Behavior-based dual dynamic agent architecture

The objective of the architecture is to make agent promptly and adaptively accomplish tasks in the realtime and dynamic environment. The architecture is composed of elementary level behavior layer and high level behavior layer. In the elementary level behavior layer, the reactive architecture is introduced to make agent promptly react to events( in the high level behavior layer, the deliberation architecture is used to enhance the intelligence of the agent. A confidence degree concept is proposed to combine the two layers of the architecture. An agent decision making process is also presented, which is based on the architecture. The results of experiment in RoboSoccer simulation team show that the proposed architecture and the decision process are successful.     

Formation Feedback Applied to Behavior-Based Approach to Formation Keeping

Approaches to the study of formation keeping for multiple mobile robots are analyzed and a behaviorbased robot model is built in this paper. And, a kind of coordination architecture is presented, which is similar to the infantry squad organization and is used to realize multiple mobile robots to keep formations. Simulations verify the validity of the approach to keep formation, which combines the behavior-based method and formation feedback. The effects of formation feedback on the performance of the system are analyzed.     

Multi-agent reinforcement learning with cooperation based on eligibility traces

The application of reinforcement learning is widely used by multi-agent systems in recent years. An agent uses a multi-agent system to cooperate with other agents to accomplish the given task, and one agent‘s behavior usually affects the others‘ behaviors. In traditional reinforcement learning, one agent takes the others location, so it is difficult to consider the others‘ behavior, which decreases the learning efficiency. This paper proposes multi-agent reinforcement learning with cooperation based on eligibility traces, i.e. one agent estimates the other agent‘s behavior with the other agent‘s eligibility traces. The results of this simulation prove the validity of the proposed learning method.     

Biomimetic Experimental Research on Hexapod Robot''s Locomotion Planning

To provide hexapod robots with strategies of locomotion planning, observation experiments were operated on a kind of ant with the use of high speed digital photography and computer assistant analysis. Through digitalization of original analog video, locomotion characters of ants were obtained, the biomimetic foundation was laid for polynomial trajectory planning of multi-legged robots, which was deduced with mathematics method. In addition, five rules were concluded, which apply to hexapod robots marching locomotion planning. The first one is the fundamental strategy of multi-legged robots’ leg trajectory planning. The second one helps to enhance the static and dynamic stability of multi-legged robots. The third one can improve the validity and feasibility of legs’ falling points. The last two give criterions of multi-legged robots’ toe trajectory figures and practical recommendatory constraints. These five rules give a good method for marching locomotion planning of multi-legged robots, and can be expended to turning planning and any other special locomotion.     

The determination of the link with the smallest end-to-end network latency in ethernet architecture

Ethernet fundamental and its data transmission model are introduced in brief and end-to-end network latency was analyzed in this paper. On the premise of not considering transmission quality and transmission cost, latency was the function of the rest of network resource parameter （NRP）. The relation between the number of nodes and that of end-to-end links was presented. In ethernet architecture, the algorithm to determine the link with the smallest latency is a polynomial issue when the number of network nodes is limited, so it can be solved by way of polynomial equations. Latency measuring is the key issue to determine the link with the smallest network latency. 3-node brigade （regiment） level network centric warfare （NCW） demonstration platform was studied and the latency between the detectors and weapon control stations was taken as an example. The algorithm of end-to-end network latency and link information in NCW was presented. The algorithm program based on Server/Client architecture was developed. The data transmission optimal link is one whose end-to-end latency is the smallest. This paper solves the key issue to determine the link whose end-to-end latency is the smallest in ethernet architecture. The study can be widely applied to determine the optimal link which is in the complex network environment of multiple service provision points.     

Study and application of reinforcement learning based on DAI in cooperative strategy of robot soccer

A dynamic cooperation model of multi-agent is established by combining reinforcement learning with distributed artificial intelligence(DAI),in which the concept of individual optimization loses its meaning because of the dependence of repayment on each agent itself and the choice of other agents.Utilizing the idea of DAI,the intellectual unit of each robot and the change of task and environment,each agent can make decisions independently and finish various complicated tasks by communication and reciprocation between each other.The method is superior to other reinforcement learning methods commonly used in the multi-agent system.It can improve the convergence velocity of reinforcement learning,decrease requirements of computer memory,and enhance the capability of computing and logical ratiocinating for agent.The result of a simulated robot soccer match proves that the proposed cooperative strategy is valid.     

Collision avoidance planning in multi-robot system based on improved artificial potential field and rules

For real-time and distributed features of multi-robot system,the strategy of combining the improved artificial potential field method and the rules based on priority is proposed to study the collision avoidance planning in multi-robot systems. The improved artificial potential field based on simulated annealing algorithm satisfactorily overcomes the drawbacks of traditional artificial potential field method,so that robots can find a local collision-free path in the complex environment. According to the movement vector trail of robots,collisions between robots can be detected,thereby the collision avoidance rules can be obtained. Coordination between robots by the priority based rules improves the real-time property of multi-robot system. The combination of these two methods can help a robot to find a collision-free path from a starting point to the goal quickly in an environment with many obstacles. The feasibility of the proposed method is validated in the VC-based simulated environment.     

Review of Simultaneous Localization and Mapping Technology in the Agricultural Environment

Simultaneous localization and mapping(SLAM) is one of the most attractive research hotspots in the field of robotics,and it is also a prerequisite for the autonomous navigation of robots.It can significantly improve the autonomous navigation ability of mobile robots and their adaptability to different application environments and contribute to the realization of real-time obstacle avoidance and dynamic path planning.Moreover,the application of SLAM technology has expanded from industrial product...     

A Cooperation-planning Model Based on Bilevel Programming Decision

This paper is based on a resource constrained active network project;the constraint of the local resource and the time constraint of the cooperation resource are considered simultaneously.And the respective benefit of the manager and cooperation partners is also considered simultaneously.And a cooperation planning model based on bilevel multi-objective programming is de- signed,according to the due time and total cost.And an extended CNP based on the permitted range for resource and time requests is presented.A larger task set in scheduling cycle is on the permitting for the request of cooperation resource and time while the task manager itself may be permitted biding for tasks.As a result,the optimization space for the cooperation planning is enlarged.So not every bidding task is successfully bid by invitee,and the task manager itself takes on some bidding tasks.Finally,the genetic algorithm is given and the validity and feasibility of the model is proved by a case.     

Dynamic simulation of hydrodynamic model of drum level wave action and sloshing

In order to build the model of the drum level wave action and sloshing, based on the method of modularization modeling, the hydrodynamic model of drum level wave action and sloshing was developed, and dynamic simulation researches were carried out based on the model. The results indicate that both drum level and drum length have functional relations with period of drum level wave action and sloshing. When the drum level decreases or drum length increases, the period of drum level wave action and sloshing increases, density of liquid and number of sub-module division have little influence on the period of drum level wave action and sloshing. The model was validated by the analytical solution theory of liquid’s wave action and sloshing in cuboid container, and the 3D graphics of drum level wave action and sloshing was also obtained. The model can dynamically reflect the rules of wave action and sloshing of water in the container exactly.     

Application of annexation principle to the study of thermodynamic properties of Ag-Bi-In metallic melts

Based on the measured activities, the phase diagrams and the annexation principle, the calculating models of mass action concentrations for Ag-Bi and Ag-Bi-In melts have been formulated. The calculated results agree with practice and obey the mass action law, showing that the models formulated can reflect the structural characteristics of both melts. Meanwhile, it confirms that annexation principle is applicable to the Ag-Bi-In metallic melts. The melts involving eulectic which give rise to phase separation, and in which activities exhibit positive deviation from Raoult's law is the basic cause of melts transforming from homogeneous to heterogeneous ones.     

Discrete-event modeling for internet multi-robotics

Internet multi-robotics is a typical discrete-event system. In order to describe joint activities between multiple operators and multiple robots, a 4-level discrete-event model is proposed in this paper based on the controlled condition/event Petri nets (CCEP). On the first or mission level, the task splitting of the system is defined; on the second or multi-operator level, a precedence graph is introduced for every operator to plan his or her robotic actions; on the third or coordination level, the above precedence graphs are translated and integrated into the corresponding CCEPs in terms of specific rules; and on the last or multi-robot level, operators can select their control range by setting the corresponding control marks of the obtained CCEPs. As a consequence, a clear mechanism of operator-robot collaboration is obtained to conduct the development of the system.     

Dynamic Region Updating Strategy in HLA-Based Distributed Interactive Simulation

The HLA(high level architecture)-based distributed interactive simulation uses interest management mechanism to reduce the traffic on network and improve the system scalability. Making region updating occur only when needed can improve the interest management. Typically a static threshold is defined before simulation to trigger the region updating. Now a dynamic threshold is used to trigger region updating , the threshold is adapted by the real-time massage in simulation, named as update lifetime. The result of experiment shows that this policy can overcome the weak point of static threshold and can meet the requirements of bandwidth and simulation correctness.     

Development of a multiple level underground limestone mine from geology through mine planning

The development of a multiple level underground mine is a complex task in which geology, engineering,ground control, and unit operations are integrated into a single mine design. The components are interdependent and must function cohesively to ensure a profitable underground mining operation. To optimize reserve recovery, mine planning should begin from the lowest level and progress up. This limits any misjudgments or oversights of a given level affecting the underlying levels and ensures the ability to maximize recovery from each level. Mine planning should start with the exploration and characterization of the geologic setting. Once the reserve geology and quality is well understood, then mine planning can begin with respect to the following:(1) orientation of mine works with respect to horizontal stress;(2)access to the reserve;(3) determination of opening widths;(4) selection of back, floor horizons and pillar centers;(5) selection of development and secondary mining heights;(6) appropriate inter-burden thicknesses; and(7) examining the stability of the multiple level mine through numerical modeling. The multiple level mine design process and decisions are presented through a case history example. The theme is that there is one opportunity to ‘‘get it right" and many chances to overlook a small aspect within the design that will plague the mine throughout all levels and through the entirety of its operating life.     

Algorithm to Form Coalition in Multi-Agent Cooperation

In multi-agent systems, autonomous agents may form coalition to increase the efficiency of problem solving. But the current coalition algorithm is very complex, and cannot satisfy the condition of optimality and stableness simultaneously. To solve the problem, an algorithm that uses the mechanism of distribution according to work for coalition formation is presented, which can achieve global optimal and stable solution in subadditive task oriented domains. The validity of the algorithm is demonstrated by both experiments and theory.     

Kinematics/Fuzzy Logic Combined Controller for Formation Control of Mobile Robots

A kinematics and fuzzy logic combined formation controller was proposed for leader-follower based formation control using backstepping method in order to accommodate the dynamics of the robot. The kinematics controller generates desired linear and angular velocities for follower robots, which make the configuration of follower robots coverage to the desired. The fuzzy logic controller takes dynamics of the leader and followers into consideration, which is built upon Mamdani fuzzy model. The force and torque acting on robots are described as linguistic variables and also 25 if-then rules are designed. In addition, the fuzzy logic controller adopts the Centroid of Area method as defuzzification strategy and makes robots＇ actual velocities converge to the expected which is generated by the kinematics controller. The innovation of the kinematics and fuzzy logic combined formation controller presented in the paper is that the perfect velocity tracking assumption is removed and real- time performance of the system is improved. Compared with traditional torque-computed controller, the velocity error convergence time in case of the proposed method is shorter than traditional torque-computed controller. The simulation results validate that the proposed controller can drive robot members to form the desired formation and formation tracking errors which can coverage to a neighborhood of the origin. Additionally, the simulations also show that the proposed method has better velocity convergence performance than traditional torque-computed method.     

Intelligent Control and Maintenance Management Integrated System Based on Multi-Agents for Coal-Preparation Plant

This paper discusses the progress of computer integrated processing （CIPS） of coal-preparation and then preserits an intelligence controlled production-process, device-maintenance and production-management system of coal- preparation based on multi-agents （IICMMS-CP）. The construction of the IICMMS-CP, the distributed network control system based on live intelligence control stations and the strategy of implementing distributed intelligence control system are studied in order to overcome the disadvantages brought about by the wide use of the PLC system by coaipreparation plants. The software frame, based on a Multi-Agent Intelligence Control and Maintenance Management integrated system, is studied and the implemention methods of IICMMS-CP are discussed. The characteristics of distributed architecture, cooperation and parallel computing meet the needs of integrated control of coal-preparation plants with large-scale spatial production distribution, densely-related processes and complex systems. Its application further improves the reliability and precision of process control, accuracy of fault identification and intelligence of production adjustment, establishes a technical basis for system integration and flexible production. The main function of the system has been tested in a coal-preparation plant to good effect in stabilizing product quality, improving efficiency and reducing consumption.     

Windows and Fieldbus Based Software Computer Numerical Control System

Computer numerical control(CNC)system is the base of modern digital and intelligent manufacturing technolo- gy.And opened its architecture and constituted based on PC and Windows operating system(OS)is the main trend of CNC sys- tem.However,even if the highest system priority is used in user mode,real-time capability of Windows(2000,NT,XP)for applications is not guaranteed.By using a device driver,which is running in kernel mode,the real time performance of Windows can be enhanced greatly.The acknowledgment performance of Windows to peripheral interrupts was evaluated.Harmonized with an intelligent real-time serial communication bus(RTSB),strict real-time performance can be achieved in Windows platform. An opened architecture software CNC system which is hardware independence is proposed based on PC and RTSB.A numerical control real time kernel(NCRTK),which is implemented as a device driver on Windows,is used to perform the NC tasks.Tasks are divided into real-time and non real-time.Real-time task is running in kernel mode and non real-time task is running in user mode.Data are exchanged between kernel and user mode by DMA and Windows Messages.     

A new model of foraging behavior in ant system

Foraging behavior in ant colonies has come to be viewed as a prototypical example to describe how complex group behavior can arise from simple individuals. In order to research the feature of self-organization in swarm intelligence (SI), a mean field model is given and analyzed in foraging process with three sources in this paper. The distance of trails and the richness of each source are considered. Both of the theoretical numerical analysis and Monte Carlo simulation show the power law relationship between the completion time and the flux of foragers. The work presented here guides a better understanding on self-organization and swarm intelligence. It can be used to design more efficient, adaptive, and reliable intelligent systems.     

Effect of deformation and cooling rate on the transformation behavior and microstructure of X70 steels

The effects of the deformation in the non-recrystallization region of austenite and the cooling rate on the transformation behavior and microstructure of low-carbon low-alloy steel for pipeline application were studied on the thermal-mechanical simulator Gleeble-1500. It was shown that an increase in deformation amount can greatly increase the nucleation site of ferrite when deformed in the non-recrystallization region of austenite, and an increase in nucleation ratio can greatly refine grains. When the cooling rate is accelerated, the driving force of nucleation is increased and the nucleation rate also improves. Ultra-refine grains can be obtained by controlled rolling. The high density of ferrite nucleus, which forms along the austenite grain boundary, twin interface, and deforma- tion band are introduced in the matrix of austenite by the control of hot rolling, after which the microstructure can be refined. It was found that the acicular ferrite has a very fine sub-structure, high dislocation density, and a thin slab with ultra-fine grains. Small M/A islands and cementite are precipitated on the matrix of the slabs by the analysis technique of TEM and SEM.