A novel road centerline recovery algorithm based on the network RTK GPS data

作者：Sun R.; Zhang Y.; Yang L.; Meng X.

来源：ICIC Express Letters, Part B: Applications , 2017, 8(1): 137-142.

• Curve fitting
• Network real-time kinematic (NRTK)
• Road centerline detection
• Smooth spline

Nowadays the requirement of performance for the Intelligent Transportation Systems (ITS) applications is increasing, especially for some safety critical applications, including lane control, collision avoidance and autonomous driving. The precise determination of road centerline underpins these applications. The Network Real-Time Kinematic (NRTK) Global Positioning System (GPS) positioning, which is able to provide centimetric positioning accuracy under an ideal observation condition, is possible for the road centerline determination. However, GPS signal obstruction can introduce gaps in the positioning trajectory. Sensor integration can certainly bridge these gaps but with a cost in terms of the complexity in data fusion and the actual cost in expenses. In this paper, a novel smooth spline based algorithm is developed to improve the accuracy and continuity of NRTK GPS solutions for the road centerline extraction and recovery. The simulation and field test results have shown that the proposed algorithm yields significant accuracy in the road centerline recovery compared to the other state-of-the-art algorithms, and remains an economically viable solution.

• 出版日期2017
• 单位The University of Nottingham