There is a bottleneck of mobile robots positioning technologies for uncertain goals in complex field environment. Owing to the disturbance of the environment, the objects are hard to be located precisely by robot manipulator. Aiming at the positioning problem, binocular stereo vision system and positioning principle of the picking manipulator in virtual environment (VE) were proposed and expatiated upon; in addition, the manipulator positioning model was built in VE, and the manipulator positioning simulation system was developed by Microsoft Visual C++ 6.0; and the binocular stereo vision system platform with a three-coordinate guideway positioning was constructed for test; what’s more the error sources of vision positioning system was analyzed, and the camera system error was established; the mathematical model of experimental error and the camera calibration matching error were also found; with the developed robot manipulator positioning simulation software and vision system hardware, an experimental platform of positioning system was constructed, and using the platform, the stereo vision data was mapped to the manipulator and was guiding the accurate positioning in VE. Finally, experiment of positioning error compensation was carried out. Results of simulation in VE and the experiment showed that the vision positioning method was feasible for positioning in the field environment; it can be applied to control robot operation and to correct the positioning errors in real-time, especially to the long-range precision modelling and error compensation of robots.

A frame of the Web-based exercise machine customization and simulation system is proposed first, and the flow of work of the system is analyzed. Then the paper adopts the modeling technology Pro/E and Virtual Reality technology as the three-dimensional display technology to realize the online customization according to customers' taste, embeds Agent design system into the performance function of Virtual Machine to realize the three-dimensional dynamic simulation based on variable and real-time interaction, and proposes an user-based mathematical model for preference entropy to solve the multi- attribute decision problems with uncertain information and satisfy the design preference of users. The development of the system platform provides a new way, means and platform for meeting customers' small lot, great varieties and high quality needs. The system which is already used in some enterprises, has improved their core competencies and efficiency.

Aiming to the deficiency of an individual tailor design system difficultly provide a platform for a customer can intuitively browse product and quickly experience product by playing game on-line, this paper propose a product tailor design platform integrating tailor design and virtual fitness game based on network using virtual reality and artificial intelligence technology. Establishing mathematic description and solid models on product family were detailed introduced based on a product family structure being analyzed. The system architecture of product tailor design was put forward, which includes product tailor design and experiencing fitness game on-line subsystem, product configuration subsystem, intelligent graphic processing subsystem, data management subsystem. The workflow of tailor design and automatic configuration were analyzed. Interactively tailoring 3D product on-line and playing network fitness game in a virtual environment were realized using EON SDK. The activity process which a customer manipulates virtually a tailor product model to participate in a network fitness game on-line was illustrated by a running machine example.

Aiming to uncertain problem in the process of picking manipulator location and behavior decision-making in complex environment, an approach for picking manipulator location and behavior simulation integrating binocular stereo vision together virtual simulation technical was discussed. Firstly, the imaging of picking target and environment were gained by binocular stereo vision, secondly, the result data of picking target and environment after the imaging being handled were mapped into virtual environment, picking manipulator behavior decision-making and motion path planning were implemented by software simulation technical, picking manipulator accurate location was realized in virtual environment. Finally, the simulation motion data in virtual environment were transmitted to control motors at manipulator joint by USB interface, control motors drive manipulator moving automatically to the picking target position. The experimental platform was consisted of CCD stereo vision and simulation software, a location simulation system was developed by VC++ and EON SDK.

Network sports fitness game is the product combining computer technology and network technology with the fitness exercise and game. It shows the trend of development of fitness movement. The engine of 3D game encloses complex substrate 3DAPI programme, which greatly reduces the complexity of exploitation. This paper explains the structure and the main templets of game engine and describes the structure of client's software of network fitness game based on 3D game engine. It also studies and analyzes some key technology such as handling of game movement, scrip of client, game data traffic, implementation of music sound and mutual system. At last through an example of running sports game proves its high efficiency to use 3D game engine to explore games.

The agricultural mobile robots are unable to gain navigation knowledge effectively to make decision for their behaviors during the performance due to complexity and uncertainty of operating environment and object. Complex behaviors of the mobile robots were studied based on artificial intelligence and virtual reality technology. The virtual environment, rule-base and case-base knowledge were built, and the complex behavior of the mobile robots was classified into simple, specific and foregone behaviors. The robot was navigated using similar cases and uncertain reasoning for the behaviors which is uncertain and has no priori knowledge. Then a multi-agent behavior-decision system was built and sextuple function of the structure was defined. News and events-driven mechanism were set up to achieve robots behavior of fruit picking and simulation modeling in virtual environment.

Aiming at complexity, unknown and uncertainty of picking object of agricultural intelligence mobile robot, extracting behavior knowledge and modeling based on the robot was put forward to help them obtain information effectively during operation, thereby to make decision for their behaviors. Firstly, behavior was studied based on picking behavior of the robot in virtual environment. Propose a behavior and method of extracting knowledge in virtual environment those are based on the association rules and, classify and express the entities such as robots, fruit trees and litchi, etc. Secondly, knowledge bases and models were built for reasoning. Thirdly, put forward for the first time to behavior knowledge classifies based on rough sets systematically, and classify the behaviors into obstacle-avoidance, picking, reasoning and fusion behavior to reduce redundant knowledge. Finally, an example for reasoning and simulation of the behavior was given. It realized picking behavior by message and route mechanism.