Search

The paper has illustrated the present situation of the soccer, and has analysed the characteristics of PID control algorithm. This paper has applied the HSIC theory and methods, and has improved the nonlinear PID controller. This paper has built the new time-slice concept for Microsoft's robot soccer platform, and has realized the optimal control process of turn-corner and to-point function for the robot soccer platform based on Microsoft Robotics Studio. The simulation experiment has proved that the nonlinear PID controller to being improved is more effective.

For the model time varying during point stabilization control of two-wheel robot with nonholonomic constraint, velocity, and acceleration limit, this paper proposes a multi-mode proportion control approach based on neural network. The point stabilization intelligence control is achieved by optimizing the multi-mode human simulated intelligence control(HSIC) controller with BP neural network online learning. With the performance variation of robot motion execution system, the simulation results compare the approach we proposed with the proportion controller and proportion cosine controller. The validity of the approach is confirmed in the simulation.

This paper analyzes the existing decision-making system of Chinese Chess computer game, and points out some of disadvantages. Then, a new three-level decision-making system of Chinese Chess computer game is constructed. In this decision-making system, affection computing is introduced. And the system achieves the high-speed searching and decision-making top-down though affection computing. Comparing with the existing decision-making system, the three-level decision-making system is proved workable.

Data in major commercial petroleum industries are complex in nature and often poorly organized and duplicated, and exist in different formats. Due to the diverse nature of business products and operations in different geographic locations, these industries demand more accurate and precise information and data. This study presents petroleum exploration domain ontology-based knowledge integration and sharing system framework to minimize the complexity of heterogeneous data, and enhances power of knowledge integration and information sharing among different operational units. The mechanism can support collaborative petroleum exploration process development by providing functions of knowledge integration and sharing. Results of this study facilitate the knowledge integration and sharing of petroleum exploration businesses to satisfy the knowledge demands of participants, and increase company's competitive capability, reduce exploration and development life cycle time and costs, and ultimately increase enterprise marketability.

Through thorough analyses of the emotion-related theories in psychology and microeconomics, the attributes of emotion are clarified whose essence is energy, function is driving behavior, sort reflects the sort of survival resources, and the emotional essence of motivation, (inner) drive and utility (subjective value) is discovered. According to the common law abided by these kinds of sense, and consulting Prospect Theory, emotion schema function is build up to obtain the principle of functional relationship between emotion and living resource. Based on these work, adopting optimistic selection principle, emotion-driven behavior selection mechanism is established. In order to validate the mechanism, emotion-driven competitive, cooperative behavior selection mechanism and fight behavior selection mechanism are constructed and simulations of them on artificial life platform Swarm are achieved. The simulation results confirm the proposed.

Introduces a prototype of wearable observation supporting system for helping human complete the observation and recognition tasks based on wearable camera. The research presented here addresses special challenge problems encountered in wearable computing environment. The implementation of new architecture and system process flow with functions will help the wearers real-time face detection, recognition and identification. The algorithm begins with capturing sufficient quality video frames from wearable camera followed by interested object detection and recognition. We demonstrate effectiveness and the feasibility of this system by experimental results.

To expand the initial angle condition in stabilizing the double inverted pendulum, the adaptive Human-Simulated Intelligent Controller is proposed. Based on it, the phase plane is divided into tow blocks according to requirement of control performance. Within the zero block the integration action works; Outside the zero block the adaptive fuzzy PD works. This control structure, called after adaptive Human-Simulated Intelligent Controller, provides the strong nonlinearity performance to satisfy the control requirement. The simulation results show this method expands the initial angle conditions 1.5 times that of the classical LQR controller and obtains the better stabilization performance.

Humanoid robot posture includes a phase where the stance heel rises from the ground and the stance foot rotates about the stance toe. The humanoid robot becomes under-actuated system in this phase. First, the dynamical model of robot is developed and the controllability of the robot is analyzed. Then, the controller can be designed based on Linear Quadratic Regulator (LQR) optimal control theory. The effectiveness of the proposed method is illustrated by simulations.

Aimming at the problem of the field of robotics soccer system and the identification of robotics soccer, based on the seed point, this paper puts forward a method of field distortion correction, analyses the characteristic of different color tag design approaches, constructs a new color tag approach and its identification algorithm, rapid calibrating field and identifying robotics soccer are realized in high-confrontational, high-dynamic robotics soccer games. Through the trainning of FIRA Mirosot platform and actual games, the effectivity and reliability of this algorithm are proved and the need for real-time of robotics soccer system is met.

Applying Emotion Behavior Selection Mechanism to Dom World fight mode, a fight behavior mechanism based on emotions is built to discover how the result of a former fight influences, in emotion pattern, the decision and achievement of a letter fight. On Swarm soft platform of Artificial Life, consulting to the design and data of Dom Word simulation, a simulation of the proposed mechanism has been conducted. The results of the simulation confirm the present mechanism, by centering on the exhibition of self-organization (emergence), a classical characteristic of Artificial Society. Adopting ADI and Netto to evaluate the stimulation results of these two mechanisms, it proved that the present excels Dom World mode. And more, by comparing and analyzing the results of simulations including and excluding feedback emotion, it’s found that feedback mechanism increases the intensity of fight, which supports Darwin’s hypothesis that emotion speeds up the natural selection.

A double-loop control system based on humansimulated intelligent control (HSIC) is designed using a mathematical model of a brushless direct current motor. The system includes a PI current controller and an HSIC speed controller, and controller parameters are set and optimized using an improved genetic algorithm. In the double-loop motor control system, a characteristic model-based multi-mode controller is adopted for the speed loop and a conventional PI controller is implemented for the current loop. The design method of the controller and its algorithm are presented. Compared with the double-loop motor subject to fuzzy control, neural-network control, adaptive PID control and conventional PID control, simulation results show the double-loop motor control system of the HSIC-based the brushless direct current motor has high dynamic and static performances as well as high adaptability and robustness.

Semi-active suspension system utilizing magneto-rheological (MR) fluid is being accepted by the automotive industry due to their controllable force characteristics. However the time delay of MR suspension system is long enough to affect the control performance. To address this problem, an experimental apparatus was founded to test the systempsilas response time. And a human-simulation intelligent control (HSIC) strategy was designed to compensation it. Simulation and road test verified effects of HSIC in solving the problem of time-delay of MR suspension.

For RoboCup middle-size league, vision processing have its particular characteristics. The requirement of real time and multi-channel image segmentation processing make the algorithm of component encode consume less processing time. In this paper, the new component encode algorithm, which is based on run-length encode, uses a new run structure and other assistant data structure to reduce the memory spaces, these structures make run movement and comparison more effective. This paper also proposes the methods of run movement, run comparison and index combination. Experiment show the proposed algorithm remarkably reduces processing time, run comparison and index combination. The proposed algorithm is very suitable for the vision processing in RoboCup middle-size league.

Through the analysis of current reactive/deliberative structures, based on schema theory, according to bionics, this paper provides a new behavior selection structure. Firstly, simply analyses the existing methods of behavior selection; Secondly, introduces the behavior selection structure based on evaluation-tradeoff in detailly and proposes two concepts, namely perception satisfied degree and exhibition desired degree; Finally, taking soccer robot as experiment platform, describes the method on how to design behavior, analyses the experiment results which proves the validity of structure proposed in this paper. Simulation experiment results show that, based on this new behavior selection structure, the artificial life not only imitates the natural life behavior realistically in unknown environment, but also has higher dynamic programming intelligence.

A new kind of intelligent control is discussed, namely, human simulated intelligent control (HSIC), which hierarchically include three levels: centrum chief level (CC), organizing and coordinating level (OC), and unit control level (UC). Then the control problem of arbitrary switch control of circular-rail double pendulum (CRDP) is studied. Finally, the control system for this control problem is designed with HSIC. Especially, the realization of unit control level (UC) is explained in detail with the example of one of the switch control - down-down to down-up. Real-time experimental results are presented.

The main process in hydroponics plants cultivating is aquatic induction. Aquatic induction is the process of cultivating plants in the right artificial environment to induce aerenchyma formation, which can make plants be grown in water. With long-time practices and researches, rules and environment parameters for aquatic induction was summarized. And a set of fuzzy control system for hydroponics cultivating was developed. Making use of the experience from practices, a method was put forward to control dissolved oxygen, CO2 and EC in some hydroponics cultivating greenhouses without oxygen detector, CO2 detector and EC detector. The result of the system running shows that the system can achieve hydroponics cultivating without watchers.

Introduces a prototype of wearable surveillance assistant system for helping human complete the surveillance tasks. The research presented here addresses the system requirements and challenge problems encountered in wearable computing environment for an assistant surveillance task. The implementation of new architecture and functions will help the surveillant in three ways: real-time object detection, intelligent appropriate alarm and automatically create the surveillance report. The algorithm begins with capturing video frame from camera followed by motion detection and interested object detection. We demonstrate effectiveness and the feasibility of this system by experimental results.

Question analysis is the primary task of question answering system. Performing Interrogative phrase classification based on the chinese interrogative phrases and their usages before parsing words can reduce the influence of the parsing errors to question analysis. Standard question notation (SQN), which composed of question classes denotation and keywords denotation, was put forward to describe a concrete question sentence, which can make the question analysis more simply. All of the process of the question analysis was described as follow: identifies the IPC, parses words, identifies the question class and gets the SQN based on the sentencepsilas IPC and matching the sentence styles, acquires the answer patterns and rewritten sentences of questions based on SQN. The result of experiments shows that the question analysis implemented by this way is effective.

It is very important to hold the wheeled robotpsilas attitude while running. The adaptive suspension system with magneto-rheological (MR) damper can optimize its attitude by providing continuously variable real-time damping control based on road profile. To address this problem, a control strategy with schema-based human simulated intelligent control (HSIC) theory is proposed and studied. According to a seven DOFs dynamic model, eight states of robotpsilas attitude are divided and relevant schemas are designed. The performances of the control system under two types of road excitations are evaluated by computer simulation. The results indicate that HSIC can achieve better attitude stability than LQG and passive system.

A hybrid simplex-genetic algorithm (HSGA) is presented to solve global numerical optimization problems. The HSGA combines the traditional genetic algorithm, which has a powerful global exploration capacity, with simplex algorithm, which can exploit the local range. Some improved mechanism are introduced in the HSGA, such as hybrid encoding, orthogonal design, and feedback mutation etc. so the HSGA can be more robust, statically sound, and quickly convergent. The proposed HSGA is applied to solve benchmark problems. The computational experiments show that the HSGA can find the optimal or close-to-optimal solutions. It is also validated that the HSGA is efficient.

Based on reduced model of double-fed direct current motor system (DFDCMS), the state space description and its key states of DFDCMS are advanced. The two wheeled robotpsilas kinematics model and motion-actuator model are analyzed, the state observer based on extended Kalman filter (EKF) of the two wheels mobile robot (TWMR) is used to filter the signal with interference and get the key states. There are tree kinds of state observers, including the direct state observer with nothing filter(NF), the state observer with Butterworth Filter(BF) and the state observer with EKF, are used to compare on two kinds of experiment platform, including the simulation system and the fact system of RoboCup soccer robot. As an application example, the motion states of TWMR have been analyzed base on EKF state observer during the course of TWMR going to one goal. The experimentation indicates that the state observer based on EKF can reflect the running states of test system more availably.

This study discusses the rapid control prototyping approach used for development and investigation of control algorithm in magneto-rheological (MR) semi-active suspension system. Firstly, a multibody dynamic model based on MR suspension system is developed with the aid of VEDYNA software. Based on the model, a new intelligent control algorithm, human simulated intelligent control (HSIC) is proposed and designed. Finally, by means of dSPACE groupware system, the semi-active control prototype of real car has been built up, road test of real car and parameters of online adjustment have also been carried out. The results of offline simulation and real time control show that HSIC can achieve better comfort under assuring stability. They also show that rapid control prototyping (RCP) on the basis of Matlab/Simulink and dSPACE can shorten the research period of vehicle semi-active controller and reduce the cost of research. It is very significant to promote the industrialization of vehicle semi-active controller.

Using fuzzy logic rules, a self-tuning structure for the self-tuning level of a human simulated intelligent controller (HSIC) has been designed. This structure has fuzzy self-tuning abilities in the swing-up control of a cart-double pendulum system. It efficiently accomplishes the online self-tuning of the control parameters of a HSIC with hierarchical and multi-mode control structure. The computer simulation and real-time experiments of the swing-up control of a cart-double pendulum system show that the online fuzzy self-tuning of the control parameters markedly enhances the robustness and adaptability of the HSIC.

Take the handstand-stabilization- control problem of Cart-Double- Pendulum System (CPS) as an example, this paper detailed a general control parameters optimization and system implementation method for human-simulated intelligent controller. Firstly, the physical and dynamic models of CDPS were analyzed, and then the research platform for the controller of CDPS is also built. Next, the parameters optimization method based on Improved Genetic Algorithm (IGA) for human-simulated controller was introduced in detail, and the effectiveness of parameters optimization method was testified by simulation and experiments. In the end, the conclusion is summarized.

Through the analysis of traditional potential field approach, and based on the bionics and schema theory, we create a behavior selection classification model of artificial life body (ALB) based on the evaluation-tradeoff, then we study on the behavior coding based on the dynamic potential filed. Therefore, the optimum behavior selection problem in live environment is solved. Simulation results indicate that put the potential field dynamically in the model, not only enables the ALB to select action according to the outside and inside environment information, and exhibits higher intelligence, but also makes the ALB to imitate the nature life behavior realistically in the unknown environment.

The self-control photo system (SCPS) synthetically combines virtual reality and computer technique in amusement photo systems. To achieve high quality image output, this paper presents a new image fusion method based on analysis of human vision perception. The experiment reveals that this method fully preserves valuable original image information and realizes smooth connection, which attributes to a satisfied vision impact.

Schema theory (ST) is a methodology of analyzing complex and dynamic systems of human and other creatures. The theory explains the formation of intelligence, the structure and organization of cognition, etc. The theory has been applied in areas of artificial intelligence, autonomous robot and autonomous agents, etc. Fundamental concepts of ST and Human Simulated Intelligent Control (HSIC) theory are introduced. The perceptual schema, motor schema, associated schema and Sensory-Motor Intelligent Schema (SMIS) based on HSIC methodology and applications are described. The fundamental idea and design principles of HSIC theory based on SMIS, and the application in the swinging-up and handstand-control of cart-double pendulum system are given in detail. Combination of HSIC theory with ST is a valid method to resolve complex system control problems.

Studied the control problem concerning the swinging-up and handstand-control of the cart-double-pendulum system (CDPS, a typical under-actuated, unsteady and nonlinear system) based on the Human-Simulated Intelligent Control (HSIC) theory. By analyzing the movement of CDPS and simulating the Human controller, designed the Sensory-motor Intelligent Schemas of HSIC controller for the CDPS. In the swinging-up process, the HSIC controller perceives the coupled motion of two pendulums and uses the multi-mode control strategy based on characteristic identification, to keep the angle and movement between the two linked pendulums minimum all the time. The double-pendulum then becomes a /spl square/uasi-cart-single-pendulum? Simulation and actual experiments successfully proved the validity of the above method.

On the basis of the human simulating intelligent control (HSIC) proposed by the author this paper discusses the swinging-up control of cart-pendulum. The theoretical analysis and simulation results have verified the universal validity of the HSIC theory which aims at the emulating human "motor sensory preview intelligence" for the nonlinear system such as cart-pendulum, walking robots, etc.