Research On Visual Intelligent Vehicle Based On Embedded System

Automatic driving technology has a wide application prospect, and is an important part of intelligent transportation. This paper aims at discussing the key technology of automatic driving though a smart car model, including:automatic obstacle avoidance technology, monocular vision measuring technology, speech recognition technology and path planning etc..The design established the smart car model of mobile platform based on embedded Linux operating system and completed vision measurement using pinhole imaging model. The car can detect the obstacle based on the different methods of filtering algorithm and ranging method. At the end, the system use the Opencv library functions to complete the initialization, image acquisition and transmission.In order to improve the intelligent vehicle detection range, the car combine the ultrasonic ranging, monocular vision measuring technology and heading data to complete extraction of environmental information around the smart car. On this basis, the system analysis the path planning method and use the fuzzy control method to realize the control of the intelligent vehicle running state, and then complete the automatic obstacle avoidance function. Finally, the speech recognition and playback technology is used in intelligent vehicle system and improve the convenience and the flexibility. The results of experiment show that the system can successfully completed the following tasks:1. The intelligent vehicle can operate under the control of voice commands, and complete the measurement of the direction of the obstacles.2. The system can be specified the direction of intelligent vehicle through voice commands, the smart car will run according to the direction, and avoid obstacles.3. Through the wireless network will collect video information and the distance information is transmitted to the PC monitoring terminal.