Adaptive Torque Control Applied to Fitness Machines

Abstract

A novel electric fitness machine which enhances the training efficiency and ensures the operator safety is proposed by this thesis Firstly Adaptive Impedance Controller (AIC) is employed for realizing the desired impedance between the operator and the fitness machine Its operation can be classified into active mode which is as same as traditional fitness machines and passive mode which can lead operators to follow an efficient trajectory consciously Secondly to replace a force/torque sensor the External Force Estimator (EFE) is used for sensorless impedance control and prevention from bandwidth reduction and high expense Thirdly if operators feel exhausted the desired impedance parameters will be adjusted by Fuzzy Inference System (FIS) in real-time to make operators finish the round of training completely under muscle contraction Lastly an electric fitness machine prototype is established to verify the overall control strategy based on the DS1104 interface setup According to the experimental results AIC can regulate the root-mean-square error of position tracking under 4 10 degree and EFE can regulate the root-mean-square error of the external torque estimation below 3 25 N-m The FIS to which the magnitude of external force is the third input can reduce the position tracking error region from [-29 49 0] degree to [-8 32 0] degree and significantly improve the capability of anti-disturbance

Date of Award	2020
Original language	English
Supervisor	Nan-Chyuan Tsai (Supervisor)