• Myo-Electric Sensor system for precise robot control

      Rana, Khaqan-Jim (University of Bedfordshire, 2019-11)
      Robotic prosthetics has been a field of great interest in recent years and much work has been conducted in the various disciplines that it consists of. Among these disciplines is the research on sensor technology which used to enable control of such prosthesis through muscle activation. Responsiveness and accuracy is vital to implement a functional sensor system for prosthesis, as such this thesis will present the research and development of a sensor system used to control a robotic prosthesis as well as a feedback system which compares the position of the robot fingers and the intended movement in order to correct the servo motor position. These sensor systems are developed to produce precise robot control of prostheses without causing amplification errors. The research will discuss the suitability of different sensors for the sensing of the muscle activity of the user and sensors for the development of the feedback system and describe their implementation and processing. In addition to this, different configuration of sensors and code will be employed and compared, so that the most suitable configuration is found, which is the configuration that is responsive to the muscle activation of the user, eliminates noise and prevents amplification errors, and enables movement of analogue manner rather than digital in order to create a natural feeling control of a prosthesis which imitates the intention of the user. Furthermore, the cost of commercially available robotic prostheses are expensive, making it inaccessible to lower-income users and people within conflict zones who are in need of such technology, thus the research will focus on using inexpensive components and material to lower the production costs in order to raise the accessibility of robotic prostheses to people in conflict zones and countries of low income. This research shows that the implementation of the proposed sensor system and feedback system indeed enables analogue mannered, responsive and accurate control of a robotic hand while preventing amplification errors, and the use of commonly available components and low-cost material is a viable option.