Loading [a11y]/accessibility-menu.js
Coding Pavement Lanes for Accurate Self-localization of Intelligent Vehicles | IEEE Conference Publication | IEEE Xplore

Coding Pavement Lanes for Accurate Self-localization of Intelligent Vehicles


Abstract:

Self-localization is a key technology for intelligent vehicles. This paper demonstrates a practical and easy solution to vehicle self-localization by simply coding paveme...Show More

Abstract:

Self-localization is a key technology for intelligent vehicles. This paper demonstrates a practical and easy solution to vehicle self-localization by simply coding pavement lane lines. Especially, the coding of pavement lane lines makes it possible to distinguish unique pavement marking within certain ranges, which is crucial for vehicle localization. Based on the coded pavement lane lines, we proposed a multi-scale strategy for accurate vehicle localization. The localization method consists of coarse localization with Real-time Locating Systems (RTLS), marking-level localization with marking matching, and metric localization by matching distinctive visual feature points around the marking area. The proposed method has been tested by using the actual data collected in the field, where we encoded the pavement lane lines with two different colors (i.e., white and yellow). The results demonstrate that the proposed method can achieve sub-meter localization accuracy by referring to the coded pavement lane lines. The results also demonstrate that advanced road infrastructure could greatly support intelligent vehicles with low-cost and reliable solutions.
Date of Conference: 26-30 June 2018
Date Added to IEEE Xplore: 21 October 2018
ISBN Information:
Print on Demand(PoD) ISSN: 1931-0587
Conference Location: Changshu, China

References

References is not available for this document.