Design And Implementation Of On-board Diagnosis Systems For Car Networks

With the rapid growth of the total amount of domestic vehicles, people pay more attention to the road information and vehicle maintenance. Car users need a smart, safe, comfortable driving environment. Nowadays, automotive electronics develops rapidly and car networks can meet the above demands of car users. The equipments to output internal automobile data through on-board automotive diagnosis interfaces are considered the first selected smart terminals of car networks.Since most vehicle terminals are expensive and ineffective, this thesis aims at the implementation of on-board automotive diagnosis systems for car networks. The main research results are listed as following:(1) The key circuit of the terminal module is designed. By comparing the performance,cost and other aspects of the main chips, STM32 is selected as the main chip. To satisfy the demand for low power, the detection circuit is proposed to wake up the terminal from standby mode. 3G module is designed to realize the communication between the terminal and the server.The CAN interface circuit and power circuit and other key circuits are designed.(2) The software of the vehicle terminal and the server are implemented. With CAN bus and K-line technology, the vehicle terminal program complete the collection of vehicle data. The main chip reads GPS integrated in 3G module to collect the GPS data. The collected real-time data are transmited to the server by 3G module. Combining C/S structure and B/S structure, the socket connection is embedded in J2 EE framework, which monitors the terminal data. The data are stored in database, and displayed on the WEB page.(3) With automobile data, driving behaviors are analyzed. AdaBoost algorithm is used for the collected data analysis. The training sets with different sample number are classified, and the classification model of driving behaviors is obtained.