Search

This article proposes a navigation scheme for a wheeled robot in unknown environments. The navigation scheme consists of obstacle boundary following (OBF), target seeking (TS), and vertex point seeking (VPS) behaviors and a behavior supervisor. The OBF behavior is achieved by a fuzzy controller (FC). This article formulates the FC design problem as a new constrained multiobjective optimization problem and finds a set of nondominated FC solutions through the combination of expert knowledge and data-driven multiobjective ant colony optimization. The TS behavior is achieved by new fuzzy proportional-integral-derivative (PID) and proportional-derivative (PD) controllers that control the orientation and speed of the robot, respectively. The VPS behavior is proposed to shorten the navigation route by controlling the robot to move toward a new subgoal determined from the vertex point of an obstacle. A new behavior supervisor that manages the switching among the OBF, TS, and VPS behaviors in unknown environments is proposed. In the navigation of a real robot, a new robot localization method through the fusion of encoders and an infrared localization sensor using a particle filter is proposed. Finally, this article presents simulations and experiments to verify the feasibility and advantages of the navigation scheme.

This article proposes a data-driven multiobjective evolutionary interval type-2 fuzzy-logic system (IT2FLS) learning approach that considers both system performance and rule interpretability. One characteristic of the evolutionary learning framework is that the IT2FLS structure is incrementally generated instead of searching from an initially huge grid-type rule base, which significantly reduces the parameter search space. Another one is that a new constrained objective function is proposed to improve the distinguishability of the generated interval type-2 fuzzy sets (FSs) without restricting them to be evenly distributed. Because of the tradeoff between system performance and interpretability, a new multiobjective ant colony optimization (ACO) algorithm is proposed to optimize the IT2FLS parameters and improve optimization performance. The evolutionary IT2FLS learning approach is applied to control a real wall-following hexapod robot. The approach shows the advantages of model-free design and interpretability and robustness to noise in the evolved type-2 fuzzy rules. Simulations with performance comparisons and experiments in controlling a real robot show the advantages of the learning approach.

Manual titration of positive airway pressure (PAP) is a gold standard to provide an optimal pressure for the treatment of obstructive sleep apnea-hypopnea syndrome (OSAS). Since manual titration studies were costly and time-consuming, many statistical models for predicting effective PAPs were reported. However, the prediction accuracies of the models associated with nocturnal parameters still remain low. This study proposes a fuzzy neural prediction network (FNPN) with input candidate variables, selected among easily available measurements (e.g., body mass index (BMI), waist circumstance (WC), and body composition) and OSAS related questionnaires, to rapidly predict an optimal PAP. The FNPN comprises fuzzy rules and is characterized with the ability of automatic rule growing and pruning from training data. A total of 147 participants from April 2018 to April 2019 were enrolled in Taichung Veterans General Hospital, Taiwan. After two selection processes for feature extraction, WC and BMI were the significant variables for entering the FNPN to predict optimal PAP. Experimental results showed that the average successful prediction rate of the proposed method was 71.8%. This study also found that Epworth sleepiness scales (ESS) and body composition, such as visceral fat area and percent body fat, were excluded in the final prediction model. Compared with existing models, the proposed prediction approach provided a rapid prediction of optimal PAP with higher accuracy.

This paper proposes a new method for the tracking of three-dimensional (3D) hand palms from the whole human standing body using fuzzy rule-based fusion of explainable fuzzy features from a monocular video. The characteristics of this method include visually and linguistically explainable fuzzy features and rules and computational efficiency. This paper first tracks the two-dimensional (2D) palms using the following four fuzzy features: optical flows, the degree of a pixel in the foreground, skin color information, and the search area around a hand palm candidate from a segmented body. Afterward, a fuzzy system is proposed to fuse the four fuzzy features to estimate the 2D palm positions. Localization of the elbows is based on the estimated palm locations, human body skeletons, and body contour. The 2D palms and elbows are tracked using a modified particle filter. To estimate the depth of each palm, the locations of the palm and elbow are fed as inputs to a neural fuzzy system. The 3D palm tracking result is applied to a robot upper-body imitation system. Experiments with comparisons of different hand palm tracking methods are performed to verify the real-time computational ability and accuracy of the proposed method.

Along with the rapid development of synthetic biology, the technology of biological computers has become increasingly mature in recent years. In this paper, we propose a structure for a biological ...

This paper proposes a new reinforcement neural fuzzy surrogate (RNFS)-assisted multiobjective evolutionary optimization (RNFS-MEO) algorithm to boost the learning efficiency of data-driven fuzzy controllers (FCs). The RNFS-MEO is applied to evolve a population of FCs in a multiobjective robot wall-following control problem in order to reduce the number of time-consuming control trials and the implementation time of learning. In the RNFS-MEO, the RNFS is incorporated into a typical multiobjective continuous ant colony optimization algorithm to improve its learning efficiency. The RNFS estimates the accumulated multiobjective function values of the FCs in a colony without applying them to control a process, which helps reduce the number of control trials. The RNFS is trained online through structure and parameter learning based on the reinforcement signals from controlling a process. Considering the influence of the current control signals on the future states of a controlled process, the temporal difference technique is used in the RNFS training so that it estimates not only the current but also the future objective function values. The colony of FCs in the RNFS-MEO is repeatedly evolved based on the RNFS estimated values or the objective function values from real evaluations until a colony of successful FCs is found. The RNFS-MEO-based FC learning approach is applied to a robot wall-following control problem. Simulations and experiments on the robot control application are performed to verify the effectiveness and efficiency of the RNFS-MEO.

This paper proposes the optimization of a fully connected recurrent neural network (FCRNN) using advanced multiobjective continuous ant colony optimization (AMO-CACO) for the multiobjective gait generation of a biped robot (the NAO). The FCRNN functions as a central pattern generator and is optimized to generate angles of the hip roll and pitch, the knee pitch, and the ankle pitch and roll. The performance of the FCRNN-generated gait is evaluated according to the walking speed, trajectory straightness, oscillations of the body in the pitch and yaw directions, and walking posture, subject to the basic constraints that the robot cannot fall down and must walk forward. This paper formulates this gait generation task as a constrained multiobjective optimization problem and solves this problem through an AMO-CACO-based evolutionary learning approach. The AMO-CACO finds Pareto optimal solutions through ant-path selection and sampling operations by introducing an accumulated rank for the solutions in each single-objective function into solution sorting to improve learning performance. Simulations are conducted to verify the AMO-CACO-based FCRNN gait generation performance through comparisons with different multiobjective optimization algorithms. Selected software-designed Pareto optimal FCRNNs are then applied to control the gait of a real NAO robot.

The blackbody cavity sensor formed by two coaxial tubes has been widely used in continuous temperature measurement of molten steel. However, due to the closed bottom of the inner tube, the temperature accuracy, response time and temperature measurement stability are seriously affected. It's necessary to redesign and improve sensing mechanism of the traditional design, which involves multidisciplinary knowledge, including materials, heat and flow science. This paper proposes a virtual verification-based design improvement method for blackbody cavity sensor. After redesigning the structure of the sensor, a virtual model for the sensor is established. Through real-world experiment, it is found that for the temperature measurement accuracy, the deviation between the simulation and the real-world experimental result is less than 1.5°C, and for the stability time of temperature measurement, the simulation result has a deviation from the real-world experimental result less than 15%. This verifies the accuracy of the virtual model. On this basis, model simulation for further possible optimal structures and parameters is carried out, and the influence of different nitrogen flow rates and inner tube lengths on the temperature measurement accuracy and the stability time for temperature measurement is further analyzed.

Objective/background Recently, several tools for screening obstructive sleep apnea-hypopnea syndrome (OSAHS) have been devised with varied shortcomings. To overcome these drawbacks, we aimed to propose a self-estimation method using an explainable prediction model with easy-to-obtain variables and evaluate its performance for predicting OSAHS. Patients/methods This retrospective, cross-sectional study selected significant easy-to-obtain variables from patients, suspected of having OSAHS by regression analysis, and fed these variables into the proposed explainable fuzzy neural network (EFNN), a back propagation neural network (BPNN) and a stepwise regression model to compare the screening performance for OSAHS. Results Of the 300 participants, three easily available features, such as waist circumference, mean blood pressure (BP) at the end of polysomnography and the difference in systolic BP between the end and start of polysomnography, were obtained from regression analysis with a five-fold cross-validation scheme. Feeding these three variables into the prediction models showed that the average prediction differences for apnea-hypopnea index (AHI) when using the EFNN, BPNN, and regression model were respectively 1.5 ± 18.2, 3.5 ± 19.1 and 0.1 ± 19.3, indicating none of the tested methods had good efficacy to predict the AHI values. The performance as determined by the sensitivity + specificity-1 value for screening moderate-to-severe OSAHS of the EFNN, BPNN and regression model were respectively 0.440, 0.414 and 0.380. Conclusions When fed with easy-to-obtain physiological features, the understandable EFNN should be the preferred method to predict moderate-to-severe OSAHS.

Due to the advantages of non-contact and low-cost, visual heart rate (HR) estimation is attracting more and more scholars to research. Recently, Some studies have shown that deep learning method could be developed for visual HR estimation. In this paper, we proposed an End-to-end deep neural network method for this task. The network is consisted of 2D convolutional (Conv2D) and LSTM (long short-term memory) operations. The Conv2D operation extract spatial feature and LSTM capture temporal information. The input of our model is facial ROI video and output is the predict HR. Experiment demonstrate that the proposed method could estimate HR value precisely.

This article proposes a new multistage evolutionary fuzzy control configuration and navigation of three-wheeled robots cooperatively carrying an overhead object in unknown environments. Based on the divide-and-conquer technique, this article proposes a stage-by-stage evolutionary obstacle boundary following (OBF) fuzzy control of each of the three robots through multiobjective continuous ant colony optimization. In the first stage, a set of evolutionary nondominated fuzzy controllers (FCs) for a single robot (a leader robot) in the execution of the OBF behavior is learned. In the second stage, a follower robot is controlled by two evolutionary FCs in combination with a switched compensation FC so that the leader and follower robots can cooperatively transport an object while executing the OBF behavior along obstacles containing corners with right angles. In the third stage, the third robot functions as an accompanying robot and is learned to enter into a predicted triangular formation with the leader-follower robots to transport a larger object while executing the OBF behavior. In the navigation of the three object-transportation robots, a new cooperative behavior supervisor is proposed to coordinate the learned OBF behavior and a target seeking behavior. Successful navigations in simulations and experiments verify the effectiveness of the multistage evolutionary fuzzy control approach and navigation scheme.

Signal frequency estimation is a fundamental issue in the domain of signal processing. In this paper, we proposed a novel framework, named FreqEnet (Frequency estimation network), for estimating frequency based on deep learning method. The signal frequency estimation refers to as a regression issue and predict it with LTSM module. The framework is exceedingly concise, consisted of only three LSTM and one fully connect layers, and the running time is less than 0.3 ms on CPU (i7-7700, 3.60 GHz). Two periodic signals are generated for training our model. In addition, uniform and Gauss white noise are introduce to original signal for evaluating the robustness and generalization of the framework. In addition, the proposed method performs extremely excellence in processing latent. Even if given only one periodic piece of signal, the method could predicts a precise result. Extensive experiments demonstrate that FreqEnet achieves competitive performance of estimating frequency.

This paper proposes an evolutionary wall-following hexapod robot, where a new multiobjective evolutionary fuzzy control approach is proposed to control both walking orientation and speed of a hexapod robot for a wall-following task. According to the measurements of four distance sensors, a fuzzy controller (FC) controls the walking speed of the robot by changing the common swing angles of its six legs. At the same time, the FC controls the orientation of the robot by applying additional changes to the swing angles of the three legs in each side. In addition to the basic requirement of walking along the wall in an unknown environment, the control objectives are that the robot should maintain a proper robot-to-wall distance and walk at a high speed. This paper formulates the control problem as a constrained multiobjective FC optimization problem. A data-driven advanced multi-objective front-guided continuous ant colony optimization (AMO-FCACO) is proposed to address the problem and find a Pareto set of optimal solutions of different FCs. The performance of the AMO-FCACO-based fuzzy wall-following control approach is verified through simulations and comparisons with various multiobjective optimization algorithms. Experiments on controlling a real robot in an unknown environment using two software-designed FCs are performed to view the control performance in practice.

The digital revolution of noncontact physiological signal monitoring in clinical and home health care is underway, and deep learning techniques are incredibly popular. Camera-based physiological signal monitoring for adults has made considerable progress in recent years. However, most of existing methods and datasets are developed for adult subjects, and until now, there has been no neonatal public database that is collected for developing deep learning method. Thus, in this paper, we introduce a large-scale newborn baby database, named NBHR (newborn baby heart rate estimation database), to fill the abovementioned knowledge gap. A total of 9.6 h of clinical videos (1130 videos totaling 921 GB) and reference vital signs are recorded from 257 infants at 0–6 days old. The facial videos and corresponding synchronized physiological signals, including photoplethysmograph information, heart rate, and oxygen saturation level, are recorded in our database. This large-scale database could be used to develop deep learning methods to estimate heart rate or oxygen saturation levels. Furthermore, a multitask deep learning method, called NBHRnet, is proposed to estimate heart rate based on the NBHR database, and the model is succinct that it can be deployed on a computer without GPUs. The experimental results indicate that NBHRnet yields competitive performance in predicting infant heart rate, with a mean absolute error of 3.97 bpm and a mean absolute percentage error of 3.28%; additionally, it can estimate heart rate almost instantaneously (2 s/60 frames). Our datasets are freely publicly available by request.

This paper proposes a neural fuzzy evaluation system (NFES) with significant variables selected from stepwise regression to predict apnea-hypopnea index (AHI) for evaluating obstructive sleep apnea (OSA). The variables considered are the change statuses of blood pressure (BP) before going to sleep and early in the morning as well as other five easily available measurements (age, body mass index (BMI), etc.) so that users can use the system for self-evaluation of OSA. A total of 150 subjects are reviewed retrospectively and categorized as training (120 subjects) and validation (30 subjects) sets by a fivefold cross-validation scheme with stratified sampling based on the OSA severity. Among the eight variables, the stepwise regression shows that BMI, the difference of systolic BP, and Epworth Sleepiness Scale were the significant factors to predict AHI. The three variables are fed as inputs to the NFES with interpretable fuzzy rules automatically generated from the training set. The average accuracy, sensitivity (Sn), specificity (Sp), and Sn+Sp-1 of the NFES were 75.6%, 77.2%, 75.0%, and 0.552, respectively, in distinguishing the OSA level of normal-mild (AHI

This paper proposes a new circuit for implementing a reduced-interval type-2 neural fuzzy system using weighted bound-set boundaries (RIT2NFS-WB) with online tuning ability. The antecedent and consequent parts of the RIT2NFS-WB use interval type-2 fuzzy sets and Takagi-Sugeno-Kang (TSK) rules with interval combination parameters, respectively. In the software implementation, the structure and parameters of the RIT2NFS-WB are learned through firing-strength-based rule generation and gradient descent algorithms, respectively. The software-designed RIT2NFS-WB is then transferred to a circuit implementation with online parameter-tuning ability; the hardware version is called the RIT2NFS-WB(HL). The RIT2NFS-WB(HL) is characterized by its online tuning ability with updatable consequent and weighting parameters. To the best of our knowledge, the RIT2NFS-WB(HL) is the first TSK-type interval type-2 neural fuzzy circuit with online parameter tuning ability in the literature. To take advantage of the inherent parallel processing property of the rules, a parallel processing technique is utilized in the RIT2NFS-WB(HL) to achieve computational speedup. The RIT2NFS-WB(HL) is applied to examples of online system modeling and sequence prediction to demonstrate the system's functionality.

This paper proposes a new multiobjective optimization approach to designing a fuzzy logic system (FLS) using process data and applies it to the wall-following control of a mobile robot. The objectives considered include both the interpretability and control performance of the FLS. It is assumed that no off-line training data are available in advance, and the rule base is initially empty. All rules are generated through an online clustering and fuzzy set merging (OCFM) algorithm using data generated online during the FLS evaluation process. The OCFM builds a reference rule base that flexibly partitions the input space with distinguishable fuzzy sets (FSs). Based on the reference rule base, a new multiobjective front-guided continuous ant-colony optimization (MO-FCACO) algorithm is proposed to optimize the FLS structure and parameters. In addition to the objective functions defined to evaluate the FLS control performance, a transparency-oriented objective function is defined with constraints imposed on the FS parameters to obtain an interpretable FLS with transparent FSs. The MO-FCACO solves the constrained multiobjective optimization problem by optimizing all of the free parameters in an FLS through ant-path selection, sampling operation, and front-guided optimization processes. The multiobjective FLS design approach is applied to control the orientation and moving speed of a mobile robot in performing the wall-following task. Optimization performance of the MO-FCACO is verified through comparisons with various multiobjective population-based optimization algorithms. Experimental results verify the effectiveness of the designed FLSs in controlling a real robot.

This paper proposes an object localization and segmentation method based on a small set of training images captured from a Kinect red-green-blue-depth (RGB-D) camera. The method consists of three stages. The first stage localizes candidate objects based on the hybrid color features of cluster-based pixel distribution and color entropy and a new fuzzy classifier (FC). In the second stage, the object candidates are then sent to another FC for filtering by using the color feature of entropies of color geometrical distributions. After the two-stage localization using the color features, the depth measurement from the Kinect is used to segment the shape of the object for final localization and shape segmentation. A histogram-based shape feature is used to filter the candidate objects from the first two stages. Experimental results show that good performance is achieved by using only a small set of training images..

Estimation of the visual pulse rate (also called heart rate) refers to extraction of the pulse rate from a facial video. With the studies on extracting photoplethysmography (PPG) signals from a facial video, the non-contacted measurement method has aroused great interest among researchers over the past few years. In this study, a novel one-stage spatio-temporal framework, namely PRnet, is proposed to estimate the pulse rate from a stationary facial video. First, visual pulse rate estimation is defined as a regression task based on deep neural networks, in which a video is mapped to a pulse rate value. Then, 3D convolutional neural networks (Conv3D) and Long short-term memory (LSTM) modules are used to extract spatial and latent temporal information that is hidden in a video. Subsequently, one fully connected layer is applied in the last layer of PRnet to estimate the pulse rate directly. Based on the exquisite framework design, our proposed method realizes competitive performance, especially in terms of processing latency, since it does not rely on power spectral density (PSD) and traditional Fast Fourier Transform (FFT) algorithms. Using our method, only 60 frames of video (2 s) are required for the robust prediction of the pulse rate, whereas 6–30 s of video are typically required for other methods. Finally, a novel visual pulse rate estimation database, which includes pulse rate range at various times of day, is collected to evaluate the proposed framework. The results of extensive experiments demonstrate that PRnet performs competitively while compared with state-of-the-art methods.

Motivation: Renal biopsy is an irreplaceable diagnostic tool for kidney diseases. Glomeruli provide important information for an accurate disease diagnosis. This paper applies deep learning techniques to automate translation of renal pathology images and glomerulus detection to improve the efficiency and accuracy on pathological diagnoses. Methods: This paper first proposes a new method for automatic translation of different renal pathology staining styles using the cycle-consistent Generative Adversarial Network (GAN). This paper then proposes the combination of faster region-based convolutional neural network (R-CNN) with an aspect ratio filter to detect glomeruli in light microscopy images processed with four different stains at various optical magnifications. Finally, this paper improves glomerulus detection at different stains by using translated image stains from the CycleGAN. Results: To show the effectiveness of the translation and detection methods, in addition to quantitative analysis of the results, the involvement of assessment from four physicians is also performed. Experimental results show that the physicians fail to differentiate real and translated stains and the automatic glomerulus detection method outperforms that manually labeled by the physicians. Conclusion: The proposed method works well and improves the efficiency of renal pathological diagnosis. This work contributes in the area of automated medical diagnosis.

This paper proposes a new approach for classifying four types of moving objects in an intelligent transportation system. Pedestrians, cars, motorcycles, and bicycles are classified based on their side views from a fixed camera. A moving object is segmented and tracked using background subtraction, silhouette projection, an area ratio, a Kalman filter, and appearance correlation operations. For the classification of a segmented object, a combination of static and spatiotemporal features based on the cooccurrence of its appearance and the movements of its local parts is proposed. To extract the static appearance features, adaptive block-based gradient intensities and histograms of oriented gradients are proposed. For the spatiotemporal features, the optical-flow-based entropy values of instantaneous and short-term movements are proposed. The former finds the spatial entropy values of the orientations and the amplitudes of optical flows in a block to extract the local movement information from two consecutive image frames. The latter finds the temporal entropy values of the tracked optical flows in different orientation bins to extract the short-term movement information from several consecutive frames. Linear support vector machines with batch incremental learning are proposed to classify the four classes of objects. Experimental results from 12 test video sequences and comparisons with several feature descriptors show the effect of the proposed classification system and the advantage of the proposed features in classification.

In this paper, the interplay and relationship between digital twin and Industrial Internet are discussed at first. The sensing/transmission network capability, which is one of the main characteristics of Industrial Internet, can be a carrier for providing digital twin with a means of data acquisition and transmission. Conversely, with the capability of high-fidelity virtual modeling and simulation computing/analysis, digital twin evolving from lifecycle management for a single product to application in production/manufacturing in the shop-floor/enterprise, can further greatly enhance the simulation computing and analysis of Industrial Internet. This paper proposes a digital twin enhanced Industrial Internet (DT-II) reference framework towards smart manufacturing. To further illustrate the reference framework, the implementation and operation mechanism of DT-II is discussed from three perspectives, including product lifecycle level, intra-enterprise level and inter-enterprise level. Finally, steam turbine is taken as an example to illustrate the application scenes from above three perspectives under the circumstance of DT-II. The differences between with and without DT-II for design and development of steam turbine are also presented.

This paper proposes a new rule-based cooperative framework for multiobjective evolutionary fuzzy systems (FSs). Based on the framework, a multiobjective rule-based cooperative continuous ant-colony optimization (MO-RCCACO) algorithm is proposed to optimize all of the free parameters in FSs. Instead of optimization using a single colony of FSs (solutions), the MO-RCCACO consists of ${r}$ subcolonies of size ${N}$ cooperatively optimizing an FS that consists of ${r}$ rules, with a subcolony optimizing only a single fuzzy rule. In addition, an auxiliary colony is created to store all of the fuzzy rules in the best-so-far ${N}$ FSs to enhance the optimization ability of MO-RCCACO. The performance ranking of different fuzzy rules in the same subcolony is performed based on the multiobjective function values of their participating FSs by using Pareto nondominated sorting and the crowding distance. The MO-RCCACO is applied to find the Pareto-optimal fuzzy controllers (FCs) of a mobile robot for wall following with multiple control objectives. The optimization ability of the MO-RCCACO is verified through comparisons with various multiobjective population-based optimization algorithms in the robot wall-following control problem. Experimental results verify the effectiveness of the MO-RCCACO-based FCs for the boundary following control of a real robot.

This paper proposes a new recurrent neural network (RNN) structure evolved to control the gait of a hexapod robot for fast forward walking. In this evolutionary robot, the gait control problem is formulated as an optimization problem with the objective of a fast forward walking speed and a small deviation in the forward walking direction. Evolutionary optimization of the RNNs through a group-based hybrid metaheuristic algorithm is proposed to find the optimal RNN controller. Preliminary simulation results with comparisons show the advantage of the proposed approach1.

Glomerulus is an important component in human kidney. The appearance of the glomeruli on light microscopic image can provide abundant information for disease diagnosis. Due to the importance of glomeruli on accurate renal disease diagnosis, this paper proposes an automatic method to detect glomeruli in light microscopy images with Periodic Acid Schiff (PAS) or hematoxylin and eosin (H&E) stains at 100x, 200x, or 400x optical magnification. The faster region-based convolutional neural network (R-CNN) is applied to the detection task. The proposed detection approach performs an end-to-end glomerulus detection without any a priori information of the stains and magnifications of the images. The training dataset contains 2,511 images with 3,956 glomeruli. The test dataset contains 482 images with 563 glomeruli. The recall and precision of the test result are 91.54% and 86.50%, respectively, which shows the effectiveness of the proposed detection method.

In several studies, fuzzy classifiers (FCs) have been shown to achieve higher generalization ability when learned through support vector machines (SVMs) compared to learning through neural networks that aim to minimize only the training error. This chapter introduces the learning of FCs using linear SVMs. Two types of FCs with zero-order and high-order Takagi-Sugeno (TS)-type consequents are considered. Given a number of rules, the antecedent parameters in the two FCs are determined using a self-splitting clustering (SSC) algorithm. Regarding the consequent parameter optimization, this chapter first describes the basic concept of linear SVMs followed by its application to the learning of the consequent parameters to endow the FCs with high generalization ability. These two types of FCs are subsequently applied to object detection and shape extraction using a red-green-blue-depth (RGB-D) camera. In this application, after the detection of an object using a color-feature-based FC, the depth information from the camera is used to extract the shape of an object. A histogram-based shape feature is proposed for improving the object detection performance. The performance of the proposed classification approach is evaluated through the detection of different objects and comparisons with various object detection approaches.

This paper proposes an interval type-2 neural fuzzy classifier learned through soft margin minimization (IT2NFC-SMM) and applies it to human body posture classification. The IT2NFC-SMM consists of interval type-2, zero-order Takagi–Sugeno (T–S) fuzzy rules established through online structure learning. The antecedent part of the IT2NFC-SMM uses interval type-2 fuzzy sets to decrease the number of rules and manage noisy data. For parameter learning, the consequent parameters are learned through a linear support vector machine (SVM) for soft margin minimization to improve the generalization ability. The proposed SVM-based learning addresses the problem that the orders of the fuzzy rules in computing the outputs of an interval type-2 fuzzy system depend on the consequent values that are unknown in advance. To address this problem, the IT2NFC-SMM uses weighted bound-set boundaries to simplify the type-reduction operation and a novel crisp-to-interval linear SVM learning algorithm. Based on the soft margin minimization, the antecedent parameters are tuned using the gradient descent algorithm. The IT2NFC-SMM is applied to a vision-based human body posture classification system. The system uses two cameras and novel classification features extracted from a silhouette of the human body to classify the four postures of standing, bending, sitting, and lying. The classification performance of the IT2NFC-SMM is verified through results in clean and noisy classification examples and through the posture classification problem, as well as through comparisons with various type-1 and type-2 fuzzy classifiers. The overall result shows that the IT2NFC-SMM achieves higher classification rates with a smaller or similar model size than the classifiers used for comparison, especially for noisy classification problems.

This paper presents a method that allows two wheeled, mobile robots to navigate unknown environments while cooperatively carrying an object. In the navigation method, a leader robot and a follower robot cooperatively perform either obstacle boundary following (OBF) or target seeking (TS) to reach a destination. The two robots are controlled by fuzzy controllers (FC) whose rules are learned through an adaptive fusion of continuous ant colony optimization and particle swarm optimization (AF-CACPSO), which avoids the time-consuming task of manually designing the controllers. The AF-CACPSO-based evolutionary fuzzy control approach is first applied to the control of a single robot to perform OBF. The learning approach is then applied to achieve cooperative OBF with two robots, where an auxiliary FC designed with the AF-CACPSO is used to control the follower robot. For cooperative TS, a rule for coordination of the two robots is developed. To navigate cooperatively, a cooperative behavior supervisor is introduced to select between cooperative OBF and cooperative TS. The performance of the AF-CACPSO is verified through comparisons with various population-based optimization algorithms for the OBF learning problem. Simulations and experiments verify the effectiveness of the approach for cooperative navigation of two robots.

In contrast with type-1 neural fuzzy systems (NFSs), interval type-2 NFSs process interval membership values are much more computationally expensive in implementation, especially for large-scale problems. Interval type-2 NFSs are conventionally implemented on a single-threaded central processing unit (CPU) with serially processed computation for each input variable and fuzzy rule. Because graphics processing units (GPUs) have many cores that can collectively run many threads in parallel, this paper proposes the implementation of interval type-2 NFSs through the parallel processing on GPUs (IT2NFS-GPU) to reduce the system training time. The structure in the IT2NFS-GPU is built through an online learning approach that is based on rule-firing strength. Parameters in the T2NFS-GPU are tuned using the well-known gradient descent algorithm; therefore, it is easier for users to follow the GPU implementation techniques. In the IT2NFS-GPU, for the parallel computation of the structure and parameter learning algorithms, blocks of threads are partitioned according to the parallel and independent properties of interval boundaries, input variables, and fuzzy rules. Specifically, the IT2NFS-GPU implements the type-reduction operation through the parallel computation of all possible system outputs instead of the traditional iterative procedure. The IT2NFS-GPU is applied to several data-driven learning problems to verify its shorter computing times.

An integrated system that segments and classifies four moving objects.A weight mask is proposed to enhance the distinguishing pixels in a segmented object.A new classification feature vector extracted from a wavelet-transformed space.A hierarchical linear support vector machine classification configuration is proposed. This paper proposes an integrated system for the segmentation and classification of four moving objects, including pedestrians, cars, motorcycles, and bicycles, from their side-views in a video sequence. Based on the use of an adaptive background in the red-green-blue (RGB) color model, each moving object is segmented with its minimum enclosing rectangle (MER) window by using a histogram-based projection approach or a tracking-based approach. Additionally, a shadow removal technique is applied to the segmented objects to improve the classification performance. For the MER windows with different sizes, a window scaling operation followed by an adaptive block-shifting operation is applied to obtain a fixed feature dimension. A weight mask, which is constructed according to the frequency of occurrence of an object in each position within a square window, is proposed to enhance the distinguishing pixels in the rescaled MER window. To extract classification features, a two-level Haar wavelet transform is applied to the rescaled MER window. The local shape features and the modified histogram of oriented gradients (HOG) are extracted from the level-two and level-one sub-bands, respectively, of the wavelet-transformed space. A hierarchical linear support vector machine classification configuration is proposed to classify the four classes of objects. Six video sequences are used to test the classification performance of the proposed method. The computer processing times of the object segmentation, object tracking, and feature extraction and classification approaches are 79ms, 211ms, and 0.01ms, respectively. Comparisons with different well-known classification approaches verify the superiority of the proposed classification method.

This paper proposes the use of evolutionary fuzzy control for a wall-following hexapod robot. The data-driven fuzzy controller (FC) is learned through an adaptive group-based differential evolution (AGDE) algorithm, which avoids the explicit usage of the robot mathematical model and time-consuming manual design effort. In the wall-following task, the inputs of the FC are measurements of three infrared distance sensors mounted on the hexapod robot. The FC controls the swing angle changes of the left- and right-middle legs of the hexapod robot for proper turning performance while simultaneously moving forward. To automate the design of the FC and to improve the performance of control, an AGDE algorithm is proposed. In the AGDE-designed FC, a cost function is defined to quantitatively evaluate the learning performance of an FC based on data generated online. In the AGDE, the solution vectors in a population are adaptively clustered into different groups based on their performances at each iteration. To improve optimization performance, the AGDE adaptively selects components from either the group-based mutant vector or a typical population-based mutant vector in the mutation operation. Simulated and experimental results are gathered to verify the effectiveness and efficiency of the data-driven AGDE-based learning approach.

This paper proposes evolutionary fuzzy control of three robots cooperatively carrying an object in executing a convex wall following behavior. The object is not connected to the robots and may fall off for a failed control. Evolutionary fuzzy control of a single robot for wall following is first performed. Then, evolutionary fuzzy control of two robots cooperatively carrying a long strip object along the wall is performed. For the carry of a larger object for wall following, a third robot is included to cooperate with the two successfully controlled robots. Fuzzy controller of the third robot is also learned through a data-driven evolutionary learning approach. To improve learning efficiency of the FC, the swarm intelligence algorithm of adaptive fusion of continuous ant colony optimization and particle swarm optimization (AF-CACPSO) is employed. Simulations show the effectiveness of the evolutionary fuzzy control approach for the three cooperative object-carrying robots.

This paper proposes a cooperative continuous ant colony optimization (CCACO) algorithm and applies it to address the accuracy-oriented fuzzy systems (FSs) design problems. All of the free parameters in a zero- or first-order Takagi-Sugeno-Kang (TSK) FS are optimized through CCACO. The CCACO algorithm performs optimization through multiple ant colonies, where each ant colony is only responsible for optimizing the free parameters in a single fuzzy rule. The ant colonies cooperate to design a complete FS, with a complete parameter solution vector (encoding a complete FS) that is formed by selecting a subsolution component (encoding a single fuzzy rule) from each colony. Subsolutions in each ant colony are evolved independently using a new continuous ant colony optimization algorithm. In the CCACO, solutions are updated via the techniques of pheromone-based tournament ant path selection, ant wandering operation, and best-ant-attraction refinement. The performance of the CCACO is verified through applications to fuzzy controller and predictor design problems. Comparisons with other population-based optimization algorithms verify the superiority of the CCACO.

We propose a new type-2 neural fuzzy system (FS) learned through structure and parameter learning.A type-1 FS is converted to a type-2 FS by merging highly overlapped type-1 fuzzy sets.A new hardware circuit is proposed to implement an interval type-2 FS with TSK-type consequents. This paper first proposes a type-2 neural fuzzy system (NFS) learned through its type-1 counterpart (T2NFS-T1) and then implements the built IT2NFS-T1 in a field-programmable gate array (FPGA) chip. The antecedent part of each fuzzy rule in the T2NFS-T1 uses interval type-2 fuzzy sets, while the consequent part uses a Takagi-Sugeno-Kang (TSK) type with interval combination weights. The T2NFS-T1 uses a simplified type-reduction operation to reduce system training time and hardware implementation cost. Given a training data set, a TSK type-1 NFS is first learned through structure and parameter learning. The built type-1 fuzzy logic system (FLS) is then extended to a type-2 FLS, where highly overlapped type-1 fuzzy sets are merged into interval type-2 fuzzy sets to reduce the total number of fuzzy sets. Finally, the rule consequent and antecedent parameters in the T2NFS-T1 are tuned using a hybrid of the gradient descent and rule-ordered recursive least square (RLS) algorithms. Simulation results and comparisons with various type-1 and type-2 FLSs verify the effectiveness and efficiency of the T2NFS-T1 for system modeling and prediction problems. A new hardware circuit using both parallel-processing and pipeline techniques is proposed to implement the learned T2NFS-T1 in an FPGA chip. The T2NFS-T1 chip reduces the hardware implementation cost in comparison to other type-2 fuzzy chips.

This paper proposes a new incremental learning approach to endow a Takagi-Sugeno-type fuzzy classification model with high generalization ability. The proposed fuzzy model is learned through incremental support vector machine (SVM) and margin-selected gradient descent learning and is called FM3. In this learning approach, training samples are fed incrementally one-by-one instead of all in one batch. The FM3 evolves from an empty rule set. A one-pass clustering algorithm is used to determine the number of rules and initial fuzzy sets in the rule antecedent part. An online incremental linear SVM is proposed to tune the rule consequent parameters to endow the FM3 with high generalization ability. The use of incremental instead of batch SVM enables the FM3 to handle online training problems with only one new sample available at a time. For antecedent parameter learning, a margin-selected gradient descent algorithm is proposed to prevent overtraining. Simulation results and comparisons with SVMs and fuzzy classifiers with different learning algorithms demonstrate the advantage of the FM3.

This paper proposes a self-splitting fuzzy classifier with support vector learning in expanded high-order consequent space (SFC-SVHC) for classification accuracy improvement. The SFC-SVHC expands the rule-mapped consequent space of a first-order Takagi-Sugeno (TS)-type fuzzy system by including high-order terms to enhance the rule discrimination capability. A novel structure and parameter learning approach is proposed to construct the SFC-SVHC. For structure learning, a variance-based self-splitting clustering (VSSC) algorithm is used to determine distributions of the fuzzy sets in the input space. There are no rules in the SFC-SVHC initially. The VSSC algorithm generates a new cluster by splitting an existing cluster into two according to a predefined cluster-variance criterion. The SFC-SVHC uses trigonometric functions to expand the rule-mapped first-order consequent space to a higher-dimensional space. For parameter optimization in the expanded rule-mapped consequent space, a support vector machine is employed to endow the SFC-SVHC with high generalization ability. Experimental results on several classification benchmark problems show that the SFC-SVHC achieves good classification results with a small number of rules. Comparisons with different classifiers demonstrate the superiority of the SFC-SVHC in classification accuracy.

This paper proposes a multi-objective, rule-coded, advanced, continuous-ant-colony optimization (MO-RACACO) algorithm for fuzzy controller (FC) design and its application to multi-objective, wall-following control for a mobile robot. In the MO-RACACO-based FC design approach, the number of rules and all free parameters in each rule are optimized using the MORACACO algorithm. This is a complex multi-objective optimization problem that considers both the optimization of discrete variables (number of rules) and continuous variables (rule parameters). To address this problem, the MO-RACACO uses a rule-coded individual (solution) representation and a rule-based mutation operation to find Pareto-optimal solutions with different numbers of rules. New solutions in the MO-RACACO are generated using a pheromone-level-based adaptive elite-tournament path selection strategy followed by a Gaussian sampling operation. The MO-RACACO-based FC design approach is applied to a multiobjective, wall-following problem for a mobile robot. Three objectives are defined so that the robot is collision-free, maintains a constant distance from the wall, and moves smoothly at a high speed. This automatic design approach avoids the time-consuming manual design of fuzzy rules and the exhaustive collection of input-output training pairs. The performance of the MORACACO- based control is verified through comparisons with various multi-objective population-based optimization algorithms (MOPOAs) in multi-objective FC optimization problems. This study also includes experiments that demonstrate robot wallfollowing control using an actual mobile robot.

This paper proposes a reduced interval type-2 neural fuzzy system using weighted bound-set boundaries (RIT2NFS-WB) for the simplification of type-reduction operations. The objective of this simplification is to reduce the system training time in software implementation and chip size in hardware implementation, especially when the number of rules is large. The antecedent part in the RIT2NFS-WB uses interval type-2 fuzzy sets (IT2FSs), and the consequent part can be of the Takagi-Sugeno-Kang (TSK) or Mamdani type. The RIT2NFS-WB is built through an online structure and parameter learning to improve model accuracy. In addition, the interpretability of the RIT2NFS-WB is improved by considering distributions of the IT2FSs in input variables. A distinguishability-oriented cost function is used in parameter learning to generate distinguishable IT2FSs and improve semantics-based interpretability. For highly overlapped IT2FSs, they are merged to reduce the number of IT2FSs and improve complexity-based interpretability. The software-implemented TSK-type RIT2NFS-WB is hardware-implemented on a field-programmable gate array chip. To accelerate the chip execution speed, the chip utilizes not only the parallel execution properties of fuzzy rules and bound-set boundaries but the pipeline technique as well. In particular, the flexibility of the chip is considered so that no redesign of the circuits is required when the RIT2NFS-WB is applied to different problems. The characteristics of the software- and hardware-implemented RIT2NFS-WB are verified through various examples and comparisons with various type-1 and interval type-2 fuzzy models.

This paper proposes a novel approach to specific object detection in complex scenes using color-based entropy features and a fuzzy classifier (FC). Appearances of the detected objects are assumed to contain multiple colors in non-homogeneous distributions that make it difficult to detect these objects using shape features. The proposed detection approach consists of two filtering phases with two different novel color-based entropy features. The first phase filters a test pattern with the entropy of color component (ECC). A self-splitting clustering (SSC) algorithm is proposed to automatically generate clusters in the hue and saturation (HS) color space according to the composing pixels of an object. The ECC value is computed from histograms of pixels in the found clusters and is used to generate object candidates. The second filtering phase uses the entropies of geometric color distributions (EGCD) to filter the object candidates obtained from the first phase. An EGCD is computed for each of the clustered composing colors of a candidate object. The EGCD values are fed to an FC to enable advanced filtering. A new FC using the SSC algorithm and support vector machine (FC-SSCSVM) for antecedent and consequent parameter learning, respectively, is proposed to improve detection performance. Experimental results on the detection of different objects and comparisons with various detection approaches and classifiers verify the advantage of the proposed detection approach using the FC-SSCSVM.

This paper proposes an evolutionary fuzzy lead-lag control approach for coordinated control of flexible AC transmission system (FACTS) devices in a multi-machine power system. The FACTS devices used are a thyristor-controlled series capacitor (TCSC) and a static var compensator (SVC), both of which are equipped with a fuzzy lead-lag controller to improve power system dynamic stability. The fuzzy lead-lag controller uses a fuzzy controller (FC) to adaptively determine the parameters of two lead-lag controllers at each control step according to the deviations of generator rotor speeds. This paper proposes an Advanced Continuous Ant Colony Optimization (ACACO) algorithm to optimize all of the free parameters in the FC, which avoids the time-consuming task of parameter selection by human experts. The effectiveness and efficiency of the proposed evolutionary fuzzy lead-lag controller for oscillation damping control is verified through control of a multi-machine power system and comparisons with other lead-lag controllers and various population-based optimization algorithms.

This paper proposes evolutionary wall-following control of a mobile robot using an interval type-2 fuzzy controller (IT2FC) with species-differential-evolution-activated continuous ant colony optimization (SDE-CACO). Both the position and speed of a mobile robot are controlled by using two IT2FCs to improve noise resistance ability. A new cost function is defined to accurately evaluate the wall-following performance of an evolutionary IT2FC. A two-stage training approach is proposed that learns a position IT2FC followed by a speed IT2FC to optimize both the wall-following accuracy and the moving speed. The proposed learning approach avoids the time consuming task of the exhaustive collection of supervised input-output training pairs. All fuzzy rules are generated online using a clustering-based approach during the evolutionary learning process. All of the free parameters in an online-generated IT2FC are optimized using SDE-CACO, in which an SDE mutation operation is incorporated within a continuous ACO to improve its explorative ability. The proposed SDE-CACO is compared with various population-based optimization algorithms to demonstrate its efficiency and effectiveness in the wall-following control problem. This study also includes experiments that demonstrate wall-following control utilizing a real mobile robot.

This paper proposes a new fuzzy classifier (FC)-based face localization approach. The FC used is a self-organizing TS-type fuzzy network with support vector learning (SOTFN-SV). The SOTFN-SV learns consequent parameters using a linear support vector machine to improve generalization ability. The FC is first applied to segment human skin pixels in scaled hue and saturation (hS) color space, after which connected skin-color regions are regarded as face candidates. The FC is then applied to detect and localize faces from the candidates. The proposed FC-based face localization approach uses shape and wavelet-localized focus color features. A best fitting ellipse of each face candidate is found to obtain shape features. Focus color features are extracted from four focus regions, including the two eyes, the mouth, and the face skin-color region. To find these focus color regions, the Haar-wavelet transformation is first applied to the face candidates in the YCb color space to localize all possible pairs of eye candidates. The mouth region is then localized according to its geometric relationship with the eyes. The hS color features of the located eyes, mouth, and face skin are extracted. These focus color features, together with shape features, serve as inputs to another FC for final face localization. Comparisons with various classifiers and face detection methods demonstrate the advantage of the FC-based skin color segmentation and face localization method.

This paper considers the problem of data-driven fuzzy controller (FC) design with the objectives of not only high control accuracy but also high interpretability in the control rules. Because the tradeoff between the two objectives, a multi-objective genetic algorithm is employed to find a set of Pareto-optimal FCs. The optimization is based on an initial FC structure online generated through clustering of input data with the input space flexibly partitioned. A constrained objective function is defined to measure fuzzy set transparency and optimization of which improves FC interpretability. Since the dimension of each FC in the parameter solution population changes with the generation of a new rule, a new solution update method is proposed in this paper. The data-driven interpretable fuzzy control approach is applied to control a nonlinear plant in simulation to verify its performance.