Search

Because of its long running time, complex working environment, and for other reasons, a gear is prone to failure, and early failure is difficult to detect by direct observation; therefore, fault diagnosis of gears is very necessary. Neural network algorithms have been widely used to realize gear fault diagnosis, but the structure of the neural network model is complicated, the training time is long and the model is not easy to converge. To solve the above problems and combine the advantages of the ResNeXt50 model in the extraction of image features, this paper proposes a gearbox fault detection method that integrates the convolutional block attention module (CBAM). Firstly, the CBAM is embedded in the ResNeXt50 network to enhance the extraction of image channels and spatial features. Secondly, the different time–frequency analysis method was compared and analyzed, and the method with the better effect was selected to convert the one-dimensional vibration signal in the open data set of the gearbox into a two-dimensional image, eliminating the influence of the redundant background noise, and took it as the input of the model for training. Finally, the accuracy and the average training time of the model were obtained by entering the test set into the model, and the results were compared with four other classical convolutional neural network models. The results show that the proposed method performs well both in fault identification accuracy and average training time under two working conditions, and it also provides some references for existing gear failure diagnosis research. [ABSTRACT FROM AUTHOR]

With the exponential growth of data, solving classification or regression tasks by mining time series data has become a research hotspot. Commonly used methods include machine learning, artificial neural networks, and so on. However, these methods only extract the continuous or discrete features of sequences, which have the drawbacks of low information utilization, poor robustness, and computational complexity. To solve these problems, this paper innovatively uses Wasserstein distance instead of Kullback–Leibler divergence and uses it to construct an autoencoder to learn discrete features of time series. Then, a hidden Markov model is used to learn the continuous features of the sequence. Finally, stacking is used to ensemble the two models to obtain the final model. This paper experimentally verifies that the ensemble model has lower computational complexity and is close to state-of-the-art classification accuracy. [ABSTRACT FROM AUTHOR]

With the widespread adoption of service-oriented architectures (SOA), services with the same functionality but the different Quality of Service (QoS) are proliferating, which is challenging the ability of users to build high-quality services. It is often costly for users to evaluate the QoS of all feasible services; therefore, it is necessary to investigate QoS prediction algorithms to help users find services that meet their needs. In this paper, we propose a QoS prediction algorithm called the MFDK model, which is able to fill in historical sparse QoS values by a non-negative matrix decomposition algorithm and predict future QoS values by a deep neural network. In addition, this model uses a Kalman filter algorithm to correct the model prediction values with real-time QoS observations to reduce its prediction error. Through extensive simulation experiments on the WS-DREAM dataset, we analytically validate that the MFDK model has better prediction accuracy compared to the baseline model, and it can maintain good prediction results under different tensor densities and observation densities. We further demonstrate the rationality of our proposed model and its prediction performance through model ablation experiments and parameter tuning experiments. [ABSTRACT FROM AUTHOR]

To ensure the safe operation of highway traffic lines, given the imperfect feature extraction of existing road pit defect detection models and the practicability of detection equipment, this paper proposes a lightweight target detection algorithm with enhanced feature extraction based on the YOLO (You Only Look Once) algorithm. The BIFPN (Bidirectional Feature Pyramid Network) network structure is used for multi-scale feature fusion to enhance the feature extraction ability, and Varifocal Loss is used to optimize the sample imbalance problem, which improves the accuracy of road defect target detection. In the evaluation test of the model in the constructed PCD1 (Pavement Check Dataset) dataset, the mAP@.5 (mean Average Precision when IoU = 0.5) of the BV-YOLOv5S (BiFPN Varifocal Loss-YOLOv5S) model increased by 4.1%, 3%, and 0.9%, respectively, compared with the YOLOv3-tiny, YOLOv5S, and B-YOLOv5S (BiFPN-YOLOv5S; BV-YOLOv5S does not use the Improved Focal Loss function) models. Through the analysis and comparison of experimental results, it is proved that the proposed BV-YOLOv5S network model performs better and is more reliable in the detection of pavement defects and can meet the needs of road safety detection projects with high real-time and flexibility requirements. [ABSTRACT FROM AUTHOR]

Ship fire is one of the greatest dangers to ship navigation safety. Nevertheless, typical detection methods have limited detection effectiveness and accuracy due to distance restrictions and ship motion. Although the issue can be addressed by image recognition algorithms based on deep learning, the computational complexity and efficiency for ship detection are tough. This paper proposes a lightweight target identification technique based on the modified YOLOv4-tiny algorithm for the precise and efficient detection of ship fires, taking into account the distinctive characteristics of ship fires and the marine environment. Initially, a multi-scale detection technique is applied to broaden the detection range and integrate deep semantic information, thereby enhancing the feature information of small targets and obscured objects and improving the detection precision. Then, the proposed algorithm employs the SE attention mechanism for inter-channel feature fusion to improve the capability of feature extraction and the precision of ship fire detection. Last but not least, picture transformation and migration learning are added to the small ship fire dataset to accelerate the convergence pace, improve the convergence effect, and reduce dataset dependence. The simulation experiments reveal that the proposed I-YOLOv4-tiny + SE model outperforms the benchmark algorithm in terms of ship fire detection accuracy and detection efficiency and that it satisfies the real-time ship fire warning criteria in demanding maritime environments. [ABSTRACT FROM AUTHOR]

Vehicle fault detection and diagnosis (VFDD) along with predictive maintenance (PdM) are indispensable for early diagnosis in order to prevent severe accidents due to mechanical malfunction in urban environments. This paper proposes an early voiceprint driving fault identification system using machine learning algorithms for classification. Previous studies have examined driving fault identification, but less attention has focused on using voiceprint features to locate corresponding faults. This research uses 43 different common vehicle mechanical malfunction condition voiceprint signals to construct the dataset. These datasets were filtered by linear predictive coefficient (LPC) and wavelet transform(WT). After the original voiceprint fault sounds were filtered and obtained the main fault characteristics, the deep neural network (DNN), convolutional neural network (CNN), and long short-term memory (LSTM) architectures are used for identification. The experimental results show that the accuracy of the CNN algorithm is the best for the LPC dataset. In addition, for the wavelet dataset, DNN has the best performance in terms of identification performance and training time. After cross-comparison of experimental results, the wavelet algorithm combined with DNN can improve the identification accuracy by up to 16.57% compared with other deep learning algorithms and reduce the model training time by up to 21.5% compared with other algorithms. Realizing the cross-comparison of recognition results through various machine learning methods, it is possible for the vehicle to proactively remind the driver of the real-time potential hazard of vehicle machinery failure. [ABSTRACT FROM AUTHOR]

This paper presents a method of high-capacity and transparent watermarking based on the usage of deep neural networks with the adjustable subsquares properties algorithm to encode the data of a watermark in high-quality video using the H.265/HEVC (High-Efficiency Video Coding) codec. The aim of the article is to present a method of embedding a watermark in a video with HEVC codec compression by making changes in a video in a way that is not noticeable to the naked eye. The method presented here is characterised by focusing on ensuring the accuracy of the original image in relation to the watermarked image, providing the transparency of the embedded watermark, while ensuring its survival after compression by the HEVC codec. The article includes a presentation of the practical results of watermark embedding with a built-in variation mechanism of its capacity and resistance, thanks to the adjustable subsquares properties algorithm. The obtained PSNR (peak signal-to-noise ratio) results are at the level of 40 dB or better. There is the possibility of the complete recovery of a watermark from a single frame compressed in the CRF (constant rate factor) range of up to 16, resulting in a BER (bit error rate) equal to 0 for the received watermark. [ABSTRACT FROM AUTHOR]

Gas utilization ratio (GUR) is an important indicator that is used to evaluate the energy consumption of blast furnaces (BFs). Currently, the existing methods cannot predict the GUR accurately. In this paper, we present a novel data-driven model for predicting the GUR. The proposed approach utilized both the TS fuzzy neural network (TS-FNN) and the particle swarm algorithm (PSO) to predict the GUR. The particle swarm algorithm (PSO) is applied to optimize the parameters of the TS-FNN in order to decrease the error caused by the inaccurate initial parameter. This paper also applied the box graph (Box-plot) method to eliminate the abnormal value of the raw data during the data preprocessing. This method can deal with the data which does not obey the normal distribution which is caused by the complex industrial environments. The prediction results demonstrate that the optimization model based on PSO and the TS-FNN approach achieves higher prediction accuracy compared with the TS-FNN model and SVM model and the proposed approach can accurately predict the GUR of the blast furnace, providing an effective way for the on-line blast furnace distribution control. [ABSTRACT FROM AUTHOR]

Analyzing vibration data using deep neural networks is an effective way to detect damages in rotating machinery at an early stage. However, the black-box approach of these methods often does not provide a satisfactory solution because the cause of classifications is not comprehensible to humans. Therefore, this work investigates the application of the explainable AI (XAI) algorithms to convolutional neural networks for vibration-based condition monitoring. Thus, the three XAI algorithms GradCAM, LRP and LIME with a modified perturbation strategy are applied to classifications based on the Fourier transform as well as the order analysis of the vibration signal. The following visualization as frequency-RPM maps and order-RPM maps allows for an effective assessment of saliency values for variable periodicity of the data, which translates to a varying rotation speed of a real-world machine. To compare the explanatory power of the XAI methods, investigations are first carried out with a synthetic data set with known class-specific characteristics. Both a visual and a quantitative analysis of the resulting saliency maps are presented. Then, a real-world data set for vibration-based imbalance classification on an electric motor, which runs at a broad range of rotation speeds, is used. The results indicate that the investigated algorithms are each partially successful in providing sample-specific saliency maps which highlight class-specific features and omit features which are not relevant for classification. [ABSTRACT FROM AUTHOR]

Metal oxide semiconductor (MOS) gas sensors have many advantages, but the main obstacle to their widespread use is the cross-sensitivity observed when using this type of detector to analyze gas mixtures. Thermal modulation of the heater integrated with a MOS gas sensor reduced this problem and is a promising solution for applications requiring the selective detection of volatile compounds. Nevertheless, the interpretation of the sensor output signals, which take the form of complex, unique patterns, is difficult and requires advanced signal processing techniques. The study focuses on the development of a methodology to measure and process the output signal of a thermally modulated MOS gas sensor based on a B-spline curve and artificial neural networks (ANNs), which enable the quantitative analysis of volatile components (ethanol and acetone) coexisting in mixtures. B-spline approximation applied in the first stage allowed for the extraction of relevant information from the gas sensor output voltage and reduced the size of the measurement dataset while maintaining the most vital features contained in it. Then, the determined parameters of the curve were used as the input vector for the ANN model based on the multilayer perceptron structure. The results show great usefulness of the combination of B-spline and ANN modeling techniques to improve response selectivity of a thermally modulated MOS gas sensor. [ABSTRACT FROM AUTHOR]

The heterogeneity of wireless receiving devices, co-channel interference, and multi-path effect make the received signal strength indication (RSSI) of Wi-Fi fluctuate greatly, which seriously degrades the RSSI-based positioning accuracy. Signal strength difference (DIFF), a calibration-free solution for handling the received signal strength variance between diverse devices, can effectively reduce the negative impact of signal fluctuation. However, DIFF also leads to the explosion of the RSSI data dimension, expanding the number of dimensions from m to C m 2 , which reduces the positioning efficiency. To this end, we design a data hierarchical processing strategy based on a building-floor-specific location, which effectively improves the efficiency of high-dimensional data processing. Moreover, based on a deep neural network (DNN), we design three different positioning algorithms for multi-building, multi-floor, and specific-location respectively, extending the indoor positioning from the single plane to three dimensions. Specifically, in the stage of data preprocessing, we first create the original RSSI database. Next, we create the optimized RSSI database by identifying and deleting the unavailable data in the RSSI database. Finally, we perform DIFF processing on the optimized RSSI database to create the DIFF database. In the stage of positioning, firstly, we design an improved multi-building positioning algorithm based on a denoising autoencoder (DAE). Secondly, we design an enhanced DNN for multi-floor positioning. Finally, the newly deep denoising autoencoder (DDAE) used for specific location positioning is proposed. The experimental results show that the proposed algorithms have better positioning efficiency and accuracy compared with the traditional machine learning algorithms and the current advanced deep learning algorithms. [ABSTRACT FROM AUTHOR]

With the development of artificial intelligence technology, the behavior decision-making of an intelligent smart marine autonomous surface ship (SMASS) has become particularly important. This research proposed local path planning and a behavior decision-making approach based on improved Proximal Policy Optimization (PPO), which could drive an unmanned SMASS to the target without requiring any human experiences. In addition, a generalized advantage estimation was added to the loss function of the PPO algorithm, which allowed baselines in PPO algorithms to be self-adjusted. At first, the SMASS was modeled with the Nomoto model in a simulation waterway. Then, distances, obstacles, and prohibited areas were regularized as rewards or punishments, which were used to judge the performance and manipulation decisions of the vessel Subsequently, improved PPO was introduced to learn the action–reward model, and the neural network model after training was used to manipulate the SMASS's movement. To achieve higher reward values, the SMASS could find an appropriate path or navigation strategy by itself. After a sufficient number of rounds of training, a convincing path and manipulation strategies would likely be produced. Compared with the proposed approach of the existing methods, this approach is more effective in self-learning and continuous optimization and thus closer to human manipulation. [ABSTRACT FROM AUTHOR]

The information about optical flow, i.e., the movement of pixels between two consecutive images from a video sequence, is used in many vision systems, both classical and those based on deep neural networks. In some robotic applications, e.g., in autonomous vehicles, it is necessary to calculate the flow in real time. This represents a challenging task, especially for high-resolution video streams. In this work, two gradient-based algorithms—Lucas–Kanade and Horn–Schunck—were implemented on a ZCU 104 platform with Xilinx Zynq UltraScale+ MPSoC FPGA. A vector data format was used to enable flow calculation for a 4K (Ultra HD, 3840 × 2160 pixels) video stream at 60 fps. In order to detect larger pixel displacements, a multi-scale approach was used in both algorithms. Depending on the scale, the calculations were performed for different data formats, allowing for more efficient processing by reducing resource utilisation. The presented solution allows real-time optical flow determination in multiple scales for a 4K resolution with estimated energy consumption below 6 W. The algorithms realised in this work can be a component of a larger vision system in advanced surveillance systems or autonomous vehicles. [ABSTRACT FROM AUTHOR]

An innovative monitoring-while-drilling method of pressure relief drilling was proposed in a previous study, and the periodic appearance of amplitude concentrated enlargement zone in vibration signals can represent the drilling depth. However, there is a lack of a high accuracy model to automatically identify the amplitude concentrated enlargement zone. So, in this study, a neural network model is put forward based on single-sensor and multi-sensor prediction results. The neural network model consists of one Deep Neural Network (DNN) and four Long Short-Term Memory (LSTM) networks. The accuracy is only 92.72% when only using single-sensor data for identification, while the proposed multiple neural network model could improve the accuracy to being greater than 97.00%. In addition, an optimization method was supplemented to eliminate some misjudgment due to data anomalies, which improved the final accuracy to the level of manual recognition. Finally, the research results solved the difficult problem of identifying the amplitude concentrated enlargement zone and provided the foundation for automatically identifying the drilling depth. [ABSTRACT FROM AUTHOR]

The problems of the neural network-based nonuniformity correction algorithm for infrared focal plane arrays mainly concern slow convergence speed and ghosting artifacts. In general, the more stringent the inhibition of ghosting, the slower the convergence speed. The factors that affect these two problems are the estimated desired image and the learning rate. In this paper, we propose a learning rate rule that combines adaptive threshold edge detection and a temporal gate. Through the noise estimation algorithm, the adaptive spatial threshold is related to the residual nonuniformity noise in the corrected image. The proposed learning rate is used to effectively and stably suppress ghosting artifacts without slowing down the convergence speed. The performance of the proposed technique was thoroughly studied with infrared image sequences with both simulated nonuniformity and real nonuniformity. The results show that the deghosting performance of the proposed method is superior to that of other neural network-based nonuniformity correction algorithms and that the convergence speed is equivalent to the tested deghosting methods. [ABSTRACT FROM AUTHOR]

The wide spread usage of wearable sensors such as in smart watches has provided continuous access to valuable user generated data such as human motion that could be used to identify an individual based on his/her motion patterns such as, gait. Several methods have been suggested to extract various heuristic and high-level features from gait motion data to identify discriminative gait signatures and distinguish the target individual from others. However, the manual and hand crafted feature extraction is error prone and subjective. Furthermore, the motion data collected from inertial sensors have complex structure and the detachment between manual feature extraction module and the predictive learning models might limit the generalization capabilities. In this paper, we propose a novel approach for human gait identification using time-frequency (TF) expansion of human gait cycles in order to capture joint 2 dimensional (2D) spectral and temporal patterns of gait cycles. Then, we design a deep convolutional neural network (DCNN) learning to extract discriminative features from the 2D expanded gait cycles and jointly optimize the identification model and the spectro-temporal features in a discriminative fashion. We collect raw motion data from five inertial sensors placed at the chest, lower-back, right hand wrist, right knee, and right ankle of each human subject synchronously in order to investigate the impact of sensor location on the gait identification performance. We then present two methods for early (input level) and late (decision score level) multi-sensor fusion to improve the gait identification generalization performance. We specifically propose the minimum error score fusion (MESF) method that discriminatively learns the linear fusion weights of individual DCNN scores at the decision level by minimizing the error rate on the training data in an iterative manner. 10 subjects participated in this study and hence, the problem is a 10-class identification task. Based on our experimental results, 91% subject identification accuracy was achieved using the best individual IMU and 2DTF-DCNN.We then investigated our proposed early and late sensor fusion approaches, which improved the gait identification accuracy of the system to 93.36% and 97.06%, respectively. [ABSTRACT FROM AUTHOR]

Considering the installation cost and coverage, the received signal strength indicator (RSSI)-based indoor positioning system is widely used across the world. However, the indoor positioning performance, due to the interference of wireless signals that are caused by the complex indoor environment that includes a crowded population, cannot achieve the demands of indoor location-based services. In this paper, we focus on increasing the signal strength estimation accuracy considering the population density, which is different to the other RSSI-based indoor positioning methods. Therefore, we propose a new wireless signal compensation model considering the population density, distance, and frequency. First of all, the number of individuals in an indoor crowded scenario can be calculated by our convolutional neural network (CNN)-based human detection approach. Then, the relationship between the population density and the signal attenuation is described in our model. Finally, we use the trilateral positioning principle to realize the pedestrian location. According to the simulation and tests in the crowded scenarios, the proposed model increases the accuracy of the signal strength estimation by 1.53 times compared to that without considering the human body. Therefore, the localization accuracy is less than 1.37 m, which indicates that our algorithm can improve the indoor positioning performance and is superior to other RSSI models. [ABSTRACT FROM AUTHOR]

The development of intrusion detection systems (IDS) that are adapted to allow routers and network defence systems to detect malicious network traffic disguised as network protocols or normal access is a critical challenge. This paper proposes a novel approach called SCDNN, which combines spectral clustering (SC) and deep neural network (DNN) algorithms. First, the dataset is divided into k subsets based on sample similarity using cluster centres, as in SC. Next, the distance between data points in a testing set and the training set is measured based on similarity features and is fed into the deep neural network algorithm for intrusion detection. Six KDD-Cup99 and NSL-KDD datasets and a sensor network dataset were employed to test the performance of the model. These experimental results indicate that the SCDNN classifier not only performs better than backpropagation neural network (BPNN), support vector machine (SVM), random forest (RF) and Bayes tree models in detection accuracy and the types of abnormal attacks found. It also provides an effective tool of study and analysis of intrusion detection in large networks. [ABSTRACT FROM AUTHOR]

This paper proposes a real-time feature extraction VLSI architecture for high-resolution images based on the accelerated KAZE algorithm. Firstly, a new system architecture is proposed. It increases the system throughput, provides flexibility in image resolution, and offers trade-offs between speed and scaling robustness. The architecture consists of a two-dimensional pipeline array that fully utilizes computational similarities in octaves. Secondly, a substructure (block-serial discrete-time cellular neural network) that can realize a nonlinear filter is proposed. This structure decreases the memory demand through the removal of data dependency. Thirdly, a hardware-friendly descriptor is introduced in order to overcome the hardware design bottleneck through the polar sample pattern; a simplified method to realize rotation invariance is also presented. Finally, the proposed architecture is designed in TSMC 65 nm CMOS technology. The experimental results show a performance of 127 fps in full HD resolution at 200 MHz frequency. The peak performance reaches 181 GOPS and the throughput is double the speed of other state-of-the-art architectures. [ABSTRACT FROM AUTHOR]

Single-shot stereo 3D shape measurement is becoming more popular due to its advantages of noise robustness and short acquisition period. One of the key problems is stereo matching, which is related to the efficiency of background segmentation and seed point generation, etc. In this paper, a more efficient and automated matching algorithm based on digital image correlation (DIC) is proposed. The standard deviation of image gradients and an adaptive threshold are employed to segment the background. Scale-invariant feature transform (SIFT)-based feature matching and two-dimensional triangulation are combined to estimate accurate initial parameters for seed point generation. The efficiency of background segmentation and seed point generation, as well as the measuring precision, are evaluated by experimental simulation and real tests. Experimental results show that the average segmentation time for an image with a resolution of 1280 × 960 pixels is 240 milliseconds. The efficiency of seed point generation is verified to be high with different convergence criteria. [ABSTRACT FROM AUTHOR]

In this paper, we propose a cost-effective localization solution for land vehicles, which can simultaneously adapt to the uncertain noise of inertial sensors and bridge Global Positioning System (GPS) outages. First, three Unscented Kalman filters (UKFs) with different noise covariances are introduced into the framework of Interacting Multiple Model (IMM) algorithm to form the proposed IMM-based UKF, termed as IMM-UKF. The IMM algorithm can provide a soft switching among the three UKFs and therefore adapt to different noise characteristics. Further, two IMM-UKFs are executed in parallel when GPS is available. One fuses the information of low-cost GPS, in-vehicle sensors, and micro electromechanical system (MEMS)-based reduced inertial sensor systems (RISS), while the other fuses only in-vehicle sensors and MEMS-RISS. The differences between the state vectors of the two IMM-UKFs are considered as training data of a Grey Neural Network (GNN) module, which is known for its high prediction accuracy with a limited amount of samples. The GNN module can predict and compensate position errors when GPS signals are blocked. To verify the feasibility and effectiveness of the proposed solution, road-test experiments with various driving scenarios were performed. The experimental results indicate that the proposed solution outperforms all the compared methods. [ABSTRACT FROM AUTHOR]

In this paper a modified wavelet synthesis algorithm for continuous wavelet transform is proposed, allowing one to obtain a guaranteed approximation of the maternal wavelet to the sample of the analyzed signal (overlap match) and, at the same time, a formalized representation of the wavelet. What distinguishes this method from similar ones? During the procedure of wavelets' synthesis for continuous wavelet transform it is proposed to use splines and artificial neural networks. The paper also suggests a comparative analysis of polynomial, neural network, and wavelet spline models. It also deals with feasibility of using these models in the synthesis of wavelets during such studies like fine structure of signals, as well as in analysis of large parts of signals whose shape is variable. A number of studies have shown that during the wavelets' synthesis, the use of artificial neural networks (based on radial basis functions) and cubic splines enables the possibility of obtaining guaranteed accuracy in approaching the maternal wavelet to the signal's sample (with no approximation error). It also allows for its formalized representation, which is especially important during software implementation of the algorithm for calculating the continuous conversions at digital signal processors and microcontrollers. This paper demonstrates the possibility of using synthesized wavelet, obtained based on polynomial, neural network, and spline models, during the performance of an inverse continuous wavelet transform. [ABSTRACT FROM AUTHOR]

Capability assessment plays a crucial role in the demonstration and construction of equipment. To improve the accuracy and stability of capability assessment, we study the neural network learning algorithms in the field of capability assessment and index sensitivity. Aiming at the problem of overfitting and parameter optimization in neural network learning, the paper proposes an improved machine learning algorithm—the Ensemble Learning Based on Policy Optimization Neural Networks (ELPONN) with the policy optimization and ensemble learning. This algorithm presents an optimized neural network learning algorithm through different strategies evolution, and builds an ensemble learning model of multi-intelligent algorithms to assess the capability and analyze the sensitivity of the indexes. Through the assessment of capabilities, the algorithm effectively avoids parameter optimization from entering the minimum point in performance to improve the accuracy of equipment capability assessment, which is significantly better than previous neural network assessment methods. The experimental results show that the mean relative error is 4.10%, which is better than BP, GABP, and early stopping. The ELPONN algorithm has better accuracy and stability performance, and meets the requirements of capability assessment. [ABSTRACT FROM AUTHOR]

This paper proposes a parallel fixed point radial basis function (RBF) artificial neural network (ANN), implemented in a field programmable gate array (FPGA) trained online with a least mean square (LMS) algorithm. The processing time and occupied area were analyzed for various fixed point formats. The problems of precision of the ANN response for nonlinear classification using the XOR gate and interpolation using the sine function were also analyzed in a hardware implementation. The entire project was developed using the System Generator platform (Xilinx), with a Virtex-6 xc6vcx240t-1ff1156 as the target FPGA. [ABSTRACT FROM AUTHOR]

The autonomous vehicle is an automated system equipped with features like environment perception, decision-making, motion planning, and control and execution technology. Navigating in an unstructured and complex environment is a huge challenge for autonomous vehicles, due to the irregular shape of road, the requirement of real-time planning, and the nonholonomic constraints of vehicle. This paper presents a motion planning method, based on the Radial Basis Function (RBF) neural network, to guide the autonomous vehicle in unstructured environments. The proposed algorithm extracts the drivable region from the perception grid map based on the global path, which is available in the road network. The sample points are randomly selected in the drivable region, and a gradient descent method is used to train the RBF network. The parameters of the motion-planning algorithm are verified through the simulation and experiment. It is observed that the proposed approach produces a flexible, smooth, and safe path that can fit any road shape. The method is implemented on autonomous vehicle and verified against many outdoor scenes; furthermore, a comparison of proposed method with the existing well-known Rapidly-exploring Random Tree (RRT) method is presented. The experimental results show that the proposed method is highly effective in planning the vehicle path and offers better motion quality. [ABSTRACT FROM AUTHOR]

Gravity matching navigation algorithm is one of the key technologies for gravity aided inertial navigation systems. With the development of intelligent algorithms, the powerful search ability of the Artificial Bee Colony (ABC) algorithm makes it possible to be applied to the gravity matching navigation field. However, existing search mechanisms of basic ABC algorithms cannot meet the need for high accuracy in gravity aided navigation. Firstly, proper modifications are proposed to improve the performance of the basic ABC algorithm. Secondly, a new search mechanism is presented in this paper which is based on an improved ABC algorithm using external speed information. At last, modified Hausdorff distance is introduced to screen the possible matching results. Both simulations and ocean experiments verify the feasibility of the method, and results show that the matching rate of the method is high enough to obtain a precise matching position. [ABSTRACT FROM AUTHOR]

One of the most powerful techniques to diagnose cardiovascular diseases is to analyze the electrocardiogram (ECG). To increase diagnostic sensitivity, the ECG might need to be acquired using an ambulatory system, as symptoms may occur during a patient's daily life. In this paper, we propose using an ambulatory ECG (aECG) recording device with a low number of leads and then estimating the views that would have been obtained with a standard ECG location, reconstructing the complete Standard 12-Lead System, the most widely used system for diagnosis by cardiologists. Four approaches have been explored, including Linear Regression with ECG segmentation and Artificial Neural Networks (ANN). The best reconstruction algorithm is based on ANN, which reconstructs the actual ECG signal with high precision, as the results bring a high accuracy (RMS Error < 13 μ V and CC > 99.7%) for the set of patients analyzed in this paper. This study supports the hypothesis that it is possible to reconstruct the Standard 12-Lead System using an aECG recording device with less leads. [ABSTRACT FROM AUTHOR]

Currently, intelligent security systems are widely deployed in indoor buildings to ensure the safety of people in shopping malls, banks, train stations, and other indoor buildings. Multi-Object Tracking (MOT), as an important component of intelligent security systems, has received much attention from many researchers in recent years. However, existing multi-objective tracking algorithms still suffer from trajectory drift and interruption problems in crowded scenes, which cannot provide valuable data for managers. In order to solve the above problems, this paper proposes a Multi-Object Tracking algorithm for RGB-D images based on Asymmetric Dual Siamese networks (ADSiamMOT-RGBD). This algorithm combines appearance information from RGB images and target contour information from depth images. Furthermore, the attention module is applied to repress the redundant information in the combined features to overcome the trajectory drift problem. We also propose a trajectory analysis module, which analyzes whether the head movement trajectory is correct in combination with time-context information. It reduces the number of human error trajectories. The experimental results show that the proposed method in this paper has better tracking quality on the MICC, EPFL, and UMdatasets than the previous work. [ABSTRACT FROM AUTHOR]

Classification is one of the data mining problems receiving enormous attention in the database community. Although artificial neural networks (ANNs) have been successfully applied in a wide range of machine learning applications, they are however often regarded as black boxes, i.e., their predictions cannot be explained. To enhance the explanation of ANNs, a novel algorithm to extract symbolic rules from ANNs has been proposed in this paper. ANN methods have not been effectively utilized for data mining tasks because how the classifications were made is not explicitly stated as symbolic rules that are suitable for verification or interpretation by human experts. With the proposed approach, concise symbolic rules with high accuracy, that are easily explainable, can be extracted from the trained ANNs. Extracted rules are comparable with other methods in terms of number of rules, average number of conditions for a rule, and the accuracy. The effectiveness of the proposed approach is clearly demonstrated by the experimental results on a set of benchmark data mining classification problems. [ABSTRACT FROM AUTHOR]

This paper presents a benchmark for peak detection algorithms employed in fiber Bragg grating spectrometric interrogation systems. The accuracy, precision, and computational performance of currently used algorithms and those of a new proposed artificial neural network algorithm are compared. Centroid and gaussian fitting algorithms are shown to have the highest precision but produce systematic errors that depend on the FBG refractive index modulation profile. The proposed neural network displays relatively good precision with reduced systematic errors and improved computational performance when compared to other networks. Additionally, suitable algorithms may be chosen with the general guidelines presented. [ABSTRACT FROM AUTHOR]

This paper provides a comparative study on the different techniques of classifying human leg motions that are performed using two low-cost uniaxial piezoelectric gyroscopes worn on the leg. A number of feature sets, extracted from the raw inertial sensor data in different ways, are used in the classification process. The classification techniques implemented and compared in this study are: Bayesian decision making (BDM), a rule-based algorithm (RBA) or decision tree, least-squares method (LSM), k-nearest neighbor algorithm (k-NN), dynamic time warping (DTW), support vector machines (SVM), and artificial neural networks (ANN). A performance comparison of these classification techniques is provided in terms of their correct differentiation rates, confusion matrices, computational cost, and training and storage requirements. Three different cross-validation techniques are employed to validate the classifiers. The results indicate that BDM, in general, results in the highest correct classification rate with relatively small computational cost. [ABSTRACT FROM AUTHOR]

This paper introduces a comparison of training algorithms of radial basis function (RBF) neural networks for classification purposes. RBF networks provide effective solutions in many science and engineering fields. They are especially popular in the pattern classification and signal processing areas. Several algorithms have been proposed for training RBF networks. The Artificial Bee Colony (ABC) algorithm is a new, very simple and robust population based optimization algorithm that is inspired by the intelligent behavior of honey bee swarms. The training performance of the ABC algorithm is compared with the Genetic algorithm, Kalman filtering algorithm and gradient descent algorithm. In the experiments, not only well known classification problems from the UCI repository such as the Iris, Wine and Glass datasets have been used, but also an experimental setup is designed and inertial sensor based terrain classification for autonomous ground vehicles was also achieved. Experimental results show that the use of the ABC algorithm results in better learning than those of others. [ABSTRACT FROM AUTHOR]

Millimeter-wave (MMW) imaging scanners can see through clothing to form a three-dimensional holographic image of the human body and suspicious objects, providing a harmless alternative for non-contacting searches in security check. Suspicious object detection in MMW images is challenging, since most of them are small, reflection-weak, shape, and reflection-diverse. Conventional detectors with artificial neural networks, like convolution neural network (CNN), usually take the problem of finding suspicious objects as an object recognition task, yielding difficulties in developing large-amount and complete sample sets of objects. In this paper, a new algorithm is developed using the human pose segmentation followed by the deep CNN detection. The algorithm is emphasized to learn the similarity with humans' body clutter applied to training corresponding CNNs after the image segmentation base of the pose estimation. Moreover, the suspicious object recognition in the MMW image is converted to a binary classification task. Instead of recognizing all sorts of suspicious objects, the CNN detector determines whether the body part images present the abnormal patterns containing suspicious objects. The proposed algorithm that is based on CNN with the pose segmentation has concise configuration, but optimal performance in the suspicious object detection. Extensive experiments confirm the effectiveness and superiority of the proposal. [ABSTRACT FROM AUTHOR]

The weighted K-nearest neighbor (WKNN) algorithm is a commonly used fingerprint positioning, the difficulty of which lies in how to optimize the value of K to obtain the minimum positioning error. In this paper, we propose an adaptive residual weighted K-nearest neighbor (ARWKNN) fingerprint positioning algorithm based on visible light communication. Firstly, the target matches the fingerprints according to the received signal strength indication (RSSI) vector. Secondly, K is a dynamic value according to the matched RSSI residual. Simulation results show the ARWKNN algorithm presents a reduced average positioning error when compared with random forest (81.82%), extreme learning machine (83.93%), artificial neural network (86.06%), grid-independent least square (60.15%), self-adaptive WKNN (43.84%), WKNN (47.81%), and KNN (73.36%). These results were obtained when the signal-to-noise ratio was set to 20 dB, and Manhattan distance was used in a two-dimensional (2-D) space. The ARWKNN algorithm based on Clark distance and minimum maximum distance metrics produces the minimum average positioning error in 2-D and 3-D, respectively. Compared with self-adaptive WKNN (SAWKNN), WKNN and KNN algorithms, the ARWKNN algorithm achieves a significant reduction in the average positioning error while maintaining similar algorithm complexity. [ABSTRACT FROM AUTHOR]

Source localization based on time of arrival (TOA) measurements in the presence of clock asynchronization and sensor position uncertainties is investigated in this paper. Different from the traditional numerical algorithms, a neural circuit named Lagrange programming neural network (LPNN) is employed to tackle the nonlinear and nonconvex constrained optimization problem of source localization. With the augmented term, two types of neural networks are developed from the original maximum likelihood functions based on the general framework provided by LPNN. The convergence and local stability of the proposed neural networks are analyzed in this paper. In addition, the Cramér-Rao lower bound is also derived as a benchmark in the presence of clock asynchronization and sensor position uncertainties. Simulation results verify the superior performance of the proposed LPNN over the traditional numerical algorithms and its robustness to resist the impact of a high level of measurement noise, clock asynchronization, as well as sensor position uncertainties. [ABSTRACT FROM AUTHOR]

Although correlation filter (CF)-based visual tracking algorithms have achieved appealing results, there are still some problems to be solved. When the target object goes through long-term occlusions or scale variation, the correlation model used in existing CF-based algorithms will inevitably learn some non-target information or partial-target information. In order to avoid model contamination and enhance the adaptability of model updating, we introduce the keypoints matching strategy and adjust the model learning rate dynamically according to the matching score. Moreover, the proposed approach extracts convolutional features from a deep convolutional neural network (DCNN) to accurately estimate the position and scale of the target. Experimental results demonstrate that the proposed tracker has achieved satisfactory performance in a wide range of challenging tracking scenarios. [ABSTRACT FROM AUTHOR]

External disturbance poses the primary threat to robot balance in dynamic environments. This paper provides a learning-based control architecture for quadrupedal self-balancing, which is adaptable to multiple unpredictable scenes of external continuous disturbance. Different from conventional methods which construct analytical models which explicitly reason the balancing process, our work utilized reinforcement learning and artificial neural network to avoid incomprehensible mathematical modeling. The control policy is composed of a neural network and a Tanh Gaussian policy, which implicitly establishes the fuzzy mapping from proprioceptive signals to action commands. During the training process, the maximum-entropy method (soft actor-critic algorithm) is employed to endow the policy with powerful exploration and generalization ability. The trained policy is validated in both simulations and realistic experiments with a customized quadruped robot. The results demonstrate that the policy can be easily transferred to the real world without elaborate configurations. Moreover, although this policy is trained in merely one specific vibration condition, it demonstrates robustness under conditions that were never encountered during training. [ABSTRACT FROM AUTHOR]

The paper presents the results of research on the hybrid industrial tomograph electrical impedance tomography (EIT) and ultrasonic tomography (UST) (EIT-UST), operating on the basis of electrical and ultrasonic data. The emphasis of the research was placed on the algorithmic domain. However, it should be emphasized that all hardware components of the hybrid tomograph, including electronics, sensors and transducers, have been designed and mostly made in the Netrix S.A. laboratory. The test object was a tank filled with water with several dozen percent concentration. As part of the study, the original multiple neural networks system was trained, the characteristic feature of which is the generation of each of the individual pixels of the tomographic image, using an independent artificial neural network (ANN), with the input vector for all ANNs being the same. Despite the same measurement vector, each of the ANNs generates its own independent output value for a given tomogram pixel, because, during training, the networks get their respective weights and biases. During the tests, the results of three tomographic methods were compared: EIT, UST and EIT-UST hybrid. The results confirm that the use of heterogeneous tomographic systems (hybrids) increases the reliability of reconstruction in various measuring cases, which is used to solve quality problems in managing production processes. [ABSTRACT FROM AUTHOR]

Transfer learning can enhance classification performance of a target domain with insufficient training data by utilizing knowledge relating to the target domain from source domain. Nowadays, it is common to see two or more source domains available for knowledge transfer, which can improve performance of learning tasks in the target domain. However, the classification performance of the target domain decreases due to mismatching of probability distribution. Recent studies have shown that deep learning can build deep structures by extracting more effective features to resist the mismatching. In this paper, we propose a new multi-source deep transfer neural network algorithm, MultiDTNN, based on convolutional neural network and multi-source transfer learning. In MultiDTNN, joint probability distribution adaptation (JPDA) is used for reducing the mismatching between source and target domains to enhance features transferability of the source domain in deep neural networks. Then, the convolutional neural network is trained by utilizing the datasets of each source and target domain to obtain a set of classifiers. Finally, the designed selection strategy selects classifier with the smallest classification error on the target domain from the set to assemble the MultiDTNN framework. The effectiveness of the proposed MultiDTNN is verified by comparing it with other state-of-the-art deep transfer learning on three datasets. [ABSTRACT FROM AUTHOR]

Traffic sign detection systems provide important road control information for unmanned driving systems or auxiliary driving. In this paper, the Faster region with a convolutional neural network (R-CNN) for traffic sign detection in real traffic situations has been systematically improved. First, a first step region proposal algorithm based on simplified Gabor wavelets (SGWs) and maximally stable extremal regions (MSERs) is proposed. In this way, the region proposal a priori information is obtained and will be used for improving the Faster R-CNN. This part of our method is named as the highly possible regions proposal network (HP-RPN). Second, in order to solve the problem that the Faster R-CNN cannot effectively detect small targets, a method that combines the features of the third, fourth, and fifth layers of VGG16 to enrich the features of small targets is proposed. Third, the secondary region of interest method to enhance the feature of detection objects and improve the classification capability of the Faster R-CNN is proposed. Finally, a method of merging the German traffic sign detection benchmark (GTSDB) and Chinese traffic sign dataset (CTSD) databases into one larger database to increase the number of database samples is proposed. Experimental results show that our method improves the detection performance, especially for small targets. [ABSTRACT FROM AUTHOR]

A dynamic time warping (DTW) algorithm has been suggested for the purpose of devising a motion-sensitive microelectronic system for the realization of remote motion abnormality detection. In combination with an inertial measurement unit (IMU), the algorithm is potentially applicable for remotely monitoring patients who are at risk of certain exceptional motions. The fixed interval signal sampling mechanism has normally been adopted when devising motion detection systems; however, dynamically capturing the particular motion patterns from the IMU motion sensor can be difficult. To this end, the DTW algorithm, as a kind of nonlinear pattern-matching approach, is able to optimally align motion signal sequences tending towards time-varying or speed-varying expressions, which is especially suitable to capturing exceptional motions. Thus, this paper evaluated this kind of abnormality detection using the proposed DTW algorithm on the basis of its theoretical fundamentals to significantly enhance the viability of the methodology. To validate the methodological viability, an artificial neural network (ANN) framework was intentionally introduced for performance comparison. By incorporating two types of designated preprocessors, i.e., a DFT interpolation preprocessor and a convolutional preprocessor, to equalize the unequal lengths of the matching sequences, two kinds of ANN frameworks were enumerated to compare the potential applicability. The comparison eventually confirmed that the direct template-matching DTW is excellent in practical application for the detection of time-varying or speed-varying abnormality, and reliably captures the consensus exceptions. [ABSTRACT FROM AUTHOR]

Currently, effective crowd management based on the information provided by crowd monitoring systems is difficult as this information comes in at the moment adverse crowd movements are already occurring. Up to this moment, very little forecasting techniques have been developed that predict crowd flows a longer time period ahead. Moreover, most contemporary state estimation methods apply demanding pre-processing steps, such as map-matching. The objective of this paper is to design, train and benchmark a data-driven procedure to forecast crowd movements, which can in real-time predict crowd movement. This procedure entails two steps. The first step comprises of a cell sequence derivation method that allows the representation of spatially continuous GPS traces in terms of discrete cell sequences. The second step entails the training of a Recursive Neural Network (RNN) with a Gated Recurrent Unit (GRU) and six benchmark models to forecast the next location of pedestrians. The RNN-GRU is found to outperform the other tested models. Some additional tests of the ability of the RNN-GRU to forecast illustrate that the RNN-GRU preserves its predictive power when a limited amount of data is used from the first few hours of a multi-day event and temporal information is incorporated in the cell sequences. [ABSTRACT FROM AUTHOR]

A binocular vision system is a common perception component of an intelligent vehicle. Benefiting from the biomimetic structure, the system is simple and effective. Which are extremely snesitive on external factors, especially missing vision signals. In this paper, a virtual view-generation algorithm based on generative adversarial networks (GAN) is proposed to enhance the robustness of binocular vision systems. The proposed model consists of two parts: generative network and discriminator network. To improve the quality of a virtual view, a generative network structure based on 3D convolutional neural networks (3D-CNN) and attentive mechanisms is introduced to extract the time-series features from image sequences. To avoid gradient vanish during training, the dense block structure is utilized to improve the discriminator network. Meanwhile, three kinds of image features, including image edge, depth map and optical flow are extracted to constrain the supervised training of model. The final results on KITTI and Cityscapes datasets demonstrate that our algorithm outperforms conventional methods, and the missing vision signal can be replaced by a generated virtual view. [ABSTRACT FROM AUTHOR]

At present, realizing high-quality automatic welding through online monitoring is a research focus in engineering applications. In this paper, a CNN–LSTM algorithm is proposed, which combines the advantages of convolutional neural networks (CNNs) and long short-term memory networks (LSTMs). The CNN–LSTM algorithm establishes a shallow CNN to extract the primary features of the molten pool image. Then the feature tensor extracted by the CNN is transformed into the feature matrix. Finally, the rows of the feature matrix are fed into the LSTM network for feature fusion. This process realizes the implicit mapping from molten pool images to welding defects. The test results on the self-made molten pool image dataset show that CNN contributes to the overall feasibility of the CNN–LSTM algorithm and LSTM network is the most superior in the feature hybrid stage. The algorithm converges at 300 epochs and the accuracy of defects detection in CO2 welding molten pool is 94%. The processing time of a single image is 0.067 ms, which fully meets the real-time monitoring requirement based on molten pool image. The experimental results on the MNIST and FashionMNIST datasets show that the algorithm is universal and can be used for similar image recognition and classification tasks. [ABSTRACT FROM AUTHOR]

As an alternative to high-cost shoe insole pressure sensors that measure the insole pressure distribution and calculate the center of pressure (CoP), researchers developed a foot sensor with FSR sensors on the bottom of the insole. However, the calculations for the center of pressure and ground reaction force (GRF) were not sufficiently accurate because of the fundamental limitations, fixed coordinates and narrow sensing areas, which cannot cover the whole insole. To address these issues, in this paper, we describe an algorithm of virtual forces and corresponding coordinates with an artificial neural network (ANN) for low-cost flexible insole pressure measurement sensors. The proposed algorithm estimates the magnitude of the GRF and the location of the foot plantar CoP. To compose the algorithm, we divided the insole area into six areas and created six virtual forces and the corresponding coordinates. We used the ANN algorithm with the input of magnitudes of FSR sensors, 1st and 2nd derivatives of them to estimate the virtual forces and coordinates. Eight healthy males were selected for data acquisition. They performed an experiment composed of the following motions: standing with weight shifting, walking with 1 km/h and 2 km/h, squatting and getting up from a sitting position to a standing position. The ANN for estimating virtual forces and corresponding coordinates was fitted according to those data, converted to c script, and downloaded to a microcontroller for validation experiments in real time. The results showed an average RMSE the whole experiment of 31.154 N for GRF estimation and 8.07 mm for CoP calibration. The correlation coefficients of the algorithm were 0.94 for GRF, 0.92 and 0.76 for the X and Y coordinate respectively. [ABSTRACT FROM AUTHOR]

It is crucial for robots to autonomously steer in complex environments safely without colliding with any obstacles. Compared to conventional methods, deep reinforcement learning-based methods are able to learn from past experiences automatically and enhance the generalization capability to cope with unseen circumstances. Therefore, we propose an end-to-end deep reinforcement learning algorithm in this paper to improve the performance of autonomous steering in complex environments. By embedding a branching noisy dueling architecture, the proposed model is capable of deriving steering commands directly from raw depth images with high efficiency. Specifically, our learning-based approach extracts the feature representation from depth inputs through convolutional neural networks and maps it to both linear and angular velocity commands simultaneously through different streams of the network. Moreover, the training framework is also meticulously designed to improve the learning efficiency and effectiveness. It is worth noting that the developed system is readily transferable from virtual training scenarios to real-world deployment without any fine-tuning by utilizing depth images. The proposed method is evaluated and compared with a series of baseline methods in various virtual environments. Experimental results demonstrate the superiority of the proposed model in terms of average reward, learning efficiency, success rate as well as computational time. Moreover, a variety of real-world experiments are also conducted which reveal the high adaptability of our model to both static and dynamic obstacle-cluttered environments. [ABSTRACT FROM AUTHOR]

Throughout the last decade, a whole new generation of powered transtibial prostheses and exoskeletons has been developed. However, these technologies are limited by a gait phase detection which controls the wearable device as a function of the activities of the wearer. Consequently, gait phase detection is considered to be of great importance, as achieving high detection accuracy will produce a more precise, stable, and safe rehabilitation device. In this paper, we propose a novel gait percent detection algorithm that can predict a full gait cycle discretised within a 1% interval. We called this algorithm an exponentially delayed fully connected neural network (ED-FNN). A dataset was obtained from seven healthy subjects that performed daily walking activities on the flat ground and a 15-degree slope. The signals were taken from only one inertial measurement unit (IMU) attached to the lower shank. The dataset was divided into training and validation datasets for every subject, and the mean square error (MSE) error between the model prediction and the real percentage of the gait was computed. An average MSE of 0.00522 was obtained for every subject in both training and validation sets, and an average MSE of 0.006 for the training set and 0.0116 for the validation set was obtained when combining all subjects’ signals together. Although our experiments were conducted in an offline setting, due to the forecasting capabilities of the ED-FNN, our system provides an opportunity to eliminate detection delays for real-time applications. [ABSTRACT FROM AUTHOR]

Recently, short-term traffic prediction under conditions with corrupted or missing data has become a popular topic. Since a road section has predictive power regarding the adjacent roads at a specific location, this paper proposes a novel hybrid convolutional long short-term memory neural network model based on critical road sections (CRS-ConvLSTM NN) to predict the traffic evolution of global networks. The critical road sections that have the most powerful impact on the subnetwork are identified by a spatiotemporal correlation algorithm. Subsequently, the traffic speed of the critical road sections is used as the input to the ConvLSTM to predict the future traffic states of the entire network. The experimental results from a Beijing traffic network indicate that the CRS-ConvLSTM outperforms prevailing deep learning (DL) approaches for cases that consider critical road sections and the results validate the capability and generalizability of the model when predicting with different numbers of critical road sections. [ABSTRACT FROM AUTHOR]

Abstract: We propose a method for building a simple electronic nose based on commercially available sensors used to sniff in the market and identify spoiled/contaminated meat stocked for sale in butcher shops. Using a metal oxide semiconductor-based electronic nose, we measured the smell signature from two of the most common meat foods (beef and fish) stored at room temperature. Food samples were divided into two groups: fresh beef with decayed fish and fresh fish with decayed beef. The prime objective was to identify the decayed item using the developed electronic nose. Additionally, we tested the electronic nose using three pattern classification algorithms (artificial neural network, support vector machine and k-nearest neighbor), and compared them based on accuracy, sensitivity, and specificity. The results demonstrate that the k-nearest neighbor algorithm has the highest accuracy. [ABSTRACT FROM AUTHOR]

One of the fundamental requirements for an artificial hand to successfully grasp and manipulate an object is to be able to distinguish different objects' shapes and, more specifically, the objects' surface curvatures. In this study, we investigate the possibility of enhancing the curvature detection of embedded tactile sensors by proposing a ridged fingertip structure, simulating human fingerprints. In addition, a curvature detection approach based on machine learning methods is proposed to provide the embedded sensors with the ability to discriminate the surface curvature of different objects. For this purpose, a set of experiments were carried out to collect tactile signals from a 2 x 2 tactile sensor array, then the signals were processed and used for learning algorithms. To achieve the best possible performance for our machine learning approach, three different learning algorithms of Naîve Bayes (NB), Artificial Neural Networks (ANN), and Support Vector Machines (SVM) were implemented and compared for various parameters. Finally, the most accurate method was selected to evaluate the proposed skin structure in recognition of three different curvatures. The results showed an accuracy rate of 97.5% in surface curvature discrimination. [ABSTRACT FROM AUTHOR]