Arquitectura de control centralizado usando un enfoque cuasi óptimo para la regulación de la marcha bípeda de un exoesqueleto de extremidades inferiores

Abstract

Robotic devices have reached an important role in modern society, where their presence is increasingly common in the industry and personal use. Specifically to the rehabilitation field, it is entirely beneficial to have robotic devices that help specific pathologies’ recovery processes. This project introduces the design and implementation of a quasi-optimal control architecture to manage distributed controllers of a single lower limb robotic exoskeleton. To achieve this, a control and communication topology of the exoskeleton is established based on a master node and three slave nodes. The communication system employs CAN protocol with Process Data Objects (PDO), which reduces communication times by not using acknowledgement (ACK). Comparing it with SDO, there is a significant improvement, so it demonstrates the potential of PDO to take full advantage of the CAN bus. This system is implemented in the central node utilizing CAN reading and writing functions for frame management. The control system is also implemented in this node, which sends the necessary orders to each of the EPOS4 distributed controllers with an internal speed and position controller for the application of gait trajectory. Furthermore, it includes a heuristic-based optimizer that analyzes the error to determine when a correction of the control signal is necessary to apply to minimize the trajectory tracking error. The result is a functional single lower limb exoskeleton with three joints: hip, knee, and ankle. Control and communication algorithms showed an improvement over previous versions regarding the time of sending and receiving data through CAN bus, as well as its processing in the central node. In consequence, this allows planning an extension of the algorithms for an additional limb. This project makes it possible to clarify the capabilities of distributed nodes to establish solid bases for future research with the prototype. In this way, the subsequent studies could focus on improving the control system and a possible paradigm change. This would imply performing several tests with the prototype in the rehabilitation area to eventually improve people’s health status and their quality of life.