Design of a Artificial Neural Network for wall following behavior of Automatic Mobile Vehicle

• Main Authors: CHIEH-LI HUANG, 黃介力
• Other Authors: Ten-Min Lee
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/21709703990941442151

碩士 === 明道大學 === 資訊工程學系碩士班 === 98 === Mainly discusses the use of Artificial Neural Network in the Supervised learning control of automatic mobile vehicle, and after automatic mobile vehicle study under the environment , it can follow the wall by itself. In this study, automatic mobile vehicle (U-Bot) on the eight sets of infrared rays capture environmental information, stored sensor value and the motor-driven values as input values and the corresponding response action, and then, changes its values into the Artificial Neural Network model of Multilayer Perceptron(MLP) to learn and to change the weight values, Finally, input the algorithm programs and weight values to the automatic mobile vehicle. After observing its study, whether to use the response which infrared receives to control the motor in the environment along the wall to walk smoothly. Therefore, the experiment aims to explore the possibility that the machine is the same as humans, that can achieve successful control of cars direction by studying in similar circumstances. In the research results show that, after the Artificial Neural Network training, the automatic mobile vehicle can successfully control and follow the wall according to the output value, and complete the task in the similar environment. Expected that the theory can still run when it transferred to large vehicles.