Design and Development of an Intelligent Collision Avoidance Platform for a Teleoperated Dual Robotic Arm

Abstract

ABSTRACT The aim of this study was to develop collision avoidance control for a teleoperated dual robotic arm to minimize collisions in human‐robot collaboration. Generally, in the absence of work cell mapping, collision checking in a teleoperated mobile robot is performed based on motor current or torque. However, this method cannot predict collisions that may occur during robot maneuvering. To overcome this problem, collision checking based on the description file model of the robot, and collision avoidance based on proximity sensors were designed for the robot's neighboring link and end‐effector, respectively. The UR10 robotic arm was modeled in MATLAB. The reachable point in the workspace of the robotic arm and the Geomagic Touch haptic device was calculated using the Adaptive Neuro Fuzzy Inference System (ANFIS) method. Two sets of experiments with different scenarios were carried out to detect and avoid collision in the neighboring link and end‐effector. The test results confirmed the effectiveness of the developed collision avoidance control performance in eliminating the risk of collision in the working environment of the teleoperated mobile robot. The approach presented in this study can be applied to almost any similar commercial robot as an independent or system‐integrated package.x