Exoskeleton Walking Control Using Tipping-State Stability Pools
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current exoskeletons for individuals with mobility issues struggle to achieve a stable and natural walking paradigm on varied terrains, often resulting in slow and unstable movement due to limitations in balance and direction management, which restricts access to uneven environments and increases the risk of falling.
Innovation Solution
A method and system for an exoskeleton that utilizes data processing means to generate commands for actuators to put the device in a tipping state, employing a database of virtual requirements parameterized by a phase variable to achieve an attracting stable trajectory, allowing the exoskeleton to walk autonomously and maintain balance by identifying suitable sets of virtual requirements for stability pools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If exoskeletons use traditional balance control methods with external assistance, then stability is maintained, but walking speed becomes extremely slow (a few meters per minute) and step length is limited to less than 30 cm
Solution Approach 1:
The exoskeleton autonomously manages its own balance by identifying tipping states and selecting appropriate control strategies from the database without requiring external assistance such as crutches or operators, enabling self-directed stable walking
Solution Approach 2:
The system pre-computes and stores multiple control strategies in the database corresponding to different tipping states and terrain conditions before actual walking, allowing rapid selection of appropriate control actions during operation without real-time computation delays
2Reliability
If exoskeletons maintain constant planar foot contact with the ground, then balance is maintained on flat surfaces, but the ability to navigate uneven terrains and obstacles is lost
Solution Approach 1:
The system dynamically adapts its control strategy based on the detected terrain conditions and current tipping state, selecting from multiple pre-computed strategies that accommodate different terrain types including uneven surfaces and obstacles, rather than relying on fixed planar contact assumptions
Solution Approach 2:
The control approach changes parameters such as foot placement position, contact duration, and body orientation based on terrain characteristics and tipping state, allowing the exoskeleton to adjust its walking pattern to suit different environmental conditions
3Device complexity
If exoskeletons use simplified control models, then device complexity is reduced, but the ability to handle unexpected terrain disturbances and maintain balance is compromised
Solution Approach 1:
Complex control strategies are pre-computed and stored in the database during the design phase, covering a wide range of tipping states and terrain conditions. During operation, the system simply retrieves and executes the appropriate pre-computed strategy, maintaining high reliability without requiring complex real-time computation
Solution Approach 2:
The database acts as an intermediary between the simple sensor inputs (tipping state detection) and the complex control actions required for stable walking, bridging the gap by providing pre-computed control strategies that map simple state observations to complex control sequences
4Reliability
If exoskeletons require external assistance for balance management, then stability is achieved on flat terrains, but autonomy and independence of operation are lost
Solution Approach 1:
The exoskeleton autonomously detects its tipping state, queries the database for appropriate control strategies, and executes the selected control actions without external assistance, achieving independent operation while maintaining balance stability
Solution Approach 2:
The system continuously monitors its state through sensors, identifies tipping states based on feedback from the environment and internal dynamics, and adjusts its control strategy accordingly, creating a closed-loop autonomous control system
Data Source
AI summary
The present invention relates to a method for moving an exoskeleton (1) receiving a human operator, said exoskeleton (1) having a plurality of degrees of freedom including at least one degree of freedom actuated and at least one non-actuated degree of freedom, the method being characterised in that it comprises the implementation of steps of:(a) when a start request is received, generating and emitting a command to at least one of said actuators so as to put the exoskeleton (1) in a tipping state;(b) in a database stored in data storage means (12) of triplets of:a set of virtual requirements on said actuated degrees of freedom,a controller for said exoskeleton (1) capable of generating commands of said actuators so as to fulfil said virtual requirements by implementing at least one attracting stable trajectory,a stability pool formed by all the points from which the execution of said controller allows a convergence to said attracting stable trajectory;identifying a set of virtual requirements such that said tipping state is included in said stability pool associated with this set of virtual requirements;(c) executing the controller associated with the set of virtual requirements identified such that the exoskeleton (1) walks.