Through the five senses (ears, eyes, mouth, nose and tongue) and skin, human beings feel auditory information, visual information, olfactory information, taste information and tactile information from the outside, which can be considered as a multimedia information processing system. Computer multimedia system only includes auditory and visual information (sound, image, graphics and text, etc.). ), and little or no other information (such as touch, smell and taste information). In the future, with the in-depth study of various sensory mechanisms and the progress of technology, there will be more perfect multimedia systems. The key technologies of multimedia are compression coding, communication, synchronization and network. Distributed optical fiber sensor is a kind of sensor network, which can monitor the time and position of the physical quantity change of the measured object as a whole. If the distributed optical fiber sensor, actuator, signal processing system, transmission system and control system are combined, an intelligent structure is formed. Intelligent structure can be divided into inherent type and integrated type. Intrinsic type is a structure composed of materials with sensing, execution and signal processing functions. The integrated structure, that is, the structure embedded with sensors, actuators and processing functions, is also called smart material. With the expansion of intelligent materials and structures from military applications to various civil fields, concepts such as intelligent building and intelligent robot structure have been established one after another.
Structural damage and assessment is an important application of integrated optical fiber sensor networks, and several methods have been developed. The most important parameter for structural state detection and active vibration control is structural strain. Strain sensors applying various principles have been developed. For example, when the high birefringence polarization-maintaining fiber is disturbed longitudinally, the change of relative phase delay of two orthogonal polarization modes is directly proportional to the strain relationship, and its strain measurement sensitivity is 0.06/(μ ε-cm); Michelson interferometric fiber optic sensor with strain measurement sensitivity of 6.5/(μ ε-cm); The Fabry-Perot interference sensor can measure the displacement of about 1000℃ with ordinary time-sensitive optical fiber, and can work at higher temperature with sapphire optical fiber. Similar timely fiber Fabry-Perot sensor array has been used to measure the strain of F- 15 fighter under flight simulation conditions and the strain of materials under liquid nitrogen temperature conditions. Other interferometers are also used to measure friction caused by fluid on the skin surface and external audio pressure. The double transverse elliptical core optical fiber sensor can be made into local type, so it can effectively suppress the interference caused by the vibration of the external environment of the probe. A sensing fiber with weighted sensitivity along the fiber length is developed, which can be used for structural vibration control through simplified control algorithm.
In intelligent structure, it is necessary to obtain the information of multiple discrete points or the distribution information of fields. Therefore, in the sensor information processing technology, multi-point sensor multiplexing technology and distributed sensing technology have been developed. Optical time domain reflectometer (OTDR) is the most commonly used technology. Due to the low spatial resolution of OTDR, a high resolution optical time domain reflectometer for photon counting is developed. Statistics show that 80% of the information that ordinary people get from the outside world comes from vision. How to generate stereoscopic images of large-scale complex virtual environment in real time is still an urgent problem in the current virtual reality (VR) research. Three indicators of virtual reality: real-time, immersion and interactivity. The so-called real-time means that the virtual reality system can change the virtual environment picture presented to the user in real time according to the user's current viewpoint position and line-of-sight direction, and produce audio-visual and tactile/tactile reactions in the user's ears and hands in real time. The so-called immersion means that the virtual environment perceived by users is three-dimensional and three-dimensional, and the perceived information is multi-channel. The so-called interactivity means that users can manipulate the objects in the simulation scene in the way they are used to in real life, and change their orientation, attributes or current motion state.
The existing virtual reality systems can be divided into three categories according to the hardware structure: helmet display is the earliest VR display, which uses helmets to isolate people's vision and hearing from the outside world and guide users to feel in the virtual environment. The resolution of helmet-mounted display reaches 1024×768, which can provide users with clear virtual scene pictures. Desktop VR system consists of stereoscopic display screen, 3D mouse and stereoscopic glasses, but the user's immersion is slightly poor. The most immersive VR system is a cave environment composed of multiple projection screens. There are at least 30 ~ 40 cave environments with multiple screens in the world.
Applying tracking technology to technical training and maintenance of productization: Olivia Catoni of Consoraio Pisa Riceche Institute in Italy reported "Maintenance System Based on Remote Operation". According to market research, complex machines continue to rely on the training of producers and the maintenance of products. Among them, training and maintenance are directly proportional to the geographical distance between producers and users, accounting for a certain proportion of the total cost. Because small and medium-sized enterprises in Europe are often unable to provide users with maintenance and training services (people need such as travel, communication, transportation, etc. ), MAESTRO plans to take this problem into consideration, taking Cabinqi International S.P.A and Betex A/S as the pilot companies to produce complex automobile chassis and liquid treatment equipment respectively. These two companies export the most products, and MAESTRO experimented with training and maintaining products through remote control. MAESTRO plan is divided into practical part, that is, to build a demonstration room, which consists of equipment and robot inspection system, and the virtual part includes equipment model and software system supporting operation. Connect the demonstration room with ISDN/ARM network. Users can check the image of the device in real time, and it has a joystick with 6 degrees of freedom. It can provide real-time TV images of the virtual model of the equipment, and remotely shoot the specified content through the CCD camera, so that the training and maintenance operation can be extended to the real environment. The virtual model is connected with a multimedia database including technical information. For example, operating instructions, or data describing the right or wrong situation of functions, input verbal commands through language discrimination to help users understand the details of the equipment and freely change the simulation environment. Shinsuke Saito and others reported the generation of virtual space based on knowledge synthesis.