Multiple manipulators path planning using double A*

Streamlining automated processes is currently undertaken by developing optimization methods and algorithms for robotic manipulators. This paper aims to present a new approach to improve streamlining of automatic processes. This new approach allows for multiple robotic manipulators commonly found in the industrial environment to handle different scenarios, thus providing a high-flexibility solution to automated processes.

The developed system is based on a spatial discretization methodology capable of describing the surrounding environment of the robot, followed by a novel path-planning algorithm. Gazebo was the simulation engine chosen, and the robotic manipulator used was the Universal Robot 5 (UR5). The proposed system was tested using the premises of two robotic challenges: EuRoC and Amazon Picking Challenge.

The developed system was able to identify and describe the influence of each joint in the Cartesian space, and it was possible to control multiple robotic manipulators safely regardless of any obstacles in a given scene.

This new system was tested in both real and simulated environments, and data collected showed that this new system performed well in real-life scenarios, such as EuRoC and Amazon Picking Challenge.

The new proposed approach can be valuable in the robotics field with applications in various industrial scenarios, as it provides a flexible solution for multiple robotic manipulator path and motion planning.