THE BRS-IMPROVED A* ALGORITHM-BASED PATH PLANNING FOR DIGITAL TWIN OF COLLABORATIVE ROBOTS

Collaborative robots play an important role in production, assembly, operating systems, and can be used in many automation tasks. Due to the repetitive nature of robot actions, the overall efficiency of the production system is mainly determined by the precision of the robot's movement. This study proposes the optimization of robotic path planning relied on its digital twin combined with the A* algorithm. Unity is used to create the digital twin of the UR3 robot, where its path planning, based on an improved A* algorithm, is trained in a virtual environment before being implemented in the physical world. Some limitations associated with the traditional A* algorithm include redundant points, jagged paths and proximity to obstacles, increasing collision risks. Therefore, we developed the BRS-improved A* algorithm that incorporates buffer distance, redundant point elimination, and smoothing optimization using Bezier curves. Realistic movements for the robot are planned by simulated training in a virtual environment. In addition, the virtual robot's available path will be flexibly adjusted based on the physical robot's motion data, allowing for the re-adjustment of the physical robot's motion trajectory. This work uses a robot path planning algorithm to optimize the robot's path by reducing errors in the physical robotic path through interaction between virtual and real data. Additionally, optimizing the robot's path reduces the distance it needs to travel, thereby increasing energy efficiency for both the robot and the entire system.