Intuitive optimal prosthesis tuning through user-tuned costs of effort and accuracy

Abstract

Clinicians and prosthesis users care about the practical attributes of movement like the physical effort required by the user, the response time of the device, the reliability, and the accuracy of the movement performed. But the calibration parameters of a prosthetic device are relatively abstract and do not directly correspond to those quantities that the users and the clinicians inherently care about. Here, we propose an intuitive tuning technique that allows clinicians to tune prostheses based on the things that end-users actually care about. We use well-established engineering techniques (optimal control) to determine the set of best possible solutions for different relative preferences of the user. This required optimizing the problem for multiple objectives, (effort, time, reliability, and accuracy) to compute the best tuning parameters for a wide range of trade-offs. By solving this optimization problem, the complexity of the relationship between the performance and the prosthesis parameters can be implemented as a mapping procedure, and thereby hidden from the user. This simplifies the calibration process and allows clinicians or users to intuitively customize the device for their individual needs.