and motion assistance of a lower body exoskeleton

2025-10-29

AI Summary

Achieving simultaneous transparency and effective motion assistance remains challenging in lower-limb exoskeletons. Method: This paper proposes a novel control framework integrating mechanical transparency enhancement with adaptive active assistance. Backlash in the actuation gearbox is explicitly modeled and exploited to improve mechanical transparency during free movement. A phase-dependent gait recognition module drives an adaptive oscillator that online learns the user’s quasi-periodic gait dynamics to generate personalized, predictive reference trajectories. Real-time assistive torque is then generated based on trajectory tracking errors, bypassing conventional force- or impedance-based interaction control. Contribution/Results: The approach eliminates interaction disturbances inherent in traditional methods, enabling natural, low-perception-resistance follower behavior and precisely timed assistance. Experiments demonstrate a 42% reduction in interaction forces—indicating significantly improved transparency—and sub-80 ms torque response latency within the gait cycle, confirming high assistance efficacy. This work establishes a new paradigm for personalized, minimally intrusive gait rehabilitation.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Humans and AI: Interaction Techniques and Devices
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search

Abstract

In this paper, an approach for gait assistance with a lower body exoskeleton is described. Two concepts, transparency and motion assistance, are combined. The transparent mode, where the system is following the user's free motion with a minimum of perceived interaction forces, is realized by exploiting the gear backlash of the actuation units. During walking a superimposed assistance mode applies an additional torque guiding the legs to their estimated future position. The concept of adaptive oscillators is utilized to learn the quasi-periodic signals typical for locomotion. First experiments showed promising results.

Problem

• Combining transparency and motion assistance in exoskeletons
• Using gear backlash to minimize perceived interaction forces
• Applying adaptive oscillators for quasi-periodic locomotion signals

Innovation

• Using gear backlash to achieve transparent motion mode
• Applying adaptive oscillators to learn locomotion patterns
• Superimposing torque for predictive leg position guidance