Adaptive Neuro-Fuzzy Control for Ionic Polymer Metal Composite Actuators

Abstract

Electroactive polymers (EAPs) have many attractive characteristics for applications, especially for biomimetics robots and bio-medical devices. Among the electroactive polymers, the ionic polymer metal composite (IPMC) is the commonly used EAPs. The IPMC is new generation of smart materials with significant potential in producing biomimetic robots and smart structures, and for medical applications. Ionic polymer metal composites (IPMC) have attracted great attention in the past years due to its large strain. IPMC materials have quite an unpredictable behavior due to several critical aspects that produce a change in their dynamic response. For modeling of controlling the IPMC, it is required to find suitable algorithm. In order to avoid difficult problems in control, a controller based Adaptive Neuro-Fuzzy Inference System (ANFIS), which combines the merits of fuzzy logic and neural network, is used for tracking position of IPMC actuator. This paper describes the using of controller based on Neuro and Fuzzy for controlling an IPMC actuator under water to improve tracking ability for an IPMC actuator like as biomimetics and biomedical devices. The results showed that ANFIS algorithm is reliable in controlling IPMC actuator. In addition, experimental results show that the ANFIS performed better than the pure fuzzy controller (PFC). Present results show that the current adaptive neuro-fuzzy controller can be successfully applied to the real-time control of the ionic polymer metal composite actuator for which the performance degrades under long-term actuation.

Intelligent Control of Electroactive Polymer Actuators Based on Fuzzy and Neurofuzzy Methodologies

“…Other control methodologies applied to IPMC (ionicpolymer metal composite) type actuators include adaptive neuro-fuzzy control [7] and the development of a non-linear black box model [8] which utilises the learning capability of neural networks. The neuro-fuzzy controller developed by Thinh et al [7] compares a pure fuzzy controller with an adaptive neuro-fuzzy controller and demonstrates that a neural-network is capable of modelling and improving the performance of a dynamic system.…”

“…The neuro-fuzzy controller developed by Thinh et al [7] compares a pure fuzzy controller with an adaptive neuro-fuzzy controller and demonstrates that a neural-network is capable of modelling and improving the performance of a dynamic system. Truong et al [8] attempts to model the IPMC actuator in order to eliminate the need for an external displacement sensor.…”

“…Step 3 -Apply Implication method Determine the implication of the antecedent parameter(s) on the consequent(s) using the T-norm MIN method (7).…”

Force control of a tri-layer conducting polymer actuator using optimized fuzzy logic control

“…Fuzzy logic (FL) controllers can be used to control polymer actuators as there are difficulties in deriving their precise mathematical models [19,20]. A disadvantage of the FL control is that designing an effective rule base requires experience, intuition and trial-and-error methods which make it difficult to optimize performances especially when the dimensions of the rule base become larger [20]. In order to overcome the problem of manual tuning of the membership functions and parameters of FL controllers, heuristic algorithms can be utilized [21,22].…”

“…Adaptive neuro-fuzzy inference system (ANFIS) method was also used as a controller for the tip displacement of both ionic polymer metal composite (IPMC) actuators [20] and trilayer bender type CPAs [19]. Recently, a data driven ANFIS model was used to design a predictive controller for a high-speed electric multiple unit due to its feature for capturing nonlinearities very effectively [23].…”