Assessment of GNSS and Map Integration for Lane-Level Applications in the Scope of Intelligent Transportation Location Based Services (ITLBS)
To enable safe and robust Intelligent Transportation Systems (ITS) applications, the integration of different sensors and techniques will certainly be a common reality. One application in this context is the lane-keeping techniques for autonomous driving systems. These systems normally use imagery sensors for lane identification, however imagery systems always depend on light and well-structured roads. One potential worldwide autonomous driving technique without any other lane and road detection/identification sensor would be GNSS positions along with accurate map information. However, this fusion depends on the accuracy and reliability of both GNSS positions and map information. The positioning accuracy that Intelligent Transportation Location Based Services (ITLBS) requires for where-in-lane and active control applications are 0.5 m and 0.1 m, respectively. To evaluate the potential of fusion, this work proposes an integration of GNSS and map information in the attempt to address the lane-keeping problem. This integration is performed by merging a GNSS solutions and lane centerline positions, acquired from aerial orthophotos, into a Kalman Filter and a simple map matching approach. To measure the positioning error, or off-track performance, a conversion of positions to the road space is necessary. To evaluate the results, a positioning accuracy limit, considering the road, vehicle dimensions, and the requirements for ITLBS is also proposed. The results showed that 95% of the time the proposed methodology off-track performances were within 1.89 m, in an average of 4 runs. Half of the runs were within 0.75 m, in average, at 95% of the time. Compared to an accurate GNSS Post Processed Kinematic (PPK) mode, an improvement of 10% was achieved.
KeywordsGNSS ITLBS Lane-level positioning Maps
To the CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) agency, which through the Brazilian Sciences without borders program provided the necessary funds for the development of this research.
- Bishop R (2005) Intelligent vehicle technology and trends. Artech House, BostonGoogle Scholar
- Grover R, Hwang PYC (1992) Introduction to random signals and applied Kalman filtering. Wiley, New YorkGoogle Scholar
- Li Q, Chen L, Li M, Shaw SL, Nuchter A (2014) A sensor-fusion drivable-region and lane-detection system for autonomous vehicle navigation in challenging road scenarios. 63(2):540–555Google Scholar
- Miller KS, Leskiw DM (1987) An introduction to Kalman Filtering with applications. Riverside Research Institute, New YorkGoogle Scholar
- Ristic B, Arulampalam S, Gordon N (2004) Beyond the Kalman filter: particle filters for tracking applications. Artech House, BostonGoogle Scholar
- Stephenson S, Meng X, Moore T, Baxendale A, Ford T (2011) Accuracy requirements and benchmarking position solutions for intelligent transportation location based services. In: Proceedings of the 8th international symposium on location-based servicesGoogle Scholar
- Takasu, T (2018) RTKLIB: An open source program package for GNSS positioning (2018-07-11). http://www.rtklib.com