Compensation Method of GPS Signal Delay Based on Factor Graph
Hangyu Chen, Jing Ma, Fang Ye · 2021
In recent years, unmanned ground vehicles (UGVs) have been widely used in urban environments. The satellite positioning system plays an irreplaceable role in UGVs. However, the tall buildings in cities may cause interference to the transmission of satellite signals, which results in positioning delays, thus reducing the positioning accuracy of UGVs. To address this issue, a compensation method for information delay based on factor graph model is proposed in this paper. In the factor graph model utilized, the corresponding GPS factor node will not be generated when the GPS signal delay occurs at the current fusion time. The subsequent factor graph model is directly generated by the combined navigation system of visual odometry (VO) and inertial measurement unit (IMU). When the delayed GPS signal arrives, the measurement data of VO and IMU are used to compensate the delayed GPS positioning information. The compensated GPS positioning information is used to generate an additional factor node at the moment when the GPS signal actually arrives. This additional factor node is connected into the factor graph model for data fusion so that the delayed GPS positioning information can still be used to correct the positioning error. The method proposed in this paper effectively reduces the positioning error while taking into account the real-time performance. Experimental simulations show that the proposed method can reduce the positioning error while GPS signal delay occurs.