Bio-Inspired Impedance Controller for User-Initiated Exoskeleton Gait
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing human-robot interaction systems for individuals with neuromuscular disorders or neurological conditions are often constraining and do not adequately mimic muscle behavior, leading to suboptimal assistance and safety issues during activities like walking, stairs climbing, and load lifting.
Innovation Solution
A finite-state controller with active impedance control is implemented, using a backdrivable actuation system and series elastic actuators to mimic muscle behavior, allowing smooth transitions and user-initiated phase changes based on voluntary motions, with a focus on hip flexion-extension during walking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a finite-state controller with state-constant impedance is used to assist gait, then the control architecture is simplified and easier to implement, but the system cannot adequately mimic muscle behavior and provides suboptimal assistance
Solution Approach 1:
The patent transitions from state-constant impedance to variable impedance control where the impedance parameters (stiffness, damping) dynamically change based on the current gait phase and user needs. This allows the system to mimic physiological muscle behavior more accurately while maintaining the simplicity of finite-state control architecture through phase-based transitions.
Solution Approach 2:
The controller modifies impedance parameters (stiffness k, damping b, equilibrium position α0) based on the detected gait phase and user state. These parameter changes enable the system to adapt its mechanical behavior to match natural muscle responses during different phases of walking, thereby improving assistance quality without increasing overall system complexity.
2Manufacturing precision
If trajectory and time dependent impedance control is used to correct gait, then the end-effector is attracted along a defined path, but the controller becomes constraining for the user
Solution Approach 1:
The system replaces rigid trajectory tracking with dynamic impedance control where the equilibrium position and impedance parameters adapt to the user's natural motion. This allows gait correction without constraining the user, as the assistance forces are generated based on the user's own movement patterns rather than imposing a predetermined trajectory.
Solution Approach 2:
The controller uses the user's own motion signals (joint angles, velocities) to generate appropriate assistance forces. The system serves itself by detecting the user's intent and natural gait characteristics, then providing complementary forces that enhance rather than constrain the user's voluntary movements.
3Speed
If triggered assistance is implemented mostly with the upper limbs, then the response can be event-driven and timely, but the system lacks adaptability for lower limb gait assistance
Solution Approach 1:
The patent develops a unified finite-state impedance control framework that can be applied to both upper and lower limb exoskeletons. The same control architecture and impedance modulation strategies work across different body segments and gait patterns, making the system universally applicable while maintaining event-driven responsiveness through phase detection.
Solution Approach 2:
The gait cycle is segmented into distinct phases (stance, swing, transition) with characteristic impedance profiles for each phase. This segmentation allows the system to provide timely, phase-appropriate assistance while maintaining adaptability across different gait conditions and body segments through the modular phase-based control structure.
4Ease of operation
If a backdrivable actuation system or series elastic actuator is used, then the system allows user-initiated motion with minimum force input, but the actuation complexity increases
Solution Approach 1:
Series elastic actuators introduce an elastic element as an intermediary between the motor and the load, allowing the system to store and release energy while maintaining backdrivability. This intermediary component enables natural user-initiated motion while the control system manages the complex interaction between user force and actuator response through impedance modulation.
Solution Approach 2:
The backdrivable actuation system dynamically adjusts its impedance characteristics to facilitate user-initiated motion during appropriate gait phases. The system transitions between high-impedance (energy storage) and low-impedance (energy release) states, creating a dynamic interaction that feels natural to the user while managing actuation complexity through phase-based control.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The method for controlling a single- or multi- powered robotic system, such as an exoskeleton, a prosthesis or a collaborative robot, that is physically interacting with a user, said system comprising at least one actuated joint; wherein the robot joint(s) is/are controlled in force by a low level controller using an impedance control; wherein the joint(s) output force(s) is/are determined by a high level controller using a finite state control; and wherein the high level controller finite state control is governed by a voluntary motion from the user reaching a predetermined trigger.