Embodied Control in Soft Robots via Stimuli-Responsive Modules

Robots typically interact with their external environment via feedback loops involving multiple sensors, microcontrollers, and actuators constructed from rigid electronic components that can be bulky, complex, and mechanically incompatible with soft materials. In this work, we explore strategies for embodying autonomous control of the trajectory and function of pneumatic soft robots. We specifically aim to embody the control system and its feedback loop via stimuli-responsive modules that can be moved, added, or removed from the robot to produce different behaviors in response to the surrounding environment. As one example, we build a kirigami-inspired robot that autonomously steers itself in response to variations in light or heat in the local environment by using liquid crystal elastomers (LCEs) in the control modules. The LCEs activate or deactivate modules to produce variable mechanical constraints in the kirigami, resulting in bending and steering of the robot. Additionally, we use analogous stimuli-responsive control modules to regulate the pneumatic logic of the soft robot, further expanding the range of autonomous functional changes that the robot can undergo. We show how decisions based on multiple stimuli can be achieved via logical combinations of these inputs.