Indoor 3D Visible Light Positioning Based on Multiple Cameras: Algorithm Design and Error Analysis
Jiaqi He, Nuo Huang, Chen Gong
Abstract
In this paper, a novel multiple-camera-based visible light positioning (MC-VLP) algorithm is proposed for indoor three-dimensional (3D) positioning. The basic idea of MC-VLP is to utilize multiple cameras to simultaneously capture the beacon signal transmitted by a light-emitting diode (LED) mounted on the mobile terminal. Since the pinhole camera can achieve high imaging accuracy and the fish-eye camera has a wide field of view (FOV), both of them can be adopted as the receiver to improve the positioning accuracy and coverage area. Meanwhile, the positioning errors along <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$X$ </tex-math></inline-formula> -axis, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Y$ </tex-math></inline-formula> -axis, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Z$ </tex-math></inline-formula> -axis are derived to evaluate the performance of the MC-VLP algorithm. Based on the derived expression of positioning error, we optimize the camera layout and tilt angle to minimize the positioning error in the coverage area. Simulation results show that higher positioning accuracy can be achieved via optimizing the camera layout and tilt angle. Experimental results show that the mean 3D positioning accuracy is 1.7 mm with 3 pinhole cameras at height 2.5 m.