Embedded Sensing and Control for Multi-unit Soft Robotic Systems (Talk)

Due to the deformable and compliant nature of soft systems, there has been an ongoing need to address the challenges of state estimation and environmental interaction using embedded and distributed sensing. In this talk, I will be sharing the steps I took to develop reliable soft sensors that can be scaled up to control the movements and interactions of various multiunit, electrostatic systems. Initially, I focused on characterizing, sensing, and controlling a single HASEL actuator. However, during this process, we discovered issues with the current method of sensing the movement of these electrostatic actuators, which prevented us from creating larger systems. To overcome this challenge, I developed a magnetic sensing mechanism that detects the deformation of these actuators and allows us to scale up the number of units in the system. With this approach, we were able to successfully create and control several multi-unit systems. Furthermore, we also demonstrate the benefits of magnetic sensing by applying it to a more widely studied type of soft system, which is a pneumatic-based soft gripper. Overall, my talk will provide insights into the development of reliable soft sensors that can be scaled up to control complex multiunit systems.