User preference optimization for control of ankle exoskeletons using sample efficient active learning
Data files
Oct 05, 2023 version files 712.05 KB
-
preference_data.zip
-
README.md
Abstract
A major challenge to the widespread success of augmentative exoskeletons is accurately adjusting the controller to provide cooperative assistance with their wearer. Often, the controller parameters are ``tuned'' to optimize a physiological or biomechanical objective. However, these approaches are resource-intensive, while typically only enabling optimization of a single objective. In reality, the exoskeleton user experience is derived from many factors, including comfort and stability, among others. This work introduces an approach to conveniently tune four parameters of the exoskeleton controller that maximize user preference. We use an evolutionary algorithm to recommend potential parameters, which are ranked by a neural network that is pre-trained with previously collected preference data. The controller parameters that have the highest preference ranking are provided to the exoskeleton, and the wearer provides feedback as forced-choice comparisons. Our approach was able to converge on controller parameters preferred by the wearer compared to randomized parameters with an accuracy of 88% on average. The result indicates that the proposed algorithm was able to identify users' preferences while requiring less than 50 queries to users. This work demonstrates user preference can be used to tune high-dimensional controller spaces easily and accurately, which shows the potential of translating lower-limb wearable technologies into our daily lives.
Methods
The data includes subjects' selections of control parameters throughout the experiment. The experiment consists of two sessions, which is the main optimization session and the validation session. The data shows users' selection process between two parameters (parent, offspring), which is composed of exploration / selection, and confirmation of the parameters they preferred throughout the session. There are a total of 13 subjects and each subject went through 3 trials of experiment. The speed of the treadmill was fixed to 1.2 m/s across all subjects. Please see the details of the protocol in the manuscript. The readme.txt describes the content of the data in each Excel file.
Usage notes
Microsoft Excel, Python