Wearable Robot Control Using Adaptive Motor Primitives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current dynamic models for wearable robots lack adaptability in managing discrete or open-ended gestures, as they have fixed time evolution and predefined kinematics, which does not allow for true cooperative interaction with users, especially in applications like lifting loads where kinematics depend on load position and weight.
Innovation Solution
A method for estimating motion variables in discrete gestures using a motor primitive library with sequences of control steps that update motor primitives based on real-time sensor data and dynamic models, allowing predictive observation and adaptive control of wearable robots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fixed time evolution and predefined kinematics are used in dynamic models, then model simplicity is maintained, but adaptability to user's natural movement is lost
Solution Approach 1:
The patent applies the Dynamics principle by transitioning from static, predefined kinematic models to dynamic models that adapt in real-time. The system continuously updates motor primitive parameters based on sensor feedback and observed user movement, allowing the model to evolve during operation rather than remaining fixed. This enables the wearable robot to adjust to individual user characteristics and natural movement patterns.
Solution Approach 2:
The patent implements Parameter changes by modifying the parameters of motor primitives (such as duration, amplitude, and temporal characteristics) based on real-time observation of user movement. The system extracts movement parameters from sensors and uses these to dynamically adjust the kinematic model, allowing the same motor primitive library to adapt to different users and movement conditions without changing the fundamental model structure.
2Ease of operation
If predefined trajectories are imposed on users, then control precision is improved, but cooperative interaction with users is reduced
Solution Approach 1:
The patent applies the Feedback principle by continuously monitoring user movement through sensors and using this information to adjust the motor primitive parameters. The system compares observed movement with predicted movement from the dynamic model and refines the model parameters accordingly. This closed-loop approach maintains trajectory accuracy while adapting to the user's natural movement preferences and capabilities.
Solution Approach 2:
The patent implements Preliminary action by pre-computing a library of motor primitives that represent typical movement patterns. These primitives serve as initial predictions that are then refined in real-time based on actual user movement. This allows the system to have ready-made trajectory suggestions while still adapting to individual user characteristics during operation.
3Productivity
If fixed duration and kinematics are used for motor primitives, then computational efficiency is maintained, but ability to handle open-ended gestures is lost
Solution Approach 1:
The patent applies the Dynamics principle by making the duration and kinematic parameters of motor primitives dynamic rather than fixed. The system uses a dynamic model that can extend or compress the temporal evolution of motor primitives based on the actual duration of the user's gesture. This allows the same computational framework to handle both discrete gestures with fixed durations and open-ended gestures with variable durations.
Solution Approach 2:
The patent implementsUniversality by designing a motor primitive library and dynamic model framework that can handle multiple types of gestures (discrete and open-ended) using the same computational approach. The system can adapt the same motor primitives to different gesture types by adjusting parameters such as duration scaling and temporal evolution, maintaining computational efficiency while increasing versatility.
Data Source
AI summary
The present invention is in the sector of wearable robotic devices, and in particular it concerns a method for the adaptive control of a wearable robot. In particular, the invention discloses a method for the adaptive control of the discrete or open-ended movements of a wearable robot such as a prosthesis or an orthesis.