Aiming at the problem that in indoor positioning process

light source signal is vulnerable to be occluded

the accumulated error of pedestrian dead reckoning(PDR)

and the traditional combined positioning method positioning can only rely on PDR when the light source signal is occluded

which leads to low positioning accuracy

a combined visible light communication(VLC) and PDR indoor positioning method is proposed.The method combines visible light fingerprint positioning with PDR through a factor graph-based data fusion algorithm to improve the localization precision and suppress the PDR error

and at the same time

it estimates and compensates the error of heading parameters and unit conversion constants in PDR.Faced with the situation where the light source signal is shielded

it utilizes the error-compensated PDR for localization to effectively inhibit the growth rate of the error.The experimental results show that compared with the extended Kalman filter(EKF) combination method

this method increased the positioning accuracy by 25.9% in the area with sufficient light source

and 51.1% in the area where the simulated light source is obscured.