Electrotactile Haptic Feedback for use with Remote Surgery Robots

This thesis documents the research, design, implementation and experimental testing of a prototype electrotactile haptic feedback for remote surgery applications, using a 3D-printed electrode array. A literature review was conducted to investigate the current knowledge surrounding the anatomy of the fingertip, electrical characteristics of the skin, somatosensory perception and current electrical stimulation and haptic feedback techniques. Following this, the design methodology was then presented to show the steps taken to achieve the final electrotactile haptic feedback device along with explanations of decisions made. Using the Voltera V-One PCB 3D printer, an electrode array was printed and designed to fit the average index fingertip. This process of printing required a high level of configuration to achieve the levels of accuracy required. This electrode array was connected to an electrical stimulation device capable of electronically controlling the polarity of each electrode. Finally, this device was connected to a VR environment. Following a Current Perception Threshold (CPT) test and a training period, participants were asked to determine the shape presented to them using only their tactile perception within a VR environment. In this single-blind experiment, the presented shape was obscured from the user's view while their fingertip was stimulated using electrotactile haptic feedback. Each of the five participants in this experiment were asked to determine five times what the randomized shape was. These participants correctly identified the shape in 56% of cases overall. When broken down by shape, the percentages were 100%, 40% and 50% for the triangle, square and circle respectively. The triangle is thought to have been identified more consistently due to its unique geometry. The square and circle were identified correctly less consistently; however, they were often mistaken for each other which can be seen in that 77.5% of non-triangle shapes were correctly identified as non-triangle shapes. There appeared to be some correlation between amplitude and the number of shapes correctly identified. Future works should investigate using a larger sample size to improve the reliability of the results across a broader demographic. In these works, it is suggested that a more intricate processing algorithm could be used to investigate creating spatial tactile cues on the fingertip as opposed to the temporal effects presented in this research. Research could also be directed at creating different textures using electrical stimulation rather than edge detection.