No abstract

No abstract

Various robotic applications require enforcing constraints, to achieve task performance or to hinder the robot from causing danger. Especially in human-robot-interaction, collision avoidance and velocity limits are crucial for safety. A promising approach to enforce adherence to safety margins is invariance control. Considering the system dynamics, it corrects a nominal control based on a switching policy. To overcome the resulting lack of smoothness in terms of control inputs and states, a control scheme is developed that adds dynamics to the input and thereby augments the system. In previous work it is shown that the closed loop system is stable in combination with an exponentially stabilizing nominal control. However, this could not be shown for impedance controlled robots. Therefore, in this paper, an augmented invariance control for robotic applications is developed and proven to be uniformly asymptotically stable with an impedance-type nominal control. The proposed control law is validated in experiments on a KUKA LWR4+ in a collision avoidance scenario demonstrating stable behaviour with improved smoothness and chattering characteristics while enforcing the imposed hard constraints.

In many applications, the controlled dynamical system is subject to constraints, e.g. to enforce a desired performance. There are various methods, which guarantee the enforcement of such constraints. They do, however, not provide satisfactory solutions if the system state has to exhibit certain smoothness properties. This work presents a novel method of designing a control input, which guarantees the adherence to constraints as well as a desired smoothness of the system state. The control scheme is implemented as an add-on to an existing task dependent control law. Augmentation of the dynamical system achieves a desired smoothness of the state whereas methods from invariance control are utilized to guarantee constraint enforcement.

No abstract