Grant: Discovery Grants Program - Individual

Over the past decade, there has been a surge of interest in soft robots due to their diverse potential applications. Soft robots, characterized by their use of pliable structures and muscle-like actuation through deformable materials, can undergo significant deformations during operation, allowing for continuous changes in shape along their length, reminiscent of natural examples such as octopus arms, earthworms, and elephant trunks. Recent advancements in soft actuation technologies have greatly facilitated the progress of soft robotics. Within the realm of soft actuation materials, Liquid Crystal Elastomers (LCEs) have emerged as a novel and innovative category of materials that exhibit substantial deformations in response to light signals. When exposed to light, the actuator can achieve significant deformation. The photo-responsive LCE actuator can be disconnected from its power source, resulting in a more lightweight and versatile design. This property makes LCEs particularly promising for photo-responsive applications, offering advantages not typically found in conventional electrically driven systems. The combination of this unique feature with its lightweight nature, high energy density, rapid response, and cost-effectiveness renders LCEs highly attractive for the advancement of soft robots. However, robots constructed entirely or partially from shape-changing materials pose greater challenges in terms of control compared to their rigid counterparts. LCEs exhibit viscoelastic behavior, meaning that their stiffness varies with strain rate and frequency. Viscoelasticity introduces complexities in modeling and controlling the material response, often characterized by phenomena like creep, hysteresis, and stress relaxation. Due to these inherent nonlinear and time-varying characteristics, achieving precise control of LCEs to determine the required light input for soft robot deformation becomes a non-trivial problem. Notably, LCEs exhibit sandwiched hysteresis nonlinearity, which can significantly degrade actuator performance. Addressing this hysteresis is challenging because most existing control methods are applicable only to smooth and differentiable systems. When hysteresis nonlinearity is sandwiched between two nonlinear systems, the control problem remains unresolved. The lack of control tools severely restricted the application of this promising material in soft robots. The primary objective of this research is to develop control theories, engineering design methods, and technology for LCE-actuated soft robots. This research aims to enable these robots to adapt to their surroundings, navigate unstructured environments, and manipulate objects using their entire arm in the presence of model uncertainties.