Implementation of object detection and recognition algorithms on a robotic arm using Raspberry Pi circuit board

This thesis constructs an experimental setup to implement object detection and recognition algorithms on a robotic arm. 4 degrees of freedom robotic arm OWI–535 which is similar type to robotic arms used in industry is chosen for the experimental setup in this study. In the experimental studies, firstly a camera coverage area is arranged to cover all workspace of the robotic arm. Local feature based algorithms such as SIFT, SURF, FAST and ORB are used on the images which are captured via the camera to detect and recognize the target object to be grasped by the gripper of robotic arm. These algorithms are implemented in the software for object recognition and localization, which is written in C++ programming language using OpenCV library and the software runs on the Raspberry Pi circuit board. Secondly, the location information of target object is sent to control unit of the robotic arm after recognition and localization of the object. Then, the gripper of robotic arm grasps the object and moves to desired location. In these processes, the angles of the robotic arm's joints are determined by the solutions of inverse kinematics equations of the robotic arm. In the experimental studies, the performance of the features which are extracted with the algorithms such as SIFT, SURF, FAST, and ORB are compared on the data set. Moreover, all process steps of grasping the object and moving to the desired location relating to robotic arm's gripper are analyzed in detail. Keywords: Object detection and recognition algorithms, Feature extraction, Raspberry Pi circuit board, OWI–535 robotic arm, Inverse kinematics analysis.