Design Of Indoor Positioning System Based On Lighting Features

The rapid development of modern science and technology makes the navigation become more and more sophisticated and approach perfection day by day. The navigation technology has attracted attentions in both research and industry fields. One of the most widely used navigation system is the Global Positioning System developed by the U.S. Department of Defense. However, Global Positioning System(GPS) is not appropriate to be applied in mansions and skyscrapers with complicated architectural structures. On the one hand, it is not easy for the GPS users to maintain the range of visibility with the orbiting satellites. On the other hand, GPS signal hardly penetrates walls of modern architectures. Therefore, people frequently get lose in buildings with complicated internal structures. In this thesis, we proposed an indoor navigation system based on the lighting features. Our system is totally independent from the GPS system. It is mainly designed to perform navigation in the indoor areas that the GPS signals cannot cover, with high stability, accuracy, low cost and easy maintenance. Thus, the proposed indoor navigation is of a great value in the development of navigation technology. In summary, the contributions of the thesis are listed as follows:Firstly, we propose a novel system model for the proposed indoor navigation system based on lighting features. Firstly, the video of lighting fixtures is captured by the optical system, and the features of lighting are extracted by the image processing techniques. Then, the lightening features extracted are transmitted to a pre-established recognition library that stores the indoor-environment lighting features. After that, we design the indoor positioning and tacking algorithms to calculate the actual positions of the moving objects and predict the next positions of the objects.Secondly, we establish both the software development system and the hardware operation platform for the proposed indoor navigation system. Our software development environment is based on the Linux system. It is used for coding and realizing the proposed positioning and tracking algorithms. On the other hand, we choose to use the embedded ARM development board and free-moving car controlled by Micro Control Unit(MCU) as the hardware platform. The embedded board is used to complete the video capture function, and extracting location information of the target object, such as the current position and forward angle. Then, the background can send signal to the target object and further directs the object to the right direction in the proper time.Thirdly, we choose an appropriate indoor environment and lightening conditions to test the proposed navigation system. We also constructed a simple but efficient coordinate system for indoor navigation. The performance of the proposed system is tested in the real platform, and the results are evaluated and analyzed in a great detail. Simulation results show that the system can stably accomplish the positioning and target tracking functions with small error in positioning and high accuracy in tracking, as well as satisfy the requirements of system design at the same time.