On-line Walking Speed Control in Human-Powered Exoskeleton Systems Based on Dual Reaction Force Sensors

On-line walking speed control in humanpowered exoskeleton systems is a big challenge, the translations of human intention to increase or decrease walking speed in maneuverable human exoskeleton systems is still complex field. In this paper, we propose a novel sensing technique to control the walking speed of the system according to the pilot intentions and to minimize the interaction force. We introduce a new sensing technology "Dual Reaction Force (DRF)" sensors, and explain the methodology of using it in the investigation of walking speed changing intentions. The force signals mismatch successfully applied to control the walking speed of the exoskeleton system according to the pilot intentions. Typical issues on the implementation of the sensory system are experimentally validated on flat terrain walking trails. We developed an adaptive trajectory frequency control algorithm to control the walking speed of HUman-powered Augmentation Lower Exoskeleton (HUALEX) within the human wearer intended speed. Based on the mismatch of DRF sensors, we proposed a new control methodology for walking speed control. Human intention recognition and identification through an sensorized footboard and smart shoe is achieved successfully in this work, the new term heel contact time H-CT is main feedback signal for the control algorithm. From the experimental walking trails we found that, the HCT during flat walking ranges from 0.69 +/- 0.05 sec and 0.41 +/- 0.07 sec while walking speed varies between 1m/s and 2.5m/s. The proposed algorithm used an Adaptive Central Pattern Generators (ACPGs) applied to control joint trajectory frequency, the different walking speeds associated with different functioning of human body CPGs frequency. We validated the proposed control algorithm by simulations on single Degree of Freedom (1-DoF) exoskeleton platform, the simulation results show the efficiency and validated that the proposed control algorithm will provides a good walking speed control for the HUALEX exoskeleton system.