A Method of Road Recognition Based on LIDAR Echo Signal
Road recognition technology is a difficult area of unmanned vehicle for autonomous navigation. Combining with the practical application of driving environment, this paper propose a method of road recognition based on laser infrared radar (LIDAR) of the un
- PDF / 413,607 Bytes
- 10 Pages / 439.37 x 666.142 pts Page_size
- 33 Downloads / 199 Views
A Method of Road Recognition Based on LIDAR Echo Signal Qinzhen Zhang and Weidong Deng
Abstract Road recognition technology is a difficult area of unmanned vehicle for autonomous navigation. Combining with the practical application of driving environment, this paper propose a method of road recognition based on laser infrared radar (LIDAR) of the unmanned vehicle. Firstly a clustering analysis with the radar echo intensity data is processed. Then the road is divided into region of interest (ROI) and non-ROI combining with the radar distance information. Finally the road information is managed by the method of straights fitting curve to accomplish the road recognition in real time. The tests on running vehicle show that the proposed method of the road recognition can work efficiently in real traffic environment.
Keywords LIDAR Echo intensity Road recognition Navigation of unmanned vehicles
42.1 Introduction Road recognition, which is a major research content of unmanned vehicles for autonomous navigation, is a difficult and heat field in the study of unmanned vehicles environment perception. Generally road recognition is a main factor Q. Zhang (&) College of Information Engineering, Capital Normal University, Beijing 100048, China e-mail: [email protected] W. Deng (&) Beijing Aerospace Guanghua Electronics Technologies Limited Corporation, Beijing 100854, China e-mail: [email protected]
W. Lu et al. (eds.), Proceedings of the 2012 International Conference on Information Technology and Software Engineering, Lecture Notes in Electrical Engineering 210, DOI: 10.1007/978-3-642-34528-9_42, Springer-Verlag Berlin Heidelberg 2013
397
398
Q. Zhang and W. Deng
reflecting the level of intelligent vehicles. Recently many domestic and foreign scholars have made a lot of effects and developed different techniques of road recognition based on the computer vision. Those techniques are usually on the basis of different land line styles (continuous or discontinuous styles), various road models (2D, 3D, lines and curves) and different segmentation technology (Hough Transform, template match or neural network). Conclusively, the techniques of road recognition mainly include the recognition methods based on road models, the methods based on road feature and the method based on color images [1–5]. At present, there are few researches based on sensor in the study of road recognition. A method checking road border based on LIDAR was proposed by Yu in [6]. According to the certain border height between road region and non-region, the method take advantage of border line and collect border data depth by Kalman filter algorithm to detect the road border. The method has advantage in instantaneity since it adopts LIDAR as the sensor. However, because it is on the basis of altitude difference between road and off road, the method is difficult to detect road when there is no or little altitude difference. Based on mobile robot localization of LIDAR, the approach proposed by Wang [7] collects the LIDAR data to bu
Data Loading...