Robots that Evolve on Demand

Robots are traditionally designed with fixed physical hardware and control policies that make them specialized for repetitive tasks and structured environments. This talk discusses foundational work toward robots that "evolve on demand," morphing their bodies and adapting their behaviors to accommodate multiple tasks in diverse environments. First, I will introduce robotic structures made from stiffness-changing materials that enable myriad shape transformations. Next, I will discuss a compact sensor capable of detecting a wide variety of stimuli relevant to shape morphing robots. I will then present models that describe shape morphing systems' highly nonlinear geometric and material behavior, allowing for more efficient inverse design of bespoke morphing components. These foundational topics inform the second part of the talk-the interplay of a robot's shape and control policy. Here, I will demonstrate how tuning shape can boost robotic performance and unlock entirely new capabilities in locomotion and combined locomotion-manipulation tasks.