Bimanual human robot cooperation with adaptive stiffness control

Abstract

We propose a human-robot cooperation scheme for bimanual robots. After the initial task demonstration, the human co-worker can modify both the spatial course of motion as well as the speed of execution in an intuitive way. To achieve this goal, speed-scaled dynamic motion primitives are applied for the underlying task representation. The proposed adaptation scheme adjusts the robot's stiffness in path operational space, i. e. along the trajectory. It allows a human co-worker to be less precise in the parts of the task that require high precision, as the precision aspect can be provided by the robot. The required dynamic capabilities of the robot were obtained by decoupling the bimanual robot dynamics in operational space, which is attached to the desired trajectory. The proposed scheme was validated in a task where two Kuka LWR-4 robot arms cooperate with a human to carry an object.