A multi-robot behavior adaptation mechanism that adapts to human intention is proposed for human-robot interaction (HRI), where information-driven fuzzy friend-Q learning (IDFFQ) is used to generate an optimal behavior-selection policy, and intention is understood mainly based on human emotions. This mechanism aims to endow robots with human-oriented interaction capabilities to understand and adapt their behaviors to human intentions. It also decreases the response time (RT) of robots by embedding the human identification information such as religion for behavior selection, and increases the satisfaction of humans by considering their deep-level information, including intention and emotion, so as to make interactions run smoothly. Experiments is performed in a scenario of drinking at a bar. Results show that the learning steps of the proposal is 51 steps less than that of the fuzzy production rule based friend-Q learning (FPRFQ), and the robots’ RT is about 25% of the time consumed by FPRFQ. Additionally, emotion recognition and intention understanding achieved an accuracy of 80.36% and 85.71%, respectively. Moreover, a subjective evaluation of customers through a questionnaire obtains a reaction of “satisfied.” Based on these preliminary experiments, the proposal is being extended to service robots for behavior adaptation to customers’ intention to drink at a bar.

In this paper, a switched Kalmanfilter (KF) is used to predict the status of feature points leaving the field of view (FOV), which is one of the most common constraints in FOV. By using the prediction of status to compensate for the real state of feature points, nonholonomic robots conduct visual servoing tasks efficiently. Results of simulation and experiments verify the effectiveness of the proposed approach.

Random deployment and inadequate numbers of sensor nodes may cause low coverage, so we propose a new priority coverage control strategy based on the distributed cooperation of mobile sensors. Sensor nodes are distributed randomly around the region of interest (ROI) and searched for independently. When nodes are found, an unbreakable group is formed under repulsion, attraction and speed consistency control, then searching is begun cooperatively. When some node finds a ROI, it guides the other nodes in the group following it to the ROI. While in the ROI, nodes choose the most important position within the sensing range, then move toward it independently while avoiding collision, eventually, reaching the most important area of the ROI. Under the premise of satisfying key area coverage, sensor nodes are adjusted based on the degree of coverage, maximizing coverage. Simulation results show that the proposed method quickly improves the coverage rate and achieves priority coverage of key areas strongly robustly without being adversely affected by sudden damage to nodes. Applications include coverage with limited amounts of nodes in unknown environments.

To address the inherent energy constraint in cognitive radio sensor networks, a novel joint optimization method of spectrum sensing and data transmission for energy efficiency is investigated in this paper. To begin with, a cooperative spectrum sensing scheme based on dynamic censoring is employed to shorten sensing time and save unnecessary spectrum sensing energy. Then to jointly optimize the energy efficiency, the distortion constrained probabilistic transmission scheme is utilized. Afterwards the sensing threshold solving issue can be formulated as a nonlinear minmax optimization problem with the detection probability and false alarm probability constraints. Solving by the Matlab software with the free OPTI toolbox, simulation results demonstrate that significant energy can be saved via the the proposed joint optimization method in various mobile cloud scenarios.

In this paper, we propose a FPGA based intelligent desk lamp system with a camera, two servo motors and LEDs. The proposed lamp system can automatically trace a moving book with a red bookmark by pan and tilt control to provide sufficient illumination. Thus, the user can read book with enough light for eye caring. The architecture of the system includes a vision system and a servo-motor control system. The image processing module inside the system can obtain the book’s position data. According to the book’s position data, two servo motors automatically adjust the pan and tilt angles of the desk lamp. The adjustable ranges for pan and tilt angles are 40° and 30°, respectively. Additionally, an App is designed and allows the user to remotely control the lamp by using a smartphone. All the processing modules are implemented on Altera DE0-Nano kit with Cyclone® IV EP4CE22F17C6N FPGA with Verilog HDL.

The rapid progress in and the expanding complexity of information and technology systems have made data analysis increasingly relevant. Data having a variety of elements are complex, and making very difficult to evaluate a state of a model from observed data generated probabilistically by the model. To evaluate these hidden states, we propose Spherical-Self Organizing Map (S-SOM) with a Hidden Markov Model (HMM) that infers such hidden states.

A fuzzy association rule mining based method is proposed for myocardial ischemia diagnosis on ECG signals. The proposal provides interpretable and understandable information to doctors as an assistant reference, while rule mining on fuzzy itemsets guarantees that the feature segmentation before rule extraction is feasible and effective. A set of fuzzy association rules is mined through experiments on data from the European ST-T Database, and classification results of myocardial ischemia and normal heartbeats on the test dataset using the extracted rules obtained values of 83.4%, 80.7%, and 81.4% for sensitivity, specificity, and accuracy, respectively. The proposed method aims to become a helpful tool to accelerate the diagnosis of myocardial ischemia on ECG signal, and to be expanded to other heart disease diagnosis areas such as hypertensive heart disease and arrhythmia.

A hierarchical force-directed graph drawing is proposed for the analysis of a neural network structure that expresses the relationship between multitask and processes in neural networks represented as neuron clusters. The process revealed by our proposal indicates the neurons that are related to each task and the number of neurons or learning epochs that are sufficient. Our proposal is evaluated by visualizing neural networks learned on the Mixed National Institute of Standards and Technology (MNIST) database of handwritten digits, and the results show that inactive neurons, namely those that do not have a close relationship with any tasks, are located on the periphery part of the visualized network, and that cutting half of the training data on one specific task (out of ten) causes a 15% increase in the variance of neurons in clusters that react to the specific task compared to the reaction to all tasks. The proposal aims to be developed in order to support the design process of neural networks that consider multitasking of different categories, for example, one neural network for both the vision and motion system of a robot.

The complexity of coking production and the correlations among the three major processes involved make it difficult to study and apply effective methods in practice. We have designed a hierarchical simulation platform for coking production in coke ovens for experiments and the validation of the methods used. To handle problems in processing and obtain the comprehensive production targets, the simulation platform provides reliable, easy-to-use conditions for coking production research, which has the functions of simulating processes, examining methods for experiments, monitoring production status and coordinating optimization. To implement these functions, the platform has a three-layer structure and flexible communication interfaces. Results of experiments have demonstrated the simulation platform’s effectiveness and feasibility.

In this paper, a robust H_∞ damping controller of multi-FACTS device for a power system is developed with considering the time delay of the remote feedback signals transmitted by wide-area measurement systems (WAMS). A free-weighting matrices method based on linear control design approach is presented to design the robust H_∞ damping controller to improve the dynamical performance of power systems. Firstly, the linearized reduced-order plant model is established, which efficiently considers the signal’s time delay and disturbance. Then, the design of multi-FACTS robust H_∞ damping controller is formulated as the standard control problem on delay-dependent state-feedback robust control based on the robust control theory. Finally, the simulation tests are carried out on the 2-area 4-machine power systems. Satisfactory test results verify the correctness of the proposed damping controller.

An event-driven on-vehicle intelligent human-computer interface has been proposed to solve the problem of complex on-vehicle human-computer interaction. After need analysis of human-computer interaction under the vehicle platform, the framework of intelligent human-computer interface is established, various modules and workflows in the system are designed, and the reasoning feature based on fuzzy cognitive map (FCM) is implemented. The on-vehicle intelligent human-computer interface could help users to complete the interactive operation which is unrelated to the driving operations. Furthermore, the system could analyze the whole information and predict the information required by the user. At last, it could display the information on the interface. So, the on-vehicle intelligent human-computer interface could not only meet the user’s demand for secondary interactive tasks, but also could ensure the driving performance and safety.

To locate Informative Bright Region (IBR) in which visual information is missing owing to limited dynamic range of image sensor, an algorithm is proposed that combines the geometric properties of visual cues into a confidence map. The location of an IBR in a road tunnel scene is estimated in real-time under the condition in which most of the vision information inside the IBR is lost. The algorithm is evaluated by comparing the estimated location of IBR with that annotated by multiple human observers in a self-built tunnel scene video dataset recorded by a car-mounted camera, and the algorithm achieves a running time of 10 ms for each frame. The algorithm aims to provide control timing of imaging sensor on a low-cost platform such as a vehicle driving recorder to enhance the visual contents captured in over-exposed regions.

Polarization degree measurement requires that the three input axes be orthogonal. This prevents this method from being used if input angles are nonorthogonal. This paper proposes a method for optimal window median filtering preprocessing to measure the polarization degree of nonorthogonal angles. A 7×7 window is selected to filter out most noise and identify an object from polarization degree images. Compared to conventional algorithms, the proposed method selects three nonorthogonal angles, which is applied to measuring the polarization degree of dynamic objects.

This paper aims at building a Computational Fluid Dynamics (CFD) model which can describe the gas flow three dimensions (3D) distribution in blast furnace (BF) throat. Firstly, the boundary conditions are obtained by rebuilding central gas flow shape in BF based on computer graphics. Secondly, the CFD model is built based on turbulent model by analyzing the features of gas flow. Finally, a method which can get the numerical solutions of the model is proposed by using CFD software ANSYS/FLUENT. The proposed model can reflect the changes of the gas flow distribution, and can help to guide the operation of furnace burdening and to ensure the BF stable and smooth production.

This paper focuses on ground-moving target tracking of an unmanned aerial vehicle (UAV) in the presence of static obstacles and moving threat sources. Due to a UAV is restricted by airspace restrictions and measurement limitations during flight, we derive a dynamic path planning strategy by generating guidance vector filed combined Lyapunov vector field with collision avoidance potential function to track target in standoff distance loitering pattern, and resolved collision avoidance, simultaneously. This method relies only on the current information of the UAV and target, and generates a single-step route plan in realtime. Its performance is simple, efficient, and fast and have low computational complexity. The results of numerical simulation verify the effectiveness of the tracking and collision avoidance process of the UAV.

A fuzzy set representation method of Kansei Texture is proposed to express individual difference of Kansei Texture feelings for the purpose of online shopping. The method provides buyers with criteria whether a request to send samples is necessary according to the variance degree of individual differences, and it also offers sellers with information regarding the possibility of returned goods in case of significant individual differences with regard to expensive prices. The correlation coefficient of the degree of individual difference and sample demand is 0.78 (P<0.05, t-test), i.e., a directly proportional relationship is observed between the two degrees. There is a tendency for expensive goods, e.g., those with price greater than $50, to be returned in the case of a large individual difference degree, i.e., the individual difference degree of Kansei Texture with price information provides a useful strategy for estimating the possibility of returned goods. Moreover, the relationship between stress and individual difference is also shown. Further validity verification is planned in order to realize practical applications in the real market.

This paper presents a new method for controlling the motion of a wheeled inverted pendulum (WIP) based on the equivalent-input-disturbance (EID) approach. Coordinate transformation first transforms the WIP into a simple nonlinear system divided into linear and nonlinear parts. The nonlinear part is then treated as a state-and-input-dependent disturbance, and the EID approach is used to estimate and compensate it. Simulation results of an NXTway-GS demonstrate the validity of the method.

In this article, a guidance problem for cooperative salvo attack of multiple missiles against a single stationary target is investigated. The proposed guidance law combines the well-known PNG law and cooperative acceleration command, which is based on the feedback of state error between the current missile and the mean value of participant missiles. The state variable in this paper is used as the approximate calculation of time-to-go. The cooperative acceleration command is designed to adjust the flight path and impact time, which leads the multi-missiles to hit the common target simultaneously. During the engagement, the velocities of missiles are not changed and presetting impact time is not needed. Simulation results show the effectiveness of the proposed guidance law.

There is a growing trend at universities to switch from conventional teaching methods, which focus on knowledge transfer, to methods based on the concept of active learning. Many such methods have been devised and tested to show the validity of this concept. In this study, a project was designed and implemented that teaches some simple principles of aeronautics by having students construct and fly a remote-controlled (RC) model airplane. The goals are to motivate students to study mechatronics and to foster teamwork and communication. This paper explains the project. Its effectiveness was demonstrated in three trials with three groups of students.

Deterministic echo state network (ESN) models integrated with particle swarm optimization (PSO) are proposed to improve the accuracy and efficiency of stock price forecasting. ESNs have been successfully applied to financial time series forecasting because of their efficient and powerful computational ability in approximating nonlinear dynamical systems. However, reservoir construction in standard ESNs is primarily driven by a series of randomized model-building stages, because of which both researchers and practitioners have to rely on a series of trials and errors to determine parameters. An ESN with a deterministically constructed reservoir is comparable in performance to a standard ESN and has minimal complexity as well as potential for optimizations with regard to ESN parameters. In this paper, forecasting performances of the proposed PSO-DESN models are compared with those of standard ESNs for stock price prediction on the benchmark dataset of S&P 500. The comparison results demonstrate that the proposed PSO-DESNs exhibit better performance in stock price forecasting in terms of both accuracy and efficiency, thereby verifying the potential of PSO-DESNs for financial predictions.

The method of extracting still corresponding points proposed in this paper uses a moving monocular camera connected to a 6-axis motion sensor. It classifies corresponding points between two consecutive frames containing still/moving objects and chooses corresponding points appropriate for 3D measurement. Experiments are done extracting still corresponding points with 2 scenes from original computer graphics images. Results for scene 1 show that accuracy is 0.98, precision 0.96, and recall 1.00. Robustness against sensor noise is confirmed. Extraction experiment results with real scenes show that accuracy is 0.86, precision 0.88, and recall 0.94. We plan to include the proposed method in 3D measurement with real images containing still/moving objects and to apply it to obstacles avoidance for vehicles and to mobile robot vision systems.