GPS Based Map Matching in the Pseudorange Measurement Domain

S. Syed

Abstract:	Accurate vehicle location is important for various Intelligent Transportation System (ITS) applications such as route guidance and collaborative driving. The Global Positioning System (GPS) has become popular for locating vehicles for many ITS applications due to the 24- hour, all weather and free-of-charge availability. Although GPS-based vehicle navigation systems are accurate in open sky conditions, it is challenging to navigate vehicles in dense urban canyon conditions where GPS is affected by many issues, with the primary being availability. The other effect which severely degrades the accuracy of GPS-derived navigation solutions in such environments is multipath. High Sensitivity GPS (HS GPS) receivers, can increase the availability, but are affected by multipath and cross correlation due to weak signal tracking. These factors may hamper the use of GPS for vehicle positioning in urban canyons. This paper describes a map matching algorithm to tackle the problems discussed above with the goal to successfully navigate a vehicle in an urban canyon, amidst low signal availability and multipath conditions. The algorithm is based on tight integration of the map information with the GPS measurements. The novelty of the approach is the use of map information to eliminate "erroneous" pseudorange measurements. An optimal map matched solution is then computed using the outlier-free measurements from GPS. The algorithm identifies the road link using a Fuzzy Inference System (FIS). The input to the system comes from a HS GPS receiver, a low cost gyro and a digital map. The algorithm was tested in an urban canyon under low signal availability conditions. including fleet management systems, in-car telematics and Location Based Services (LBS) require vehicle position information in real time. Many vehicle location sensors are used to support ITS including GPS and Dead Reckoning (DR) sensors like gyros, speedometers and accelerometers. GPS is being widely used for many ITS applications due to its high accuracy, and free-of-charge availability (Basnayake and Lachapelle, 2003). However, it suffers from line-of-sight (LOS) issues that make it less effective in urban canyons. Conventional GPS receivers may track/acquire a lower number of satellites in such conditions, which may be insufficient to obtain a navigation solution. Other problems include higher Geometric Dilution of Precision (GDOP) and multipath. High Sensitivity GPS receivers (HS GPS) were developed to track and acquire weak signals and thus increase the satellite availability. However, HS GPS-derived positions in an urban canyon may still have errors due to interference effects, namely signal cross-correlation, multipath and echo-only signal tracking (MacGougan, 2003). Thus GPS alone cannot be used for vehicle navigation systems in an urban canyon (Leung et al., 2003). Vehicle navigation solution has some special properties because of particular constraints originating from the fact that the vehicles are mostly traveling on roads. Map matching is the process of imposing these constraints on the navigation solution. Conventionally, the navigation output from GPS (i.e. position and velocity) is used for The performance of the map matching algorithm was evaluated in terms of number of correct and false fix on the road map. It is found that the proposed algorithm gave 98% correct and no false fixes, showing the effectiveness of the algorithm in mitigating navigation errors and increase in the availability of reliable navigation solution.
Published in:	Proceedings of the 17th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2004) September 21 - 24, 2004 Long Beach Convention Center Long Beach, CA
Pages:	241 - 252
Cite this article:	Syed, S., "GPS Based Map Matching in the Pseudorange Measurement Domain," Proceedings of the 17th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2004), Long Beach, CA, September 2004, pp. 241-252.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In