基于激光雷达的军用智能车

AbstractAbstractMilitary intelligent vehicle research is the one of the main development ofmilitary vehicles.It has important military and economic potential,pursued hotlyby each developed country of the race.In the past few decades,intelligent vehicletechnology has made considerable progress,but there are still many limitations inpath planning.This article is a preliminary exploration of these issues.In the past path planning of the structure,the methods of obtainingenvironmental information by researchers is to identify a clear road signs and drivelines,but it is often face to how to solve the problem of road recognition in theabsence of drive lines and in the failure of the camera.On the other hand,becauseof the complex road environment and limited computing resources,there must bean effective way to achieve real-time information on path planning.How todistinguish the validity of the information in different regions around vehicles,andassign a different degree of concern,in order to decrease the computation time andensure the important regional focus in path planning.It is another major problemof this paper.To solve these problems,this paper carried out the following:Firstly,the related Domestic and foreign path planning algorithm is analysisedand The region search algorithm based on voting method is improved.Thealgorithm get the radar scan data into a straight line fitting the road,and identifyThe region judging by whether the data is in the straight line.Secondly,variable-scale grid method is proposed in structure of the road.Themethod focuses on the relationship between the size of the grid and the vehiclelocation,relative distance,vehicle performance.Based on the comparisons andexperiments,a computational grid size method is proposed independently.It has aII