Soft robots are promising mechanisms for human-robot-interaction due to their inherent safety and natural movement behaviors. Bionic Handling Assistant (BHA) robot is one of the largest soft continuum robots which has pneumatically actuated modules in the form of air chambers. Each module is able to bend, stretch, and grasp in all directions. The control of this robot is very challenging due to its specific characteristics such as parallel actuation, complex movement dynamics, slow pneumatic actuation, non-stationary behavior, and a lack of analytic models. In this topic, we cover the basics of one available control architecture in which standard classical controller and state of the art machine learning approaches are integrated.
Literature: Rolf et al.