Exosuit Control Optimization for Personalized Gait Assistance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Wearable exosuits and exoskeletons exhibit significant variability in effectiveness across different individuals due to the need for individualized assistance patterns, leading to inconsistent metabolic benefits and reduced efficacy.

Innovation Solution

The development of systems and methods to optimize actuation parameters in real-time using wearable sensors, allowing for adjustments in actuation timing, amplitude, rate, and profile shape to maximize or minimize objective functions such as energy expenditure, comfort, and operational efficiency, through gradient-descent or Bayesian optimization approaches.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a one-size-fits-all control approach is used for exosuits, then device complexity is reduced, but effectiveness and metabolic benefits vary significantly across different individuals

Engineering Contradiction:
Improvecontrol approachVSAvoideffectiveness
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The control system dynamically adapts actuation parameters in real-time based on individual user characteristics and gait patterns. The controller modifies actuation timing, amplitude, and profile shape during operation to optimize effectiveness for each user, transforming a static one-size-fits-all approach into a dynamic personalized system.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes multiple actuation parameters simultaneously including timing, amplitude, rate, and profile shape to tailor the assistance pattern to each individual user. These parameter adjustments are made in real-time based on measured performance metrics, allowing the exosuit to adapt to individual variability without requiring complete redesign for each user.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If actuation parameters are optimized in real-time for each individual, then effectiveness and metabolic benefits are maximized, but device complexity and computational requirements increase

Engineering Contradiction:
ImproveeffectivenessVSAvoidcontrol system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The control system uses feedback from sensors measuring metabolic cost, gait parameters, and performance metrics to continuously adjust actuation parameters. This closed-loop feedback mechanism enables real-time optimization of effectiveness while keeping the control architecture manageable through systematic parameter adjustment based on measured outcomes.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary identification of individual user characteristics and gait patterns during initial operation or calibration phases. This preliminary action establishes a baseline model for each user that guides subsequent real-time parameter optimization, reducing the computational burden during actual operation by pre-characterizing user-specific parameters.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If multiple actuation parameters are adjusted simultaneously, then individualized assistance patterns are achieved, but control difficulty and optimization complexity increase

Engineering Contradiction:
Improveindividualized assistanceVSAvoidcontrol optimization
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The control system segments the optimization process into distinct parameter groups or adjustment stages. Rather than optimizing all parameters simultaneously, the system divides the complex multi-parameter optimization into manageable segments, adjusting timing, amplitude, and profile shape in a structured sequence or through modular optimization routines.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3487666B1Controls optimization for wearable systems
Publication Date: 2024.11.13 PRESIDENT & FELLOWS OF HARVARD COLLEGE
  • EP3487666B1 patent drawingFigure 1
  • EP3487666B1 patent drawingFigure 2A~2B
  • EP3487666B1 patent drawingFigure 3A~3B

AI summary

A wearable system comprising an exosuit or exoskeleton; an actuator(s) configured to generate force in the exosuit or exoskeleton; a sensor(s) configured to measure information for evaluating an objective function associated with providing physical assistance to the wearer, an interaction between the wearer and the exosuit or exoskeleton, and/or an operation of the exosuit or exoskeleton; and a controller(s) configured to: actuate the actuator(s) according to an actuation profile(s), evaluate the objective function based on the information measured by the at least one sensor to determine a resulting change in the objective function, adjust a parameter(s) of the actuation profile(s) based on the resulting change in the objective function, and continue to actuate, evaluate, and adjust to optimize the actuation parameter(s) for maximizing or minimizing the objective function. Wearable systems configured to assist or promote an improvement in the wearer's gait and optimized using a gradient descent or Bayesian approach.