A hybrid autonomous positioning system for public transportation

Transit systems commonly use traditional technologies such as Dead Reckoning (DR) and signpost systems for vehicle position determination. Unfortunately, these systems suffer from a number of drawbacks, which affect the operational efficiency of the transit system. To overcome the limitations of these conventional systems, a hybrid positioning system is developed in this paper. The integrated positioning system consists of a low-cost autonomous GPS system, supplemented by the already existing conventional DR/signpost systems. An optimal positioning solution is obtained using the Kalman filtering technique, which utilizes all the available sensor information. The biases of the conventional sensors are modelled as first order Gauss-Markov and random walk processes. In addition, the frequent bus stops are taken into consideration to estimate the gyro bias offset. To further increase the accuracy of the integrated system, the signposts are used as reference stations to correct for the positioning error. Two field trials were conducted to evaluate the performance of the integrated system: one in an open area, the other in an urban area with high-rise buildings. It is shown that the DR solution was rather poor but could be improved by frequent signpost updates. A GPS-only solution was not adequate, by itself, to aid or update the DR system in the urban area. The integrated system, GPS, DR, and Signpost, gave a reliable solution in the open area, with GPS outages for 100 seconds with maximum error equal 22.9 m and 22.6 m in the north and east directions, respectively. The downtown test results were not accurate compared with the simulated outages and its deviation from the road centre line was in range from 30 m to 50 m. However, taking the bus dimension into consideration, it is clear that the obtained positioning solution is still sufficiently precise.