Robin Passama

Abstract: Although the concept of industrial cobots dates back to 1999, most present day hybrid human-machine assembly systems are merely weight compensators. Here, we present results on the development of a collaborative human-robot manufacturing cell for homokinetic joint assembly. The robot alternates active and passive behaviours during assembly, to lighten the burden on the operator in the first case, and to comply to his/her needs in the latter. Our approach can successfully manage direct physical contact between robot and human, and between robot and environment. Furthermore, it can be applied to standard position (and not torque) controlled robots, common in the industry. The approach is validated in a series of assembly experiments. The human workload is reduced, diminishing the risk of strain injuries. Besides, a complete risk analysis indicates that the proposed setup is compatible with the safety standards, and could be certified.

TL;DR: The proposed framework for real-time safe human-robot collaboration, using static hand gestures and 3D skeleton extraction is designed and integrated in the physical human- robot interaction library OpenPHRI.

TL;DR: This letter proposes a framework that allows the robot to use environmental contacts for shaping the cable, and introduces an index to quantify the contact mobility of a cable with a circular contact.

TL;DR: A multimodal sensor-based controller enabling a robot to adapt to changes in the sensor signals, based on a unified task formalism, and in contrast with classical hybrid visicn-force-position control, enables smooth transitions and weighted combinations of the sensor tasks.

TL;DR: A multimodal sensor-based control framework, enabling a robot to recognize human intention, and consequently adapt its control strategy, based on a unified task formalism is proposed and validated in a mock-up industrial scenario.