Bioinspired Hierarchical Electronic Architecture for Robotic Locomotion Assistance: Application in Exoskeletons

Abstract

This paper presents the conceptualization, design and development of a bioinspired electronic architecture based on the human motor system for the control of mobility assistance systems, such as assistance or rehabilitation exoskeletons. The proposed architecture is divided into three hierarchical levels: perception-intention, pattern generator, and execution, facilitating modularity, scalability, and parallelism in the execution and operation of the system. ROS2 was chosen as the middleware for communication management due to its ability to handle a large amount of data, robustness, and scalability. For validation in a real world scenario, the proposal was implemented in the gait rehabilitation robotic platform Discover2Walk. Among the advantages offered by this architecture, we highlight greater modularity, improved compatibility with programming languages, and scalability, in addition to ease of supervision and control. The architecture presented can be adopted in future robotic platforms and exoskeletons with built-in interoperability.