An Optical Curvature Sensor for Tip Position Trajectory Tracking Control of Soft Robots

As soft robotics applications become more complex, control requirements increase. However, advanced control approaches require sensors which do not infer with the softness of the robot. In order to measure the current configuration of the robot, curvature measurements are of great importance, since bending is usually the main deformation of soft robots. This paper presents the design, fabrication, and trajectory tracking control of a novel low-cost fiber-optic soft curvature sensor that can measure the magnitude and direction of the curvature in two spatial directions. The curvature sensor can be seamlessly integrated into the silicone body of soft robots, preserving their inherent softness. The trajectory control of a beam-shaped soft robot is considered here as an example. For trajectory tracking control, an inverse kinematic feedforward controller combined with a simple feedback controller is investigated. Experiments show that the combined controller provides accurate trajectory tracking control with good disturbance rejection.