Development and Implementation of a Collaborative Workspace for Industrial Robots Utilizing a Practical Path Adaptation Algorithm and Augmented Reality

The idea of humans and robots coexisting in manufacturing environments has increased the importance of personnel safety. Current practice to provide human safety in a conventional industrial robotic application is to put the robot in a cell enclosed by fences that occupies a large area in the environment. If someone enters this robotic cell, the robot ceases its movement immediately that leads to significant delays in the production cycle. In this paper, we present a new collaborative workspace concept for conventional industrial robots in order to eliminate fences and to avoid delays as much as possible. The proposed concept relies on a novel path adaptation algorithm and an augmented reality-based warning system. The algorithm, which makes the robot adaptable to human interventions, is unique in a sense that it can be applied to variety of industrial robots without requiring changes in the robot controller. Besides, the warning system provides additional safety features to the collaborative workspace through a projector and a head-mounted display. We demonstrate the feasibility of the proposed concept with an industrial robot in a realistic palletizing application. Several experiments were carried out to test the performance along with a user study. Performance test results indicate that the robot is able to perform three actions according to the minimum distance between the human and the robot in case of a human violation: slowdown, avoidance, and full stop. Qualitative performance tests results show that augmented reality improves the perception of safety and collaboration but negatively affects physical comfort. Overall, the proposed concept provides a practical and implementable solution for conventional industrial robots missing collaborative features.