Abstract
Automation of industries has made robotic handling a significant enabler of the precision-oriented manufacturing systems. The paper presents an assembly automation system founded on robot manipulation, including an intelligent control algorithm, machine vision, and adaptive motion planning to achieve high precision in case of applying on a complex assembly job. The suggested system, consisting of a multi-axis robotic manipulator and trajectory optimization based on reinforcement learning, will ensure sub-millimeter assembly tolerance, reduced cycle time, and operational error. Simulation and experiment validation was used to test the performance of the systems with the help of ROS-Gazebo and MATLAB. Results indicate that the proposed solution is more efficient than conventional PID and rule-based automation systems with an error of 35 percent in assembly and a throughput increase of 28 percent. The ability to merge the feedback-based control and simulation of the digital twins enhances reliability, flexibility, and scalability.