Autonomous Vehicle Control in Urban Environment by Map-Based Driving Lane Detection

Highly-precise ego-localization and mapping techniques from the road shape features are key elements in order to realize an autonomous driving system for vehicle in urban area which has complex environments. The objective of this study is to develop an autonomous driving system based on mapping and ego-localization using a LIDAR. To handle curved path tracking scenario, this paper proposes a desired steering angle generator considering a constructed map using the LIDAR in real time combined with the feedback control of the preview lateral deviation. The effectiveness of the proposed control method is verified by simulation and test drives using the autonomous path tracking control system.