Optimal Control of Actuator Arrays with Electric and Thermofluidic Inputs.
The objective of this research is to characterize the performance of an array of Shape Memory Alloy (SMA) actuators with multiple energy domain inputs. The SMA actuators will be embedded in a vascular network that can provide both electric and thermofluidic energy. The vascular system delivers and removes thermal energy from compact muscle]like actuators, allowing them to be integrated into large DOF systems. This research is anticipated to advance the capabilities of a variety of robotic systems such as prosthetics, exoskeletons, haptic devices and biomimicing robots. Specifically, we propose two distinct objectives: 1) to characterize and optimize the performance of a single wet SMA actuator using electric and thermofluidic inputs, 2) to characterize and optimize the performance of these wet SMA actuators in bundles/arrays where different combinations of actuators in the array can be addressed with different inputs in order to maximize a combination of speed and energy consumption. We plan to accomplish this through several research tasks including thermomechanical and computational fluid dynamic modeling of wet SMA actuators, and analysis of intelligent algorithms for optimal multi-input control of actuator arrays. This research sponsored by an NSF Grant.
Design of a Self-Sustaining SMA Thermally Powered Robotic Heart with SMA Muscles.
The goal of this project is to develop a robotic heart that can deliver hot and cold fluid at relatively low pressures to power the SMA actuator arrays described above. A fraction of the fluidic output will be circulated back to the heart muscles, which themselves are wet SMA actuators. The heart will be self-sustaining as long as heat is added and removed from hot and cold reservoirs, respectively. Research to date has involved dynamic modeling of the heart system, optimization of design parameters, simulation, and experimental validation. We have just recently achieved the objective of experimentally implementing a system that is capable of a sustained net output of 66 mL/min of fluid.
Integrated Control of Growth/Actuation for the Recursive Optimization of EAPs
The general research objective of this proposal is to experimentally characterize a method for recursively optimizing the performance of a class of electroactive polymer (EAP) actuators through an integrated control of growth and actuation. EAPs are part of a larger class of smart materials that have the potential to function as artificial muscles in bioinspired robotic and mechatronic systems. However, unlike other smart materials such as Shape Memory Alloys, EAPs have the potential to grow and self-repair in situ, much like biological muscles. We specifically propose to work with Conducting Polymer (CP) actuators that are electrochemically grown on top of a metal helix support structure, a technique that is already in use by other researchers. We first propose to advance this technique by designing a novel cell, such that these helical CPs can be iteratively grown and actuated in the same cell, by alternately filling the cell with different electrochemical solutions. We then propose to develop an intelligent model-based control algorithm characterized by alternating cycles of growth and actuation, such that the experimentally measured performance of the actuators (i.e. speed, strength) drives the growth process until an optimal actuator size is achieved. In the same way that human muscles autonomously self-repair and grow in response to increased workload, we anticipate robotic muscle systems that could be similarly capable of self-repair and growth to meet changing demands placed on them.