The Research About Ultrasonic Measuring Technique Of Circum-Obstacle

Along with the development of mobile robot technology, in order to satisfy the artificial intelligence, the technology of obstacle detection has more and more attention because of the basement place in path planning.Combining with the inherent favorable characteristics of ultrasonic sensor, including simple structure, mature technology, high performance-to-price ratio and good at working in the conditions of dark beam and dirt oil, a kind of ultrasonic measuring technique for obstacle based on mode identification of neural network machine is developed. It cans separation, orientation, distance measuring and mode identification about obstacle. On basis of referring and analyzing a large quantity of literatures, the following work has been carried out.Aim at the characteristics of ultrasonic sensor, according to lots of experiment, finds that the wave angle and side wave have effect on obstacle's outline. On base of experiment, the technology of separation, orientation, distance measuring and mode identification about obstacle have been developed.The application of neural network to mode identification is discusses, and a neural network structure is designed for this system. Based on experiment, a lot of data have been collected, dispose and extracted feature. A program is designed for neural network by Matlab's nntool, and then writes the weights and biases of the NN into VB program to make on-line training, and makes PC has capability to mode identification.A machine about obstacle detection has been made; it's based on STC51 singlechip which can resist disturbance. It's made up of serial communication, motor control and sonar control. VB program is work to deal with and show data. The results show that the instrument is efficient and accurate, and run stably.