Anticipatory Robot Assistance for the Prevention of Human Static Joint Overloading in Human–Robot Collaboration

TLDR

The letter proposes a novel human–robot collaboration control approach that alerts and reduces static joint torque overloading in a human partner during shared tasks. The method uses a statically equivalent serial chain model to compute center‑of‑pressure and ground‑reaction‑force variations, evaluates joint‑torque overloads in real time, and applies online optimization to adjust robot trajectories for ergonomic human poses, with the optimized configurations displayed live and validated in a human–robot load‑sharing task.

Abstract

This letter proposes a novel human–robot collaboration (HRC) control approach to alert and reduce the static joint torque overloading of a human partner while executing shared tasks with a robot. Using a preidentified statically equivalent serial chain model, variations of the centre-of-pressure and ground reaction force are calculated, and the overloading joint torques are evaluated in real time to initially alert the human about consequent injuries. An online optimization technique is implemented to adjust the robot trajectories facilitating the human to achieve more ergonomic body poses throughout the HRC task. The optimized human configurations are calculated by taking into account the human stability, the robot, and the human workspaces, and the task constraints, and illustrated to the human in real time. The experimental evaluation of the proposed technique is achieved in a human–robot load sharing task as a representative example of coassembly or transportation scenarios in industrial settings.