When creating walking human-machine systems for rehabilitation process of patients with musculoskeletal disorders, an important task is to plan the trajectories of the executive links. This paper considers a human-machine unit type exoskeleton for lower extremities rehabilitation equipped with controlled electric drives. The framework of the study introduces a method for obtaining the setting angles for the position control loops of the robot links based on the use of a fully connected neural network. The paper describes the data acquisition process for training and validation of the neural network, as well as the structure of the neural network model and the results of training and testing. A control diagram of the system based on a neural network strategy is proposed, and also a comparative analysis between the proposed method and a direct analytical method is being performed. In order to evaluate the effectiveness and applicability of the proposed method for solving trajectory planning strategy problems for the links of a walking human-machine system, a discussion of the main obtained results is carried out.
CITATION STYLE
Jatsun, S. F., Malchikov, A. V., Postolniy, A. A., & Yatsun, A. S. (2022). Trajectory Planning Strategy for the Links of a Walking Human-Machine System Using a Neural Network. In Lecture Notes in Networks and Systems (Vol. 324 LNNS, pp. 255–261). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-86294-7_22
Mendeley helps you to discover research relevant for your work.