Abstract
The objective of this paper is to develop an estimation methid to extract wrist joint angle from a deformation of the forearm skin surface based on muscle contraction. We have been focused on the muscle bulge movement longitudinally along the forearm that derived from muscle contraction. We confirmed on to one relathinship between the muscle bulge movement and wrist joint angle and validate the feasibility to estimate joint angle using this relationship. However, the relationship is difficult to identify because of the misalignment of the sensor to a position of the muscle. Therefore, in this paper, we use the tactle sensor that can measure three-dimensional data of the foream skin surface for considering the estimation method from a muscle bulge mapping. We measured the large area (32[mm]×96[mm]) on the foream skin and the sensor has 48 distance sensor in this area. We calculated x and y coordinates of center of gravity from the measured data for deriving an equation that express the wrist joint angle. We compared the RMSE of estimated angle between the equation that uses x and y coordinate of the COG and the equation that uses only y coordinate of the COG. From the result, RMSE using the equation that uses only y coordinate of COG was smaller than anothre equation. Finally, we validated the fesibility of the estimation method that uses y coordinate of the COG from the measured data by tactile sensor. However, further research is needed to reduce the estimation error by improving the extraction of the relate expression.
