Development and analysis of optical-based tactile sensor for measuring surface traction kinetics / Bakri Ali

• Main Author: Ali, Bakri
• Format: Thesis
• Language: English
• Published: 2019
• Subjects: Mechanical movements; Robotics. Robots. Manipulators (Mechanism)
• Online Access: http://ir.uitm.edu.my/id/eprint/28287/1/28287.pdf; http://ir.uitm.edu.my/id/eprint/28287/

Research on tactile sensors is expanding due to the need for autonomous dexterous robotic gripper to perform various tasks that need manipulation skills in medical operation, humanoid robots, household work as well as space exploration. These applications need a tactile sensor that can react to the variable force or pressure similar to human tactile sensing capability. To date, several researchers have explored several materials for tactile sensors ranging from the semiconductor based materials to silicon based rubber materials. This research is focused on the development and analysis of a spherical silicone tactile sensor based on a deformation image to measure the surface traction kinetics. A spherical tactile sensor was formed from a silicon rubber. It was marked with a circular mark inside the spherical dome to show changes in the diameter and shifts of the centroid. The tactile sensor was subjected to various normal and shear forces, and the circular mark image was recorded using an optical camera probe. The deformation image was digitized using a CCD camera and analysed using several digital image processing techniques such as image conversion from color to grayscale image, thresholding and boundary detection. The final image was then analysed to find the changes in the diameter and the shifts of the centroid, which were related to the normal and shear forces applied to the tactile sensor in x-axis, y-axis and z-axis. The results were tabulated and plotted in several graphs. The graphs showed that there was a significant correlation between the deformation image and the applied forces. This newly developed tactile sensor was proven to have a stiffness (kt) of 0.7238 N/mm and also to be able to detect the stiffness of the soft workpiece used in this research which was kw=0.156 N/mm. To conclude, the sensor can distinguish between hard and soft surfaces as well as between rough and smooth surfaces.