Abstract: The number and orientation of light-emitting diodes (LED) in visible light positioning have a great impact on positioning performance. To circumvent the challenge, a photo diode (PD) assisted perspective-three-point (P3P) algorithm is proposed. The basic idea of the proposed algorithm is to utilize visual information to estimate the incidence angles of visible lights first. Then, estimating the distances between the LEDs and the receiver by typical P3P algorithms, we can obtain up to four solution sets for the distances. Finally, the desired solution set can be determined by strength information, and based on the estimated distances, the position of the receiver can be obtained. Simulation results show that the proposed algorithm can effectively reduce the positioning error caused by LED's tilt. To further verify the performance of the proposed algorithm, a visible light positioning platform is designed and implemented. Experimental results verity that the proposed algorithm can achieve indoor positioning using three LED point light sources when the LEDs tilt within 15°, and the positioning error is less than 20 cm.

Key words: visible light positioning, photo-diode, perspective-three-point algorithm, image sensor