Closed-loop control of soft robot based on machine learning

In this paper, we present a new strategy to control the soft robot with elastic behavior, piloted by 4 actuators. The main contribution of this work is the use of neural network to get approximated model soft robots, based on which a robust controller is then proposed. In this paper, we proved that if the approximated model satisfies certain conditions, then the proposed robust controller can always drive any given point of interest of the robot to the desired position, without knowing the exact model. Finally, the proposed result is experimented and validated by a 3D printed silicone soft robot.