Research On Personalized Indoor Navigation

With the development of the society, there are more and more buildings with large scale and complex internal structure, the demands for the indoor navigation service are also increasingly urgent. Existing indoor navigation systems are not practical due to the weakness of locating and navigation technology, and the lack of personalized service reduces their commercial value. To solve above problems, this thesis does some research on personalized indoor navigation, specific work has the following several aspects:(1)For there is no mature indoor map platform until now, an indoor map production scheme which can support locating and navigation is proposed. By reading the AutoCAD drawing of an indoor map, the proposed scheme can obtain the data of frame, route and graphics primitive in the map, and then draw the indoor map on the mobile device, the scheme is easy and feasible which can also support the follow-up navigation effectively.(2)In order to solve the weak-signal and multi-disturbance problem of indoor environment, this thesis proposes a technology scheme which is based on two-dimension barcode locating and combines the color landmark navigation and inertial navigation. After simple decoration for indoor scenes, such as two-dimension barcode and color mark, the scheme can achieve locating and navigation by using the camera and sensor built in the mobile device, so it is almost network-independent, and can solve the high-price and low-accuracy problem in the existing navigation systems primarily.(3)Regarding large-scale indoor shopping center as the research domain, a personalized navigation scheme is proposed. The scheme fuses the recommendation thought into routing algorithm, firstly, it makes use of user interest and preference information, adopts mixed recommendation strategy and achieves personalized recommendation which can fit user interest and preference,at the same time it solves the cold-start problem,this thesis also determines the related coefficient in the collaborative filtering algorithm by the experiment; Secondly, the scheme improves the traditional A* algorithm, user interest and preference information is used to improve the estimating function in A* algorithm, experimental result shows that the improved algorithm can cover more positions which users are interested in, so it can achieve the personalized service in indoor navigation.(4)Based on the above research, an integrated personalized indoor navigation prototype system based on iOS is designed and implemented. The system is built with C/S architecture, it can achieve indoor locating, position recommendation and personalized navigation, so it can verify the feasibility and effectiveness of the map drawing, locating and personalized navigation schemes.