Bulletin of Surveying and Mapping ›› 2022, Vol. 0 ›› Issue (6): 6-11,75.doi: 10.13474/j.cnki.11-2246.2022.0163.

Previous Articles     Next Articles

An indoor parking lot vehicle positioning method assisted by environmental semantic information

ZHOU Baoding1,2, YANG Chengjing1,2, GU Zhining3,4, LIU Xu2,3   

  1. 1. Institute of Urban Smart Transportation & Safety Maintenance, Shenzhen University, Shenzhen 518060, China;
    2. College of Civil and Transportation Engineering, Shenzhen University, Shenzhen 518060, China;
    3. Guangdong Key Laboratory of Urban Informatics, Shenzhen University, Shenzhen 518060, China;
    4. School of Architecture & Urban Planning, Shenzhen University, Shenzhen 518060, China
  • Received:2021-07-07 Revised:2022-04-06 Published:2022-06-30

Abstract: Aiming at the problem that the GNSS signal is limited in the indoor parking lot environment and cannot be used for vehicle positioning, this paper proposes an indoor parking lot positioning method based on environmental semantic information. This method uses smartphone to identify the semantic information(bumps and turns) in the indoor parking lot by machine learing, establishes the topological structure of the indoor parking lot road network. Then, particle filter algorithm is used to fuse the traditional vehicle dead reckoning data. Finally, experimental results show that this method effectively eliminates the cumulative error of vehicle dead reckoning, with an average positioning accuracy of less than 3 m, and reduces the computational time of traditional particle filtering methods through semantic information matching.

Key words: vehicle positioning, smartphone, dead reckoning, environmental semantic information, particle filter

CLC Number: