The Study on Hybrid Intelligent Automatic Guided Vehicles

Abstract

The purpose of this paper is to study the hybrid intelligent automatic guided vehicles (HIAGV), which guidance and navigation method is composed of vehicle dynamic motion model and mentor model from driving expertise. The vehicles are navigating in a beeline by dynamic motion model, which derived from the turning angle of front wheels of vehicle (δ ) with a fixed speed. But, while vehicles are changing a direction or a curve, it’s attitude is varied a large-scale range, could be leaded to exceed out the lane boundary unless adjust the vehicle speed appropriately. A vision-based vehicle dynamic behavior acquisition algorithm (VVDBA) had been developed to capture δ in our previous study. In this paper, we used this achievement to derive the dynamic motion model and incorporated with a fuzzy inference system to guide the vehicle navigating along the lane smoothly. The captured δ from VVDBA was compared by a threshold value to find more efficient ways of utilizing hybrid automatic guided strategy. If δ is below the threshold value, vehicles are guided by dynamic motion model. Otherwise, HIAGV is using the fuzzy inference system to handle the speed of vehicle that can be moving smoothly and avoiding exceed out the lane boundary. HIVAG was simulated by MTALAB in this paper, and it would obtain robust and precise tracking capabilities along any arbitrary lane.