Modelling and control of IPMC actuators for biomedical robotics applications : a compliant stepper motor, an artificial muscle joint, a microfluidic pump and a cell microtool/gripper and manipulator

• Main Author: McDaid, Andrew John
• Other Authors: Aw, Kean
• Published: ResearchSpace@Auckland 2011
• Online Access: http://hdl.handle.net/2292/8440

The natural progression for future robotic devices is to work closely with humans in their own environment and even become embedded with humans for improving quality of life. Some already existing biomedical devices include hearing aids, pacemakers and prosthetic limbs. The main motivation for this research was to develop new innovative biomedical robotic devices which have the ability to improve the standard of living for humanity. Traditional robotic actuators, for example electro-magnetic drives and hydraulic/pneumatic machines, have all been extensively investigated. Although these devices and their control systems are well understood and have advanced performance which can, in some aspects, surpass that of humans, they are simply inadequate for developing beneficial devices which are capable of operating together with humans to augment their capabilities. The main limiting factors for these devices are size, weight, power requirements, stiffness and scalability, most of which cannot be resolved through incremental research. New approaches to device development must therefore be taken. Bio-inspired transducers which have similar properties to human tissue and muscle, in particular mechanical compliance, structural simplicity to allow easy scalability, high power-to-weight and power-to-volume ratios, precise and embedded control capabilities are demonstrating much promise as candidates for replacing traditional robotic actuators. IPMCs, a type of EAP whose actuation mechanisms can mimic biological muscle, were utilized in this research due to their extensive desirable characteristics when compared with traditional and other smart material actuators. The aim of this research was to further the state of art in smart material transducers towards their implementation as replacements for traditional robotic actuators in real life applications, outside the laboratory. Fundamental scientific research into advanced modelling and control of IPMCs has facilitated their implementation into a new generation of biomedical robotic devices developed in this research. The first step was the development of a new conclusive scalable model to predict the actuation response of IPMC actuators. This novel model was developed based on real material parameters and physical transduction mechanisms and can predict the mechanical response with external loads and hence is very useful for designing mechanical devices. Five novel biomedical robotic devices have been designed, utilizing the developed IPMC model to optimise their performance and validate the designs before implementing the real devices. These devises are a bio-inspired compliant stepper motor, an artificial muscle finger joint, a microfluidic pump, a cell microtool/gripper and a cell micromanipulation system. Each application has specific performance specifications and so new control strategies were developed to meet these requirements. The new controllers are based on an IFT algorithm which adaptively tunes the controller to optimise device performance. The controllers extend the current capabilities of the IFT algorithm in order for it to be implemented on nonlinear systems, online throughout normal operation and to be robust by rejecting external disturbances and adapting to drift in system dynamics over time. Finally the proposed devices and their corresponding controllers have been built and implemented to verify their performance in real life situations. This has validated the entire design process from the IPMC model and mechanical system simulations through to controller performance and device functionality. By successfully developing and implementing the devices in this research the feasibility of integrating IPMCs into real world applications as solutions to engineering problems has been proven.