Skip to main content

Über dieses Buch

This book presents selected research papers from the 2015 Chinese Intelligent Systems Conference (CISC’15), held in Yangzhou, China. The topics covered include multi-agent systems, evolutionary computation, artificial intelligence, complex systems, computation intelligence and soft computing, intelligent control, advanced control technology, robotics and applications, intelligent information processing, iterative learning control, and machine learning. Engineers and researchers from academia, industry and the government can gain valuable insights into solutions combining ideas from multiple disciplines in the field of intelligent systems.



Chapter 1. Observer Design for Discrete-Time Switched Lipschitz Nonlinear Singular Systems with Time Delays and Unknown Inputs

In this paper, the state estimation problem for a class of discrete-time switched nonlinear singular systems simultaneously subject to Lipschitz constraints, state delays, unknown inputs (UIs), and arbitrary switching sequences is considered. A mode-dependent observer is constructed and, based on the idea of exact state and UI decoupling, sufficient conditions for the existence of the proposed observer are given in terms of linear matrix inequalities. By defining a decay-rate-dependent switched Lyapunov function, the convergence rate of the state estimation error is proved to be exponential.

Jinxing Lin

Chapter 2. Anti-disturbance Control for Nonlinear Systems with Mismatched Disturbances Based on Disturbance Observer

An anti-disturbance control scheme based on disturbance observer is proposed for a class of nonlinear system with mismatched disturbances. With such control approach, the disturbances can be rejected and the semi-global uniformly ultimate bounded (SGUUB) stability of the closed-loop system can be achieved. Finally, simulations for a numerical example are given to demonstrate the feasibility and effectiveness of the proposed scheme.

Lingyan Zhang, Xinjiang Wei

Chapter 3. Distributed Optimization for Continuous-Time Multi-Agent Systems with External Disturbance and Discrete-Time Communication

In this paper, distributed optimization problem for continuous-time multi-agent systems with external disturbance and discrete-time communication is considered. A distributed algorithm is developed to achieve the exact optimal solution by completely rejecting the disturbance. An upper bound for the discrete-time communication period is obtained to ensure the exponential convergence for this optimization problem. Finally, a numerical example is given to illustrate the effectiveness of the proposed algorithm.

Zhenhua Deng, Yiguang Hong

Chapter 4. Camera Calibration Implementation Based on Zhang Zhengyou Plane Method

Camera calibration is a crucial step in computer vision, the main determinant of the visual measurement effect, laying the basis for three-dimensional reconstruction. In order to know calibration’s precision exactly, pinhole model and relations of the four coordinates are used, camera’s internal and external parameter matrices can be solved by Zhang Zhengyou plane calibration method, camera’s distortion coefficients are then easily solved, further considered radial distortion and tangential distortion. The paper establishes a simple but clear error assessment system to evaluate the accuracy of the results and compares them with MATLAB toolbox. The experiment demonstrates that the method has high accuracy, establishing a foundation for seeking depth by binocular stereo vision.

Pingping Lu, Qing Liu, Jianming Guo

Chapter 5. Online Identification and Robust Adaptive Control for Discrete Hysteresis Preisach Model

By using the concept of online goal adaptation, we develop a new online identification and a robust adaptive control for hysteresis system which is described by discrete Preisach model. The proposed identification requires triangle matrices to simply the calculation for Preisach model. In addition, a robust adaptive control is adopted without inverse Preisach model based on the weighting factor recursive least square (WFRLS) method. A Lyapunov function candidate guarantees the stability of this hysteresis system. Finally, simulations perform on a typical system which clarify the validity of the proposed approach.

Xuehui Gao, Xuemei Ren, Chengyuan Zhang, Changsheng Zhu

Chapter 6. Switching Control for Multi-motor Driving Servo System with Uncertain Parameters

This paper presents a novel switching control for multi-motor driving servo systems with uncertain parameters, such that it is sufficient to achieve speed synchronization among motors and load tracking. In order to solve the problem of complex coupling relationship between synchronization and tracking, a switching plane (SP) is introduced to implement switching between speed synchronization control and load tracking control. In design of synchronization control, the adaptive scheme based on speed negative feedback is proposed to achieve synchronization errors convergence in finite time. Due to the properties of robustness, the adaptive robust algorithm is utilized to attain load tracking accurately, where dual-observer and adaptive laws are presented to estimate unknown friction state and uncertain parameters. Comparative simulation results are included to verify the reliability and effectiveness.

Wei Zhao, Xuemei Ren

Chapter 7. Method of Fault Diagnosis Based on SVDD-SVM Classifier

Aiming at the problem of incomplete fault data samples, a fault diagnosis method based on Support vector data description and Support vector machine (SVDD-SVM) is presented. First, the data description model is build based on the normal data samples and known fault data samples, and SVM model is built based on known fault data samples. Then the test data samples are tackled by the data description model to reject or accept. The specific categories of accepted samples are diagnosed by the SVM model and the rejected samples are unknown fault types. Tests show that this method can efficiently solve the fault diagnosis problem of incomplete fault samples.

Feng Lv, Hua Li, Hao Sun, Xiang Li, Zeyu Zhang

Chapter 8. Second-Order Sliding Mode Control for BUCK Converters

This paper presents a second-order sliding mode (SOSM) control method for DC-DC Buck converters. First of all, by taking a subtraction between the output voltage and the desired voltage, the sliding mode variable with freedom of degree two can be constructed. Secondly, by using adding a power integrator method, the second-order sliding mode controller can be developed for the Buck converter. Under the proposed controller, it can be shown that the output voltage will track the desired voltage in a finite time. Finally, the theoretical considerations have been verified by simulations.

Jiadian Wang, Shihong Ding

Chapter 9. Open-World Planning Algorithm Based on Logic

Existence of certain objects or fluent is often unknown before planning in many domains. Plan synthesis in such open worlds is challenging since we have to take various scenarios into account before searching plans. One way to do this is to employ sensors to observe unknown objects or fluent, assuming the sensors are capable of correctly capturing all information needed for planning. We aims at solving automated planning problem in open world, and call goal state with variables as query-goal. Instead of using sensors, we proposed a novel algorithm PQG (Planner with query-goal) to solve automated planning problem with query-goal, by encoding the planning problem into a planning logic problem, and then applying planning inference method to solve it. Finally, inference result is transformed into a planning solution. We empirically exhibit that our approach is effective in several planning domains.

Jie Gao, Ya-song Liu, Rui Bian

Chapter 10. Apple Nighttime Images Enhancement Algorithm for Harvesting Robot

In order to enhance the applicability and efficiency of harvesting robot to ensure that people can timely pick ripe fruit, the robot need to have an ability of continuous recognition and harvest at night. For some disadvantages of night vision images, Retinex algorithm for image enhancement based on bilateral filter is presented. Bilateral filter which has a function of edge preservation is adopted to improve the smooth, evaluate the illumination and remove unfavorable illumination effects from the original image. Then the reflectance of the image from above that contains just the characteristics of the object itself can be obtained. Finally, apple nighttime image enhancement is implemented. The experimental results show that the above method can more accurately evaluate the illumination of high-contrast edge regions, to suppress noise, enhance image contrast and improve overall visual effects of the image.

Xingqin Lv, Bo Xu, Wei Ji, Gang Tong, Dean Zhao

Chapter 11. Research on Grasping Planning for Apple Picking Robot’s End-Effector

Aiming at the lack of all-purpose and effective planning study for apple picking robot’s end-effector during the grasping process, which has caused great inconvenient in the accuracy of fruit picking process and design of end-effector. This paper studies the contact process of three-finger end-effector with apples. Taking contact of apples with fingers as point contact of friction between hard objects, and based on the fact that the contact force is decomposed into the orthogonal operating force component and internal force component, the regulation of internal force on contact stability is discussed. Stability would be attributed to the existence of the internal force of concurrent polygon and the position of the internal force concurrent node within the concurrent polygon. Regarding the circle center of concurrent polygon of the maximum inscribed circle as the intersection of three internal force action lines, we get the size of each internal force, and calculate the internal force meeting the friction cone constraints to avoid complex operation such as matrix operation. Eventually, a numerical example shows the feasibility of the method.

Feiyu Liu, Wei Ji, Wei Tang, Bo Xu, Dean Zhao

Chapter 12. Extended State Observer Based Sliding Mode Control for Mechanical Servo System with Friction Compensation

This paper proposes a tracking control method based on the extended state observer for the nonlinear mechanical servo system with friction compensation. The friction nonlinearity is described by a continuously differentiable LuGre model and compensated by using neural network (NN). Then, an extended state observer (ESO) is employed to estimate the system states and uncertainties including friction compensation error. A sliding mode control (SMC) scheme is developed based on ESO estimation to guarantee the convergence of the tracking error. Comparative simulations are conducted to show the superior performance of the proposed method.

Chenhang Li, Qiang Chen

Chapter 13. Local Linear Discriminant Analysis Using $$ \ell_{2} $$ ℓ 2 -Graph

A recently proposed method, called Local Fisher Linear Discriminant Analysis (LLDA), the experiment showed that compared with the traditional Fisher Linear Discriminant Analysis, it has a better result. However, it uses Euclidean distance selecting nearest neighbor samples which has some flaws in the way, such as the robustness is not good and not sparse, and so on. The paper presents an improved approach, called Local Linear Discriminant Analysis Using $$ \ell_{2} $$ℓ2-Graph (L2G_LLDA). It remains reconstructed coefficient of samples to select the nearest samples, which enhances the robustness of the algorithms and makes it sparse. The extensive experimental results over several standard face databases have demonstrated the effectiveness of the proposed algorithm

Ya Gu, Caikou Chen, Yu Wang, Rong Wang

Chapter 14. A New Detecting and Tracking Method in Driver Fatigue Detection

In order to reduce the traffic accidents caused by driver fatigue, this paper extends the TLD tracking algorithm to the case of detecting and tracking driver’s face and local areas which is based on the machine vision. We address problems of detecting driver’s face and local areas, which contain the driver’s fatigue information, via adaboost cascade classifier based on haar-like features and track the target areas by the TLD method. The main results are given in terms of the accuracy of detecting driver’s face and local areas. Lay a foundation for the driver fatigue detection.

Yang Yu, Xiaobin Li, Haiyan Sun

Chapter 15. A Multi-objective Dual-Resource Shop Scheduling Model Considering the Differences Between Operational Efficiency

Multiple resource constraints exist widely in the actual job shop under the complicate manufacturing environment, so a multi-objective dual-resource shop scheduling model considering the differences between operational efficiency is established, which contains two resource constraints: the machines and workers. An improved SPEA2 is proposed for the model which achieves three goals: the shortest completion time, the lowest cost and the minimum total tardiness. Since the evolutionary operation in the traditional algorithm for single-resource model has not been able to meet requirements of the model in this paper, the chromosome coding, chromosome decoding, initialization of the population, crossover operation and mutation operation are especially improved so that the machines and workers can be assigned to the process reasonably. Finally, the feasibility and efficiency of the model are proved by the simulation experiment.

Weizhong Wang, Jialian Wang, Chencheng Ma, Zhenqiang Bao, Richao Yin

Chapter 16. A Vision-Based Traffic Flow Detection Approach

Traffic flow detection plays an important role in Intelligent Transportation System (ITS). However, the conventional traffic flow detection approaches are high cost or complex installation. In this paper, a reliably vision-based traffic flow detection approach is proposed. In this approach, Gaussian mixture model (GMM) is employed to model the dynamic background of traffic scene. Then, the binary foreground contours are extracted by image subtraction. Comparing the binary vehicle contours’ location and the current frame, the real and complete vehicles are obtained for detecting and monitoring. In the part of vehicle counting, to gather the vehicle flow parameter in each lane of the road and avoid the trouble of counting vehicles repeatedly, a discriminative method is presented to classify vehicles into different lanes. Experiment shows that a desired result can be achieved in the traffic flow detection system by the vision-based approach.

Hongpeng Yin, Kun Zhang, Yi Chai

Chapter 17. The Research of High-Definition Video Processing System Based on SOC

As people puts forward more requirements for high-definition video industry, capture, transcoding, storage and display of high-definition video then become progressively imperative. This paper presents a parallel video SoC system for faster and flexible 1080P video processing. The key elements of this system, including parallel RGB to YUV transcoder, two-level pipeline and high speed memory controller, are elaborated. 0.028 um CMOS technology node is applied for implementing the SoC architecture. It is competitive in providing high quality images and is proved to be well supportive for DVI and HDMI signal.

Jian Gao, Xiaofu Zou, Yue Zhang, Fei Tao

Chapter 18. A Study on Feature Extraction of Surface Defect Images of Cold Steel

Feature extraction is one of the important characteristics used in classifying images. But the extracted features have big numbers and high dimension easily due to various type defects, complicated features and diverse methods of feature. Big numbers and high dimension of features are adverse for feature extraction. The effect of feature extraction decides the effect of image classification directly. According to these problems, experimental investigations are carried out on computer aiming at three typical surface defect images of cold steel strip, and this paper choose the gray features, textural features and Hu invariant moment features as the basis of classification finally. Experimental results demonstrated that features in this paper can be classification basis correctly.

Yingjun Guo, Xiaoye Ge, Hexu Sun, Xueling Song

Chapter 19. Consensus of Multi-agent System with Singular Dynamics and Time Delay

In this paper, the consensus problem of high-order multi-agent systems with singular dynamics and time-varying time delay is investigated. By the restricted equivalent transformation, a differential-algebraic system is introduced which is equivalent to this singular system on consensus. Based on dynamic state information and time delay, a consensus protocol is proposed. A sufficient condition of consensus is given in terms of LMI. Furthermore, the consensus state is also obtained by calculating the solution of the subsystem. Finally, an example is presented to illustrate the theoretical results.

Hua Geng, Zengqiang Chen, Zhongxin Liu, Qing Zhang

Chapter 20. Data-Driven Filter Design for Linear Networked Systems with Bounded Noise

Considering the case that the mathematical model of control plant is unavailable, this paper is concerned with the problem of data-driven filtering for linear networked systems with bounded noise and transmission data dropouts. One merit of the design is that the filter can be directly employed without identifying the model. To overcome the effect of data dropouts during the transmission, an output predictor is designed based only on the received output and input of the system. By utilizing the predicted output, a direct worst-case almost-optimal filter within the set membership framework is presented.

Yuanqing Xia, Li Dai, Wen Xie, Yulong Gao

Chapter 21. Distributed Finite-Time Coordination Control for 6DOF Spacecraft Formation Using Nonsingular Terminal Sliding Mode

This paper is devoted to the finite-time coordination control problem of 6DOF spacecraft formation with directed networks in the presence of external disturbances, and a distributed control algorithm using nonsingular terminal sliding mode (NTSM) is proposed. Based on Lyapunov methods, it is proved that all the spacecrafts achieve formation flying in finite time, which means the states of all the followers simultaneously converge to the states of the virtual leader. Simulation results are provided to validate the effectiveness of the theoretical analysis.

Fangya Gao, Yingmin Jia

Chapter 22. A Comparative Study on Optimal Transient Control of Aircraft Engines

Sequential quadratic programming has been widely used in the optimization of transient control of aircraft engine. On the other hand, the study of nonlinear optimization theory and method has been made great advance in recent years, of which active set method is one of the effective measures. In this article, the two algorithms are used to optimize the inputs based on one turbofan engine model. Results from the simulation comparisons of the two algorithms, advantages and disadvantages are given.

Yang Gao, Jiqiang Wang, Xin Wu, Zhifeng Ye, Zhongzhi Hu, Georgi Dimirovski

Chapter 23. Region-Based Shape Control for Multi-Robot Systems with Uncertain Kinematics and Dynamics

This paper is devoted to the shape control problem of multi-robot systems with uncertain kinematics and dynamics. An adaptive control method is proposed, based on which the closed-loop systems satisfy the region shape and collision avoidance requirements in task space. Using the Lyapunov-like function and the Barbalat’s lemma, a theoretical analysis on the motion of the robot end-effectors is proposed. Simulation results illustrate the feasibility and effectiveness of our controller.

Jia Yu, Yingmin Jia, Changqing Chen

Chapter 24. Image Threshold Processing Based on Simulated Annealing and OTSU Method

This chapter analyzes Maximum between-Cluster Variance method to conduct image threshold, coming up with an optimizing searching method of image segmentation with simulated annealing optimization algorithm. This algorithm determines the optimal threshold adaptively, and has strong adaptability and good effect of image segmentation, and it can greatly reduce the computational complexity. And it is optimized by multi-threading, which improves the parallel algorithm, and speeds up the efficiency of the algorithm.

Yue Zhang, Hong Yan, Xiaofu Zou, Fei Tao, Lin Zhang

Chapter 25. Fault Tolerant Tracking Control for a Team of Non-Minimum Phase VTOL Aircrafts Based on Virtual Leader Structure

This paper develops a fault tolerant approach for a team of thrust vector vertical take-off and landing (VTOL) aircrafts based on virtual leader structure. First, for each VTOL aircraft, a cascaded observer system with two extended state fault diagnosis observers (ESFDO) is designed. Then, local fault tolerant tracking controllers are developed. Finally, simulation results for a team of three VTOL aircrafts are presented to demonstrate the effectiveness of our proposed method.

Shuyao Cheng, Hao Yang, Bin Jiang

Chapter 26. Sliding Mode Fault-Tolerant Control for Air-Breathing Hypersonic Vehicles with External Disturbances

This paper studies the fault-tolerant control for air-breathing hypersonic flight vehicles(AHFVs) subject to both actuator faults and external disturbances. The overall procedure includes the feedback linearization of the model and the adaptive sliding mode control design. Specifically, a feedback linearized model of AHFVs is firstly obtained with the actuator faults and external disturbances. Then, two adaptive laws are designed to approximately estimate their unknown bounds. Based on the Lyapunov analysis method, it is proved that adaptive sliding mode control law can achieve the asymptotic tracking of all commands signals and guarantee the stability of the system. Simulations are conducted to confirm the efficacy of the proposed control method.

Wei Huang, Yingmin Jia

Chapter 27. On Indirect Model Reference Adaptive Learning Control

This chapter deals with tracking problem of single-input single-output linear time-invariant (SISO LTI) system with parametric uncertainties by using a model reference adaptive iterative learning control (MRAILC) scheme. The tracking error converges to zero pointwisely after infinite iterations. A major feature of this proposed method is that system output could track the desired trajectory whether the plant system is minimum-phrase or not.

Wen Du

Chapter 28. Stereo Vision Pose Estimation for Moving Objects by the Interacting Multiple Model Method

The stereo vision measurement system is very widely employed to obtain the 6 DOF pose information for the moving objects in space. However, the linear and angular velocities are impossible to estimate using these systems while the dynamic model is unknown and disturbances exist, and their applications is limited. To overcome this disadvantage, we propose an approach based on the IMM algorithm for moving objects. Our approach is verified in the feature points of a moving object. And the simulating results show its validity.

Ying Peng, Shihao Sun, Yingmin Jia, Changqing Chen

Chapter 29. A Carrier Tracking Algorithm of Kalman Filter Based on Combined Maneuvering Target Model

Since the traditional current statistical Kalman (CS-Kalman) filter doesn’t perform well enough when used in high-dynamic carrier tracking, an carrier tracking algorithm of Kalman filter based on combined maneuvering target model composed of high dynamic CS-Kalman filter and steady-state self-adaptive CS-Kalman filter is presented. The proposed algorithm achieves stable and accurate carrier synchronization by adjusting the CS-Kalman filter type and parameter corresponding to the dynamic condition in real time. Simulation results show that compared with the traditional CS-Kalman algorithm, the proposed algorithm is more realistic, practical valuable and adaptable in high dynamic environment.

Zhulin Xiong, Shijie Ren, Celun Liu, Jianping An

Chapter 30. Smooth Time-Varying Formation Control of Multiple Nonholonomic Agents

In this paper, we investigate the position/orientation formation problem of multiple nonholonomic agents. Coordinate transformations are first presented to obtain the matrix form of formation error model, then a distributed smooth time-varying control law is designed based on a Lyapunov-like function. We prove that the closed-formation-system is globally asymptotically stable by Barbalat’s lemma if the communication graph is undirected, time-invariant and connected. Simulation results verify the effectiveness of the proposed control scheme.

Chenghui Yang, Wenjing Xie, Chen Lei, Baoli Ma

Chapter 31. Robust Stability Analysis for Uncertain Time-Delay System

In this paper, the robust stability problem of time-delay system is discussed by constructing a Lyapunov-Krasovskii function. And two different systems are given in this paper, one of them has not consider the interference and unmodeled dynamics components, the other one add this uncertainties into the system. Based on this two different systems model, several general and powerful algorithms can also be given as theorems.

Wenfang Xin, Yaoyong Duan, Yi Luo, Qingmei Ji

Chapter 32. Interconnection Power Converter for Multi-Rail DC Distribution System

This paper presents an interconnection bi-directional dc-dc converter for multi-rail DC distribution system. The principles of the energy conversion process, small signal modeling and controller design for the new topology are discussed in detail. The system parameters are carefully calculated for an intended application of automotive 14 V/42 V dual-bus DC distribution system. A double-zero-double-pole PI controller is designed to control the high rail voltage as well as bidirectional power. The simulation and experimental results show that the proposed topology has the advantages of power expansibility, smaller capacitance current stress and bidirectional operation. It is proved to be applicable in high current high temperature automotive DC distribution systems.

Huiyang Lu, Yuru Zhu, Ying Hua, Wei Jiang, Nailu Li

Chapter 33. Sensitivity Analysis and Simulation of Performance for M / G / 1 / K Queuing Systems

In this paper, the performance potential and performance derivatives are analyzed for M / G / 1 / K queuing systems by the embedded Markov Chain. A computation algorithm is also given basing on a single sample path. To demonstrate the effectiveness of this algorithm, a special M / G / 1 / K queuing systems is given in the simulation section, which indicates the estimation error is very small.

Liping Yin, Hongyan Zhang, Lihua Wang, Li Zhou

Chapter 34. Bayesian Network Structure Learning Algorithms of Optimizing Fault Sample Set

As a representative heuristic search algorithm of structure learning algorithms for Bayesian network, K2 algorithm is easily trapped in local optimum in the process of Bayesian network structure optimization. This paper proposes a K2 algorithm based on chaotic perturbation. First of all, define the order of the node based on mutual information to construct a Bayesian network structure. Then introduce distribution function skew tent map combined with chaotic search. And ergodicity is used to jump out of local optimum and achieve global optimization. Finally, taking steering shaft control structure of B777 control system as an example, the proposed algorithm presents a good effect in Bayesian network structure learning.

Hongyang Guo, Rupei Zhang, Jiaqi Yong, Bin Jiang

Chapter 35. The Face to Face and Teleoperation by Using Remote Control of Robots

In this paper, we propose a system capable of remotely controlling mobile robots based on the wireless. In this system, we can not only control mobile robots to monitor each corner of home when we are out, but also establish interactive videos among friends and family members when we are far-flung geographically. In addition, we can also teleoperate the robot to accomplish certain task at home in case some switches are not well cut-off when leaving. Meanwhile, we can also make use of sensor to ensure the robot security in the case of network delay.

Xue Wang, Chaoli Wang

Chapter 36. Robust Adaptive Control for Robotic Systems with Guaranteed Parameter Estimation

In this paper, we propose a novel adaptive control scheme for robotic systems by incorporating the parameter error into the adaptive law. By carrying out filter operations, the robotic system is linearly parameterized without using the measurements of acceleration. Then a new adaptive algorithm is introduced to guarantee that the parameter error and control error exponentially converge to zero. In particular, we provide an intuitive method to verify the standard PE condition for the parameter estimation. The robustness against disturbances is also studied and comparisons to several adaptive laws are provided. Simulations with a realistic robot arm are presented to validate the improved performance.

Baorui Jing, Jing Na, Guanbin Gao, Guoqing Sun

Chapter 37. Cross-Media Big Data Tourism Perception Research Based on Multi-Agent

In this paper, we introduced the design and implementation in detail the cross-media big tourist perception system based on agent. We used the technology based on agent to implement the parallel in the process of data collection. This system mainly consists of a set of cooperation agent, including the data collection agent; the URL Agent; the data update agent and the management agent. Compared with the ordinary distributed information collection system, we designed and implemented the URL agent and data update agent, we used the algorithm to analyze the page based on agent, and then improves the efficiency of the cross-media big tourist perception.

Dong Guan, Junping Du

Chapter 38. Location and Navigation Study of Laser Base Station

Underground navigation is one of critical underground intelligent mining technologies. On account of underground specialty and working conditions, this article presents a research on location mode based on laser ranging and laser-based active navigation mode. First, the hardware design of base station control system used in laser location navigation is introduced. Second, the design of base station motion controller and communication module are described. Third, based on the motion controller, a moving target acquisition algorithm, laser location tracking algorithm, and laser-based active navigation algorithm are presented. Finally, some experimental results conducted in base station are illustrated.

Peng Wang, Weicun Zhang, Yuzhen Zhang

Chapter 39. Sentiment Analysis Based on Evaluation of Tourist Attractions

Tourists satisfaction has become more and more an indicator of tourism development. Sentiment analysis on data of comments and hot discussion on travel site and Weibo can help judge real-time satisfaction trend of tourists with scenic spots according to the intensity of sentimental tendency. Considering the deficiencies in current sentiment analysis, in this paper, firstly the polarity value and strength value are used to calculate the sentimental intensity of the sentimental words. HIT-CIR Tongyici Cilin (extended) is used to expand the synonyms of the sentimental words in order to reduce the impact of words not in HowNet and some sentimental words with low frequency in the corpus. Then we improved the traditional semantic similarity method based on HowNet according to the characteristics of sentiment analysis, combining it with the method based on Point Mutual Information (PMI) and syntactic dependency relations. High accuracy is shown by the experimental results.

Zhicheng Ma, Junping Du, Yipeng Zhou

Chapter 40. Target Localization and Tracking of Unmanned Mining Equipment Based on Multi-sensor Information Fusion Technology

The accurate localization is very important for mobile devices to make right decisions about autonomous path planning, avoiding obstacles and finishing other complex tasks. This paper presents a research on the localization and tracking technology of autonomous underground mining equipment. Two types of Kalman filters are considered as information fusion method: multi-sensor multi-model adaptive Kalman filtering and weighted adaptive multiple model Kalman filtering.

Xuan Li, Jiannan Chi, Weicun Zhang

Chapter 41. Improved Robust Multiple Model Adaptive Control of the Two-Cart Mass-Spring-Damper System with Uncertainties

A new weighting algorithm called Posterior Possibility Generator (PPG) is proposed to replace PPE algorithm in robust multiple model adaptive control (RMMAC) architecture, resulting in the improved robust multiple model adaptive control (IRMMAC) architecture, and a two-cart mass-spring-damper system with uncertainties is used to illustrate the advantages of PPG against PPE.

Ya Wang, Baoyong Zhao, Weicun Zhang

Chapter 42. An Improved Laplacian Eigenmaps Algorithm for Nonlinear Dimensionality Reduction

Manifold learning is a popular recent approach to nonlinear dimensionality reduction. While conventional manifold learning methods are based on the assumption that the data distribution is uniform. They are hard to recover the manifold structure of data in low-dimension space when the data is distributed non-uniformly. This paper presents an improved Laplacian Eigenmaps algorithm, which improved the classical Laplacian Eigenmaps (LE) algorithm by introduce a novel neighbors selection method based on local density. This method can optimize the process of intrinsic structure discovery, and thus reducing the impact of data distribution variation. Several compared experiments between conventional manifold learning methods and improved LE are conducted on synthetic and real-world datasets. The experimental results demonstrate the effectiveness and robustness of our algorithm.

Wei Jiang, Nan Li, Hongpeng Yin, Yi Chai

Chapter 43. A Distributed Charging and Discharging Coordination for Large-Population Plug-In Electric Vehicles

Motivated by the economic and environmental benefits of plug-in electric vehicles (PEVs), the PEVs have been regarded as adjustable bilateral auxiliaries for the power grid. They can not only charge from the grid but also discharge back to the grid. This work studies the problems of PEV coordinations, considering the battery degradation cost. We analyze the characteristics of distributed strategy curve and propose an updated algorithm to implement the optimal distributed coordination in the case of the infeasibility of fully centralized formulation. As key contribution of this paper, we show that under certain mild conditions, the system converges to an optimal charging and discharging strategy. Simulation examples illustrate the results developed in the paper.

Huiling Li, Zhongjing Ma

Chapter 44. 3D Model Classification Based on Transductive Support Vector Machines

With the development of graphical modeling methods and acquirement technology of 3D models, the numbers of 3D models increased exponentially. 3D model classification is crucial to effective management and precise retrieval. Due to the shortage of annotations, we adopted semi-supervised learning methods to solve this problem. This paper introduces Transductive Support Vector Machines (TSVMs) for 3D model classification. Our proposed method is verified efficient.

Qiang Cai, Zhongping Si, Haisheng Li

Chapter 45. Robot Localization Based on Optical-Flow Sensor Array

Recently, Optical-flow sensors commonly used as a PC input device have been explored in robot localization. One common difficulty is that ground-height variations in mobile robot moving can inevitably deteriorate the sensing performance of optical-flow sensors when used in a trivial fashion. In this paper, a novel deployment configuration is presented to build an optical-flow sensor array for robot localization, which is advantageous in the robustness to ground-height variation. In addition, the imaging capability of optical mouse sensors is exploited in developing a beacon based calibration approach for eliminating accumulative errors. The experimental results are reported to validate the proposed method.

Shaoxian Wang, Mingxiao He, Guoli Wang, Xuemei Guo

Chapter 46. Composite Continuous Anti-disturbance Autopilot Design for Missile System with Mismatched Disturbances

In this paper, the problem of autopilot design for missile system with mismatched disturbances is considered via a sliding mode control method and finite time disturbance observer. Firstly, a finite time disturbance observer is designed to estimate the mismatched disturbances. Secondly, based on disturbance estimation values and traditional sliding mode surface, a novel sliding mode surface is constructed. Thirdly, a composite continuous anti-disturbance autopilot is developed, which can guarantee system output converge to reference signal. Finally, a simulation result is presented to demonstrate the effectiveness of the proposed scheme.

Xinge Liu, Haibin Sun, Jiayuan Shan

Chapter 47. On Three-Player Potential Games

In this paper, some new criteria for detecting whether a 3-player game is potential are proposed by solving potential equations. When a 3-player game is potential, the potential function is constructively expressed.

Xinyun Liu, Jiandong Zhu

Chapter 48. Flocking of Multi-Agent Systems with Multiple Virtual Leaders Based on Connectivity Preservation Approach

In this paper, we investigate the problem of flocking with multiple virtual leader based on connectivity preserving. The basic idea is stated as follows: according to design stable control law with a navigation feedback term the follow agents equipped with virtual leaders. The velocity of the center of the mass of agents will exponentially converge to weighted average velocity of the virtual leaders by the information navigation of virtual leaders, finally approach to the same. The certain distance between every agent is kept and eventually form flocking due to the introduction of the artificial potential function. And on this basis we assume that the initial network is connected and the control law make the network of the multi-agent system preserving connectivity.

Zonggang Li, Chen Na, Guangming Xie

Chapter 49. An Adaptive Weighted One-Class SVM for Robust Outlier Detection

This paper focuses on outlier detection from the perspective of classification. One-class support vector machine (OCSVM) is a widely applied and effective method of outlier detection. Unfortunately experiments show that the standard one-class SVM is easy to be influenced by the outliers contained in the training dataset. To cope with this problem, a robust OCSVM is presented in the paper. In consideration that the contribution yielded by the outlying instances and the normal data is different, a robust one-class SVM which assigns an adapting weight for every object in the training dataset was proposed in this paper. Experimental analysis shows the better performances of the proposed weighted method compared to the conventional one-class SVM on robustness.

Jinhong Yang, Tingquan Deng, Ran Sui

Chapter 50. An Improved Kalman Filter for Fractional Order System with Measurement Lévy noise

In this paper, taking the measurement noise as the non-Gaussian Lévy noise, an improved Kalman filter for discrete linear stochastic fractional order system is proposed. By eliminating the maximum of the noise, the Lévy noise can be approximated by a series of Gaussian white noises. Then, based on the principle of least square, an improved Kalman filter is developed for discrete linear stochastic fractional order system with measurement Lévy noise. Finally, simulation results are provided to illustrate the effectiveness and usefulness of the proposed filter designing algorithm, where a better filtering performance could be found.

Yi Wang, Yonghui Sun, Zhenyang Gao, Xiaopeng Wu, Chao Yuan

Chapter 51. Nonlinear Control of Inertial Wheel Pendulum and Its Implementation Based on STM32

The swing-up and balance control problems of inertial wheel pendulum are investigated in this work. We firstly introduce the dynamic model of inertial wheel pendulum and then propose a swing-up control law by using an energy function. Theoretical analysis shows that the system gradually moves into an invariant set during swing-up process. Later, we propose the balance control law with approximate linearization and introduce a self-built experimental platform based on STM32 microprocessor. The final combination of swing-up control law and balance control law leads to a hybrid control law. We verify the efficiency of the proposed control law on the apparatus.

Yan Lixia, Ma Baoli

Chapter 52. Intermittent Synchronization of Cascaded Boolean Networks with Time Delays

In this paper, we considered the intermittent synchronization of cascaded Boolean networks (CBNs) with time delays and get its criteria based on the results of synchronization of delay-free CBNs. We further discuss complete synchronization of CBNs with time delays from intermittent synchronization.

Zhengdong Yang, Tianguang Chu

Chapter 53. Overhead Transmission Line Condition Evaluation Based on Improved Scatter Degree Method

As an important part of the transmission network, the running state of overhead transmission line will directly affect the reliability of the whole power system. With a single element reliability model, it is hard to exactly reflect the true condition of the complicated system with multi hierarchies and multi indicators, which means it is difficult to conduct an accurate comprehensive evaluation on overhead transmission line. This paper proposes an overhead transmission line evaluation model based on improved scatter degree method. The proposed model combines the subjective and objective combination weighting method, which can not only calculate the line units, but also evaluate the running state of the whole overhead transmission line. Finally, this model is used to assess overhead transmission lines in a certain area and results show that the proposed method is effective and feasible.

Guoqiang Sun, Chao Zhu, Ming Ni, Zhinong Wei

Chapter 54. A Novel Driving Method for Switching Control Pulse Signal

A novel driving method implemented with high frequency electromagnetic isolation is proposed in this paper. And the hardware designing schematic diagram and the software configuration method are presented. Then the corresponding operating principle is analyzed in detail based on digital control chip and analog circuits. The proposed driving method can be applied to real-time varying pulse width of the control signal and the duty-cycle ratio of this control pulse can be regulated in the range of 0–100 %. This novel driving method can achieve not only the advantages of high insulation and high common-mode suppression, but also the merits of low cost and fast dynamic response speed, and it is suitable for the isolation and amplification of high frequency switching control pulse signal. Finally, the experimental results show that the proposed driving method in this paper is effective.

Jin-Yan Zheng, Yu Fang, Pan Xu, Qi-Qi Zhao

Chapter 55. Study of Transformer Fault Diagnosis Based on DGA and Coupled HMM

In this paper, a coupled hidden Markov model (CHMM) based on dissolved gas analysis (DGA) is proposed for transformer fault diagnosis in power systems. By using the theory of scalar quantization, the collected date of dissolved gas is preprocessed and put into two observation channels in the coupled two-chain hidden Markov model (HMM) with the form of gas content and ratio. Transformed to the form of HMM equivalently, CHMM can be easily trained and work as the condition classifier for power transformer. Finally, it follows from the experimental results and comparison with HMMs that the proposed model is successful and effective.

Hao Guo, Zhinong Wei, Ming Ni, Guoqiang Sun, Yonghui Sun

Chapter 56. On Regular Subspaces of Boolean Control Networks

This paper investigates some fundamental problems on regular subspaces of Boolean control networks (BCNs). A new necessary and sufficient condition for regular subspaces is obtained. A new method to compute complementary subspaces is proposed. An example is given to illustrate the obtained theoretical results.

Jiandong Zhu, Pengjing Jü

Chapter 57. Research on Evolution and Simulation of Transaction Process in Cloud Manufacturing

Cloud manufacturing is a new service-oriented networked manufacturing mode. The participants of cloud manufacturing platform conduct transactions in an open environment. This paper focuses on the cooperation of supply-sides and demand-sides in cloud manufacturing platform from the perspective of evolutionary simulation. By combining the agents and evolutionary simulation, we design and implement cloud manufacturing platform simulation model based on service agents. Meanwhile, transactions and resource utilization based on cloud manufacturing and traditional manufacturing modes have been compared through evolutionary simulation. The experimental results demonstrate that cloud manufacturing simulation platform based on service agents could represent the main characteristics of cloud manufacturing transactions. And in this way, the advantages of cloud manufacturing cloud also be illustrated.

Chun Zhao, Lin Zhang, Bowen Li, Jin Cui, Lei Ren, Fei Tao

Chapter 58. Robust Output Feedback Control of Civil Aircrafts with Unknown Disturbance

In this paper, the disturbance attenuation problem is investigated for the flight control system of civil aircrafts. The unknown disturbance is generated by a linear exogenous system, and the civil aircraft dynamic model with unknown disturbance is established. The disturbance is augmented to the state vector and an augmented state observer is designed for estimating the unmeasured augment state vectors. Based on the designed augment state observer, the robust output feedback control scheme is developed for the civil aircrafts with unknown disturbance. Focusing on the B747-100/200, simulation results show that the developed controller could attenuate disturbance effectively.

Mengjie Wei, Mou Chen, Qingxian Wu

Chapter 59. Photovoltaic System Power Generation Forecasting Based on Spiking Neural Network

A forecasting model based on Spiking neural network (SNN) was proposed to tackle with the problem of the forecasting of photovoltaic system (PVS) power generation. This neural network uses temporal encoding scheme with precise times of spikes, which is closer to the real biological neural system and has powerful computing ability. Considering the main influencing factors such as season types, weather types, sunshine intensity and temperature etc., this model use the method of grey correlation analysis to select similar days. The high accuracy and robust applicability of the proposed forecasting model are verified by the simulation using actual operating data of PVS.

Tong Chen, Guoqiang Sun, Zhinong Wei, Huijie Li, Kwok W. Cheung, Yonghui Sun

Chapter 60. Stochastic Stabilization of a Class of Nonlinear Systems with Sampled Data Control

In this paper, the stabilization problem is investigated for a class of nonlinear stochastic systems with sampled data control. By means of the Lyapunov stability theory, a sufficient condition is derived to ensure that the considered stochastic nonlinear system with sampled data is asymptotically stable in mean square. The allowable bound of the sampling interval is obtained and the control gain matrix is solved in terms of convex optimization methods. Finally, a numerical example is presented to further demonstrate the effectiveness of the proposed approach.

Yueling Shen, Wenbing Zhang

Chapter 61. A Reconfigurability Evaluation Method for Satellite Control System

The reconfigurability of satellite control systems is a fundamental index to the capability of spacecraft stability after the faults occur in the process of satellite operating in-orbit. This paper addresses the problem of reconfigurability evaluation for faulty control systems. Firstly, The reconfiguration indicator based on system stability is presented by regarding the faults that occur in the system as model uncertainties. And then the method for normalized coprime factorization is introduced to describe the maximal reconfigurability boundary, which causes the control system instability when the fault exceeds the allowable range of the given controller and model. Finally, the efficacy of the proposed method is tested through an numerical simulation on the Hubble telescope.

Heyu Xu, Dayi Wang, Wenbo Li

Chapter 62. LS-SVM Generalized Predictive Control Based on PSO and Its Application of Fermentation Control

Fermentation process is a complex time-varying, nonlinear and multivariable biochemical process. The traditional fed-batch fermentation conditions are difficult to satisfy the control request. A control algorithm based on Generalized Predictive Control (GPC) is proposed. Firstly, the algorithm utilizes least square support vector machine (LS-SVM) and GPC to construct the prediction model and forecast the output value. And then, the particle swarm optimization (PSO) algorithm is applied to realize rolling optimization and obtain the control values. Finally, the control algorithm is applied to control the substrate concentration (S) of lysine fermentation. The simulation results show that the LS-SVM Generalized Predictive Control based on PSO has an excellent adaptive ability with rapid control response speed, high precision, and good performance.

Li Huang, Zhaohua Wang, Xiaofu Ji

Chapter 63. Delay Consensus of Second-Order Nonlinear Leader-Following Multi-Agent Systems

In this paper, the delay consensus of second-order nonlinear leader-following multi-agent systems is discussed. The considered multi-agent system has an active leader and the information exchange between two different agents possesses directional. A simple input control law is proposed. Based on the matrix theory and Lyapunov stability theory, the effectiveness of this control law is proved and a sufficient condition is obtained to realize delay consensus of the second-order multi-agent system.

Jiezhi Wang, Qing Zhang, Hang Li

Chapter 64. An Investigation of Model-Based Design Framework for Aero-Engine Control Systems

Throughout the design of aero-engine control systems, modeling and simulation technologies have been widely used for supporting the conceptualization and evaluation. Due to the increasing complexity of such systems, the overall quality management and process optimization are becoming more important. This in particular brings the necessity of integrating various domain physical models that are traditionally based on different formalisms and isolated tools. In this paper, we present the initial concepts towards a model-based design framework for automated management of simulation services in the development of aero-engine control systems. We exploit EAST-ADL and some other existing state-of-the-art modeling technologies as the reference frameworks for a formal system description, with the content ranging from requirements, to design solutions and extra-functional constraints, and to verification and validation cases, etc. Given such a formal specification of system V&V (Verification and Validation) cases, dedicated co-simulation services will be developed to provide the support for automated configuration and execution of simulation tools. For quality management, the co-simulation services themselves will be specified and managed by models in SysML.

Jinzhi Lu, DeJiu Chen, Jiqiang Wang, Wentao Li

Chapter 65. Container Throughput Time Series Forecasting Using a Hybrid Approach

This paper proposed a novel two-stage hybrid container throughput forecasting model. Time series in reality exhibits both linear and nonlinear characteristics and individual models are not able to describe the two features simultaneously. Therefore, we combine linear model SARIMA (seasonal autoregressive integrated moving average) and nonlinear model ANN (artificial neural network). In order to break through the limitations of traditional hybrid models, based on the identified parameters of SARIMA in first stage, the structures of several ANN in second stage could be decided. Finally, we validate the proposed hybrid model 5 performs best with case study in Shanghai port.

Xi Zha, Yi Chai, Frank Witlox, Le Ma

Chapter 66. A Dynamic Performance Analyzing Method of Intelligent Fire Control System Based on SCPN

To describe tank future fire control system logic level generally and exactly, a modeling method is advanced which based on Stochastic Colored Petri Net (SCPN). Firstly, the constitutive structure, information structure and working flow of future fire control system are studied. Secondly, the tank future fire control system is formalized described. The rule which transform tank fire control system to Petri Net is established. Finally, Petri Net model of tank future fire control system is established. Its dynamic performance is analyzed. It proves the modeling method can describe tank future fire control system logic level effectively.

Tianqing Chang, Junwei Chen, Bin Han, Kuifeng Su, Rui Wang

Chapter 67. Forecast of Train Delay Propagation Based on Max-Plus Algebra Theory

The forecast of delay time is of great assistance to decision making in train operation adjustment when the schedule is disturbed either by infrastructure fault or natural hazard. This paper presents a railway delay propagation model to forecast the delay time, described by discrete event dynamic system (DEDS) and formulated by max-plus algebra theory. On the basis of the train operation regulations and headway constraints, a system matrix of max-plus algebra is acquired to illustrate the mechanism of delay propagation. And then a function to predict the delay time is proposed to solve the model, with two advantages: Firstly, the specific delay time is able to be calculated; secondly, the result of the prediction is comparatively precise due to the highly match of the model to the actual operation. Finally, by analysis of the prediction, this paper offers the decision support in train adjustment, from which the dispatcher can proactively conduct countermeasures to alleviate the propagation and even stop it.

Hui Ma, Yong Qin, Guoxing Han, Limin Jia, Tao zhu
Weitere Informationen

Premium Partner