Autonomous Navigation Of Unmanned Ground Vehicles In Campus Scenes

Unmanned ground vehicle, as a set of automatic control, artificial intelligence, image processing and many other technology as one, is a product of the high development of computer science, pattern recognition and intelligent control, and has extensive application prospect in field of national defense and economy. Autonomous navigation of the unmanned intelligent vehicle is the core technology to realize the autonomous driving, and is the precondition of achieving other advanced tasks. According to the driving condition of intelligent vehicle and the trust degree of the GPS data, this paper devises the corresponding autonomous navigation strategy which makes the intelligent vehicle can complete autonomous navigation in unstructured campus scene effectively.This paper puts forward an autonomous algorithm based on laser scan line which is applied to the situation that the vehicle is stop and the GPS data distortion is high. According to the distance information returned from the laser data, objects in safe distance are regarded as obstacles and segment information about the obstacles in front of the vehicle is obtained which will be used to find feasible direction. In this paper, the navigation algorithm is proposed aiming at the impact of dynamic obstacles on the movement trend of the intelligent vehicle. By forecasting different conditions that dynamic obstacles meet the vehicle, the original navigation algorithm is updated respectively and consequently the vehicle could avoid in advance.This paper puts forward an autonomous algorithm based on local grid map which is applied to the situation that the vehicle is normal driving. three lasers scanning ground in order to gain the information of ground height and identify feasible grid and unfeasible grid, the algorithm determines whether it is a feasible region or not and converts the question to find a solution of optimal direction by establishing navigation evaluation function. In addition, given the particularity of the road environment, the navigation strategy dedicated for road environment is designed based on the above navigation algorithm, which can guide the vehicle run by certain traffic regulations via detecting the lane. The results of experiments show that the method in this paper enforces the autonomous navigation of unmanned intelligent vehicle effectively.