2.11.1.1 Control Applications
To date, type-2 fuzzy logic has been widely used in control applications, and most appli- cations are using type-2 interval fuzzy sets with the Karnik-Mendel iterative algorithms and the Wu-Mendel minimax uncertainty bounds, allowing fast execution of type-2 fuzzy systems.
Many researchers have begun to use type-2 fuzzy logic in control applications. For example, Melin and Castillo [80, 81] and Castillo et al. [82] have used type-2 interval systems in the context of plant control. Hagras [83] presented type-2 fuzzy logic control application to three challenging domains including industrial, mobile robots, and ambient
2.11. Recent Work on Type-2 Fuzzy Sets and Systems 52 intelligent environments control. Lynch et al [84, 85] are continuing to build a type-2 interval control system for large marine diesel engines. Hagras et al. [86, 87] and Doctor et al [88] used a type-2 interval system to model and adapt to the behaviour of people in an intelligent dormitory room. Wu and Tan [89] applied type-2 interval systems to the control of a complex multi-variable liquid level process and in [90] they simplified type-2 fuzzy logic control to real-time control applications. Melgarejo et al. [91] have developed a limited hardware implementation of a type-2 interval controller. Lin et al. [92, 93] designed type-2 fuzzy controller for buck DC-DC converters.
2.11.1.2 Time Series Forecasting Application
There are more researchers interested in using type-2 fuzzy sets to deal with forecasting applications. Uncu et al. [94] proposed a system modelling approach based on type-2 fuzzy sets to predict the price of a stock. Baguley et al. [95] found that a model with type-2 fuzzy sets can leverage design process knowledge and predict time to market from performance measures is a potentially valuable tool for decision making and continuous improvement. Kim and Park [96] used type-2 fuzzy logic system to forecast the Box- Jenkin’s gas furnace time series and compare the results with type-1 fuzzy logic system. Huarng and Yu [97] proposed the use of a type-2 fuzzy time series model to improve the prediction performance by using the TAIEX, Taiwan stock index, as the forecasting target.
Medina and Mendez [98] presented an application of the interval singleton type-2 fuzzy logic system to one-step-ahead prediction of the daily exchange rate between the Mexican Peso and US dollar (MXNUSD). Mencattini et al. [99, 100] used type-2 fuzzy systems for meteorological forecasting. Li et al. [101] proposed a new method for short- term traffic forecasting using type-2 fuzzy logic.
Karnik and Mendel [102] used a type-2 interval system to predict the next value in a chaotic time series. Musikasuwan et al. [103] investigated the effect of number of model parameters on performance in type-1 and interval type-2 systems. Both systems were designed to predict a Mackey-Glass time series.
Liang and Wang [104] presented a new approach for sensed signal strength forecast- ing in wireless sensors using interval type-2 fuzzy system and compare with type-1 fuzzy
2.11. Recent Work on Type-2 Fuzzy Sets and Systems 53 system. Pareek and Kar [105] demonstrated an application of type-2 fuzzy system to predict a critical parameter of Gas Turbine in a power plant, that is the compressor dis- charge pressure. Mendez et al. [106] presented the experimental results of the application of type-2 fuzzy systems for scale breaker entry temperature prediction in a real hot strip mill.
2.11.1.3 Medical Applications
There are researchers using type-2 fuzzy logic to model in medical application. John et al. [107–109] used type-2 fuzzy sets to assist in the pre-processing of tibia radiographic images, while John and Lake investigated the use of type-2 fuzzy sets in modelling nurs- ing intuition. Innocent et al. [110–112] represented the perceptions of lung scan images by experts in order to predict pulmonary emboli by using type-2 fuzzy relations. Garibaldi et al. [10–14] have done extensive work on assessing the health of a new born baby us- ing knowledge of acid-base balance in the blood from the umbilical cord. Di Lascio et al. [113] presented a model of differential medical diagnosis for the pathologies based on type-2 fuzzy sets to indicate the elements needs to have more precise diagnosis and it can control its same accuracy. Finally, Herman et al. [114] examined the potential of the type-2 fuzzy system methodology in devising an EEG-based brain-computer interface to classify imaginary left and right hand movements.
2.11.1.4 Mobile Robot Applications
Type-2 fuzzy systems were successfully applied in mobile robot controllers. Phokharatkul and Phaiboon [115] implemented the type-2 fuzzy logic controller to process the data out- put to control the direction of the mobile robot movement. Hagras [116,117] implemented the type-2 fuzzy logic controller in different types of mobile robots navigating in indoor and outdoor unstructured and challenging environments. Coupland et al. [118] designed and compared three fuzzy logic control using type-1 , interval type-2 and general type-2 fuzzy logic to the robot control for completing the task of following the edge of a curved wall, and found that both type-2 fuzzy systems outperformed the type-1 system. Figueroa et al. [119] explored how the type-2 fuzzy logic controller, in the context of robot soccer games, overcomes uncertainty in the control loop without increasing the computational
2.12. Summary 54