Enabling Proprioception in Multistable Soft Machines through Embedded Soft Capacitive Sensors

Transforming soft machines into fully autonomous soft robots capable of complex interactions with the environment requires the integration of soft electronics and sensors, enabling feedback control while preserving mechanical compliance. Multistable mechanisms, such as mechanical metamaterials and snapping beams, are great candidates for such systems due to their programmable nonlinear responses. In this work, we introduce the integration of a sensing technology based on 3D-printed soft capacitive sensor in 3D printed multistable metastructure composed of four bistable unit cells - enabling proprioceptive monitoring of state transitions by detecting snap-through events in multistable mechanisms. Under cyclic tensile loading, the metastructure exhibits snapping events at forces up to 7N, accompanied by capacitance changes from about 0.025 pF to 0.750 pF. We further demonstrate the integration of the same sensing principle in a 3D printed monolithic bistable soft gripper, reliably grasping delicate objects such as popcorns, tomatoes and strawberries, spanning masses from 0.5 g to 16 g. These results confirm that embedded capacitive soft sensors are a viable technology to detect snap-through transitions and to be seamlessly integrated in soft structures, representing a crucial step toward fully compliant, self-aware robotic systems.