Curvature Estimation of Soft Grippers Based on a Novel Highly Stretchable Strain Sensor With Worm-Surface-Like Microstructures

This paper presents a high-stretchable strain sensor based on the worm-surface-like microstructures and the principle of resistance sensing to estimate the curvature information of the soft pneumatic gripper with a large bending angle. The proposed sensor comprises a central viscoelastic fiber with worm-surface-like microstructures made of Ecoflex-10 silicone and an outside conductive layer of Carbon Black (CB) powder. The microstructures are generated by a 3D printing customized mold, and they can effectively increase the surface area and adhesion energy of the viscoelastic fiber so that more conductive powder can be adsorbed. Therefore, the proposed sensor produces lower resistance in the initial state and generates a more stable circuit connection and electrical signal transmission in the extremely stretched state, significantly improving the sensor sensitivity. Meanwhile, the microstructures can reduce the surface strain of the viscoelastic fiber during stretching and further increase the sensor’s measurement range. The flexibility improvement of the proposed sensor brought by the microstructures can additionally reduce the required tension for achieving equivalent deformation, reducing the impact on the motion of the measured soft grippers. The electrical performances of the highstretchable strain sensor have been characterized, and it can reach an excellent strain range of up to 500% with a gauge factor of 206.6. The prototyped strain sensor has been integrated on a soft pneumatic gripper’s outside surface and well reflected the unknown objects’ curvature radius with an average error of 4.07% during the grasping experiments.