GPS Map Matching Technology Research And Its Application In Scenic Navigation

Recent years have witnessed a rapid growth of smart tourism. Products and applications have been brought out in the market. But so far as the status, these applications are not so good and can be improved a lot on both navigation experience and positioning precision. So Map matching technique is introduced. Map matching algorithm (MMA) has been widely applied in intelligent transportation systems, locating the vehicle on the road through the combination of positioning systems and digital road maps. Existing algorithms are mainly intended for vehicles and not suitable for pedestrians using low cost positioning devices because of the complex location environment, vulnerability of location devices and irregular movement of the user.As to the problems mentioned above, this research has made some improvement and optimization on electronic map, road segment identification and user location update, according to the characteristic of scenic navigation. Firstly, this paper propose an extended design of electronic map. Based on conventional map design, we consider the width of the road and introduce a new element type, rectangle road as well as its rule of map matching.Secondly, this research counts in the environment variation as a new contributor to location error based on probabilistic map matching method. Then a new method of error region estimation is presented and this method is applied in initial matching, intersection detection and intersection matching. Thirdly, a score based rule is draw out from the error region estimation and confidence on roads selection. This rule can be applied to determine the right road segment that the user is moving on with high confidence level. Dynamic location experiments demonstrates the validity and effectiveness of the proposed method. Compared with map matching in ITS as well as current mobile map applications, the proposed algorithm can effectively improve the accuracy of road selection and location, meanwhile improve map display.Finally, a three dimensional navigation system is developed with score based map matching algorithm integrated. The scene model is from the university campus which can be seen as a scenic spot and the road network is derived from the campus roads. Real experiments demonstrate validity, high efficiency and sensitivity. The proposed method is of great value in practice.