An approach to path planning and real-time redundancy control for human-robot collaboration

This paper presents a control system for redundant robot manipulators that handles the path planning and redundancy control in two separate loops. The goal is to simplify industrial programming of redundant robots, by decomosing it to a familiar path planning loop, and an autonomous redundant control loop. The familiar path planning is done using an industrial robots native control system, which is already known to the operator. An external feedback loop acts as a secondary task control loop so that the redundancy in the robot is used automatically. The secondary task control is expanded from the null space projection method to a method where a parallel space to task space is modeled in joint space. The system was implemented and experimentally verified on a NACHI MR20 7-axes industrial robot with AX20 Controller. The secondary task in the experiments was an obstacle avoidance task that was tested in unknown dynamic and static environments.