산업용 로봇암의 잔류진동 저감을 위한 포텐셜 필드 기반 충돌 회피 궤적 계획

Abstract

In industrial robots, preventing collisions with obstacles and creating smooth trajectories are critical issues. In previous research, an S-curve profile has been studied to reduce jerk. Many potential field method approaches have been researched to avoid collisions with obstacles in mobile robots. Typical potential fields generate a repulsive force for a robot to avoid obstacles inside the assumed boundary near the obstacle, but this repulsive force can affect robot motion that is above the maximum velocity because the repulsive force adds to the attractive force after avoiding an obstacle if the attractive and repulsive forces are in the same direction. This paper presents an asymmetric S-curve profile with a modified potential field, which generates a smooth velocity trajectory while avoiding the static obstacle and being over maximum velocity, as well as reducing residual vibration. The 6-axis robot uses a damped least square for singularity avoidance, and an augmented Jacobian is used with wrist parts added to the end-effector for a fixed pose. Simulations were conducted on MATLAB. We applied the simulation results to the NEUROMEKA INDY7 (6-axis) robot. According to the results, the proposed method produces a smoother trajectory than existing methods.