Experimental validation of a collision avoidance strategy for collaborative robots

Collaborative robotics is revolutionizing the modern workspace by enabling safe and efficient human-robot interaction. However, the inherent complexity of shared environments presents significant challenges in ensuring both productivity and safety. This thesis aims to build upon existing state-of-the-art solutions by proposing a novel collision avoidance strategy that leverages the capabilities of a depth and RGB camera to monitor human motion. The proposed strategy is validated through a simulation framework, analyzing the robot’s kinematics and the performance of the collision avoidance algorithm. Additionally, experiments are conducted in a laboratory setting to apply the method to a real-world industrial robot the Techman TM5-700 cobot. The results demonstrate the system’s effectiveness in facilitating seamless human-robot collaboration, highlighting its potential applications in manufacturing, healthcare, and service industries. In conclusion, this research underscores the critical role of integrated perception and motion planning in advancing collaborative robotics, paving the way for safer and more efficient shared workspaces.