• cooperative manipulation
• inference engine
• load sharing
• orienting response
• pHRI
• user intent

Since the very beginning of industrial robotics, safety has been a fundamental point
and robots have been kept in closed work-cells, avoiding interaction with humans.
However, many of the future and emerging applications of robotics will indeed require
robots to work in direct contact with humans.
Robotic systems designed to share an environment with humans must fulfill different requirements from those typically met in industry.
To effectively interact with people in a physically assistive role, robots will need to be able to cooperatively manipulate objects with a
human partner. For example, it can be very difficult for an individual to manipulate a long object and a robotic assistant can help to improve the manoeuvrability of the load.
Herein, non-verbal communication in a human-robot coordinated manipulation task is addressed using a robot arm equipped with a wrist mounted force/torque sensor.
To aid collaborative manipulation we implement an impedance nonholonomic constraint at the robot end-effector with an attractive force to a path. We utilize force/torque sensor measurements in order to identify intentional user communications specifying a change in the task direction and the path is recomputed on-line as a function of user input. Moreover, we consider the effects of path recomputation and the resulting robot haptic feedback on user physiological response.