CVT Based Design of a Knee Exoskeleton to Provide Partial Assistance During a Normal Walking Gait Cycle

Abstract

According to population statistics it is estimated that the people aging more than 65 years will rise to almost 98 million in the next 40 years which fosters the need of advancement in technology that can help old people in maintaining their independence in terms of mobility and strength. For this reason researchers have developed many exoskeleton devices which use actuators and sensors to help the people in sitting, standing and walking. This thesis provides a crisp review of state-of-the-art in exoskeletons by providing a classification methodology, comparing their constructional features and also critically analysing the pros and cons of systems which have been realized. Regulatory issues, important for realizing new markets for the technologies, are also explored in this work. Main focus of the thesis is to explore new technologies to make the exoskeleton more cost effective so that it can be accessed by the masses. Torque variation methods of the exoskeletons are surveyed and a new cost effective solution is investigated by using Continuous Variable Transmission (CVT) and conventional mechanisms. The torque range data for the optimization purpose is taken from the experimental results from the Swedish group. CVT is then coupled with a Scotch Yoke mechanism for motion reversal and four-bar chain for poly-centricity at the knee joint by developing a mathematical model. Then geometric modeling of the complete assembly is made and simulation is done in MATLAB/Simulink. For validation, the results are compared with walking gait cycle of the normal healthy human being and observations form KINECT sensor. Future work will focus on finding more alternate cost effective methods and the experimental validation of the results by prototyping and testing on human subject.