Experiments of Composite Learning Admittance Control on 7-DoF Collaborative Robots.

With the increasing demand for collaborative robots in various industries, the implementation of interaction control algorithms with superior performances for collaborative robots becomes significant. This paper develops a composite learning admittance control (CLAC) approach and implements it to an industrial collaborative robot with 7 degrees of freedom named Franka Emika Panda. The interaction performance of the CLAC in the task space is verified by both force tracking and compliance control experiments. Experimental results show that the CLAC enables the Panda robot with favorable force tracking capability under a specific admittance model, high tracking accuracy in the free state, and satisfied compliance under motion constraints.