Non-Coordinate Robotic Arm Control

Lim, Xian Zhu (2020) Non-Coordinate Robotic Arm Control. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text LimXianZhu_Full Text.pdf Restricted to Registered users only Download (6MB)

Abstract

Robotic arms have become the popular tools to help most of the industries to perform daily task, those tasks included pick and placed process, assembly process and etc. Usually, these robotic arms are controlled with an adapted joystick, or pre-programmed method. A robotic system that using coordination value to generate path planning can be known as coordinate base robotic arm control. Those method require more computation power and time consuming to setup. It is also sensitive to the external interference. Therefore, in order to overcome those disadvantages. A non-coordinate control of robotic arm was addressed as a new solution to replace old method (coordinate robotic arm control). In order to achieve the objective, a lot of factor included positioning, brightness of the environment and external interference to the robotic arm was taking into consideration. The biggest different of coordinate control and non-coordinate control is when they perform pick and place process, non-coordinate control system does not focus on the coordination of the target object, but only focus on the relationship of the end effector position and the target object position. In order to perform this function, an extra feedback system was added to the system to have a continuous checking and comparison between the position of the end effector and the position of the target object. The robotic arm movement changed will be made base on their relationship until the end effector reached to the exact position of the target object. Another Part attachment analysis system was also designed to collaborate with noncoordinate robotic arm control system. Where the part attachment analysis system has to understand the combination of the Lego, generate a series of instruction to operate the robotic arm, the instruction will only focus on which object need to be pick, and which object need to be place without requiring any coordinate value. The experiment was also carried out to measure the capability of the path attachment analysis system in term of its accuracy and duration of path planning. The proposed scheme is also compared with standard coordinate robotic arm control to compare the task execution duration and success rate. The result shows that it took around 41.42s to complete the task under unstructured environment during daytime however took lesser time which is around 36.03s during night time. The success rate of complete the task are around 85.48%. success rate. The overall performance of non-coordinate control are actually higher than coordinate control which had it only 64.25% success rate to complete the task. Even though coordinate control took lesser execute duration to control around 25s, however it took longer time to setup around 30 min to setup. This shows the advantage of non-coordinate control where it does not require any setup process and calibration process.