Singularity-Free and Collision-Free Optimal Trajectory Planning for Multi Industrial and Collaborative Robots in Common Workspace

In this paper, a novel optimization algorithm was developed that plans collision-free, singularity-free and time-optimal trajectories with constrained position, velocity, acceleration, and jerk in the case of various operations of industrial and collaborative robots sharing a common workspace. The structure of the optimization algorithm used here includes Batch Belief Tree (BBT), Artificial Potential Field (APF), and A*. The proposed algorithm is tested on challenging trajectories. The performance of the proposed algorithm is compared with RRT*, RRT, BIT and Kinodynamic RRT algorithms. In the performance tests, seven industrial, collaborative, and redundant collaborative robots were simulated on RoboDK. The results indicate that the algorithm surpasses all compared tree-based algorithms in terms of success rate per trial (e.g., achieving a solution in 29 out of 30 attempts). Additionally, it consistently delivers reliable outputs while ensuring both jerk and time optimality.