In order to make intelligent vehicle trajectory tracking achieve lane-level accuracy requirements in the city, this paper takes two axle vehicle as research object, the prediction model of motion posture was derived. The driver's manipulate habits of ideal turning mode was analyzed. According to information of GIS and GPS, based on drivers' manipulate habits, the turning trajectory tracking control algorithm of lane level was puts forward. In order to improve the accuracy and reduce tracking error and vehicle swing, based on the vehicle dynamics constraints the algorithm was improved. Finally, through the analysis and comparison of driving turning trajectory data and actual turning intersection data, the effectiveness of algorithm was verified. In order to better display the lane level control effect, the trajectory of the turning control was shown on the GIS map.
[1]
Matthew J. Barth,et al.
Next-Generation Automated Vehicle Location Systems: Positioning at the Lane Level
,
2008,
IEEE Transactions on Intelligent Transportation Systems.
[2]
Yinsong Wang,et al.
Lane-Level Vehicle Trajectory Reckoning for Cooperative Vehicle-Infrastructure System
,
2012
.
[3]
Rafael Toledo-Moreo,et al.
Lane-Level Integrity Provision for Navigation and Map Matching With GNSS, Dead Reckoning, and Enhanced Maps
,
2010,
IEEE Transactions on Intelligent Transportation Systems.
[4]
Bo Cheng,et al.
Mimicking human driving behaviour for realistic simulation of traffic flow
,
2010,
Int. J. Simul. Process. Model..