Simultaneous Optimization of Placement Planning and Motion Planning for a Single Robotic Arm Using Genetic Algorithm

Industrial robots are commonly used in factories for efficient high-mix low-volume production. It is required to determine the optimal placement and posture of a robot arm to pick up and place a workpiece to minimize the total operation time. We propose an efficient optimization approach for solving the placement planning and motion planning problems of a six-axis robot arm using a genetic algorithm (GA). To obtain a feasible solution to the problem of minimizing the total operation time, the proposed method conducts placement planning using GA and motion planning using Robot Operation System (ROS) simulation with a Rapidly-exploring Random Tree Star (RRT*) for a six-axis robot arm. The performance of the proposed algorithm is compared with that of a conventional method that uses particle swarm optimization. The computational results show that the proposed algorithm can reduce approximately 26% of the motion planning time compared to the conventional method. The proposed method can provide a near-optimal layout and motion planning solution within a practical computing time.

I. Introduction

In recent years, industrial robots have been commonly installed in limited spaces to realize high-mix low-volume production. Therefore, it is essential to determine the optimal placement and posture of the robot arm to reduce the processing time and costs of transporting workpieces. Such a decision problem can be regarded as an optimization problem in which the placement and posture of the robot arm should be determined at the factory design stage to reduce work time and cost. The robot placement planning problem involves decisions on where and how to place a robot with obstacles such that a pick-and-place operation for a robot arm is ensured. This process is typically performed using specialized equipment or manual labor, with a system centered around robots. The robot is then placed in a work area that maximizes its capabilities, and peripheral equipment such as fixtures and conveyors are arranged to plan an efficient production system that meets these objectives [1]. This study considers the placement planning problem for a pick-place operation using a single six-axis robot arm.