Application of Learning Vector Quantization and Trajectory Planning On a 4-DoF Robotic Arm to Move the Object

Abstract

The robotic arm is a type of statistical robot with a limited range of movement. Robotic arms are generally within the scope of Cartesian coordinates, according to the specified link length. The development of robot technology leads us to continue to upgrade soft computing. Intelligent systems in robots can improve good navigation detection systems or carry out the operator's tasks. On the other hand, using a camera is an important part of finding clear information about objects or capturing the environment around the robot. In this research, we implemented an intelligent system and computer-based camera on a 4-DoF robotic arm system. This robotic arm consists of a computer as the main processor, a microcontroller to adjust the joint angle, additional electronics, and a camera to detect objects and classify them by color. The colors used are red, green, and blue. The learning process uses these colors using Learning Vector Quantization (LVQ). The implementation of LVQ also carries out pre-processing, training, and testing stages. In the experiments that have been carried out, the robotic arm successfully navigates toward the target object and moves the object using the Trajectory Planning method. This computing process is done on a computer and connected to the robot arm's microcontroller. The experiment was carried out 60 times, and the success rate was 95%. Overall, the robot successfully picked up objects and grouped them by color.