Dynamic Human Detection and Adaptive Path Planning for Improved Human-Robot Collaboration in Industrial Settings

Abstract

Due to recent technological advancements in industrial working environments, human-robot collaboration has become a rising research topic in order to improve the efficiency and smoothness of the industrial workflow. One of its major concerns is the safety of both robots and humans. Thus, in this study a safe human-robot interactive algorithm is developed, by proposing a real-time human awareness, detection and avoidance algorithm, enabling the system to respond real-time to human presence and movement. In this study all of the links and joints of the robotic system is taken into consideration. The research exclusively focuses on human obstacles, prioritizing human safety in industrial environments. In addition, the Artificial Potential Field (APF) algorithm is employed for dynamic obstacle-free path planning of the robotic manipulator. This approach ensures that the robot can adapt its behavior based on dynamic environment changes. The proposed system is tested using simulation and real-life experiments on a human subject. Results validated the system objectives and its adaptability in a dynamic environment. The human detection algorithm demonstrated robustness to occluded body parts and utilized digital twinning for validation, ensuring correct behavior between the simulated twin and the hardware setup.