Study On The Driving Condition Supervision Based On The Multi-Sensor Information Fusion

Preventing traffic accident is always the emphasis of the traffic safety works. With the fast development of economy, the number of vehicle increases dramatically and the influence caused by traffic accident is becoming more serious. When analyzing the cause of traffic accident, most of overseas and domestic experts think that artificial factor is the most important factor in the three factors of human, vehicle and road. In order to assure safety traveling and improve the security of vehicle, it is urgent to develop and perfect general intelligent monitor for driving state of drivers and state of vehicle. It is also urgent to research safe monitor and control system for vehicle of pre-alarm and prevention system. The research of old accident prevention system focused on sending the information of sensor to lonely disposal system. Since nineties of the twentieth century, information processing technology has changed from single sensor processing to multi-sensor processing and sensor information fusion technology has become one aspect of sensor technology development, so multi-sensor information fusion technology becomes the hot spot of research in recent years. Based on the theory of multi-sensor information fusion technology and by capturing the information of whether the driver is in fatigue driving, corner of the steering wheel, stress level and driving speed, the paper can monitor the driver real-timely from the multi-sensor fusion signal. When the driver is in stress state or fatigue driving, the driver is prompted or warned, and the vehicle will be limited speed when necessary.The paper is part of "Intelligent Monitoring Device and Method for Vehicle Driving State" supported by Shanxi Province Science and Technology Industrialization Environment Construction Plan Project which is applied by author's tutor. The paper is also hardware improvement in the National Invention Patent "Fatigue Driving State Monitoring Device and Method" (ZL: 200610012623.6) which is applied by the author of the paper.Beginning from the research of vehicle's active and passive safety, the paper firstly compares and consults the advanced and effective safety prevention measures all over the world and applies the multi-sensor information fusion technology to the monitoring of the driver's state. Also,the paper deeply researches the multi-sensor information fusion technology. Combining the fusion method of multi-sensor fusion in the different hierarchy and levels, the paper applies the evidence theory to information fusion judgment for driver's state monitoring.Analyzing the characteristics of auto electronic power and the choosing principle of auto-loaded electronic product, the paper design the system power module aiming at the sensor characteristic used in the system. The paper researches the work principle of CCD sensor comprehensively and compares the advantages and disadvantages between CCD sensor and CMOS sensor and finally chooses NEC company's CCD image sensor to monitor the driver's state. The paper researches the work principle of acceleration sensor and compares different acceleration sensor and finally chooses the Shanghai Langshang company's LAM-TD-360 single axis sensor to monitor the steering wheel corner. The paper researches the work principle of piezoelectric transducer and the faintness signal's characteristic of human pulse, and finally chooses Hefei Huake Electronic Research Institute's HK-2000 series HK-2000A integrated pulse sensor to monitor's tense state. The paper also researches the Hall's wheel speed sensor and the work principle of Anti-lock Brake System (ABS). The paper does some technique analysis and experimental research for every kind of vehicle-used speed measurement Hall's components and chooses Micronus's 320A Hall's wheel speed sensor and does some experimental research. The paper applies the evidence theory of multi-sensor information fusion to multi-sensor information fusion examples for monitoring the driver's state and proves that it is valid in the monitoring system by MATLAB calculation.