Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Soft robotic systems are composed of active and passively deformable structures which are intrinsically compliant, flexible, and elastic. Although these features offer benefits of adaptability, robustness, and safety, controlling these types of robots is a significant challenge, in part from the difficulty of obtaining feedback from sensors which provide state information without hindering the advantageous material properties which grant these systems their unique mechanical behavior. We demonstrate here the first integration of a flexible, stretchable, liquid metal-based strain sensor with vacuum powered soft pneumatic actuators (V-SPAs) for simultaneous controlled feedback of the soft actuators as well as user input and soft robotic device interaction. The soft sensors which are encapsulated within a Polydimethylsiloxane (PDMS) membrane are directly embedded in the outer body skin of the soft actuators, and can be used to correlate the deformation of the body under vacuum actuation to overall actuator strain or to detect external disturbances. This information is used to compute and control the angle of a rotational 3-DoF actuator module, as well as detect implicit user input control signals by direct interaction without the need for an external control interface. The dual use of embedded sensing shown in this work provides a fundamental strategy for soft collaborative robot applications.
Josephine Anna Eleanor Hughes, Francesco Stella
Mahmut Selman Sakar, Mehdi Ali Gadiri, Junsun Hwang, Amit Yedidia Dolev
Dario Floreano, Bokeon Kwak, Markéta Pankhurst, Jun Shintake, Ryo Kanno