On-board validation of BDS-3 autonomous navigation using inter-satellite link observations

The third generation of the BeiDou navigation satellite system (BDS-3) has accomplished a complete constellation with 30 BDS-3 satellites in 2020. All of the BDS-3 satellites are equipped with Ka-band inter-satellite link (ISL) payloads, which have the capability for inter-satellite measurement and communication. With the support of ISL, autonomous navigation for BDS-3 satellites can be achieved. In this contribution, we develop an on-board centralized autonomous navigation processing schedule for BDS-3 satellites. Unlike traditional batching mode, we adopt an extended Kalman filter (EKF) to estimate the satellite's orbit and clock parameters using two independent filters to ensure real-time performance, in each of which we simultaneously estimate all satellites' orbit or clock parameters in a centralized processing mode. From the view of engineering, we also optimize the algorithm to reduce the computing load and resource usage of the on-board satellite's processor. The proposed approach is validated using ISL observations in an 85-day period from 18 BDS-3 satellites, and a comprehensive evaluation was conducted from orbit accuracy, clock accuracy, and positioning performances. The results demonstrate that an average root-mean-square (RMS) of 0.60-m orbit-only user range error (URE) can be achieved, and the average RMS of 2-h predicted orbit URE is 0.65 m. For time synchronization, the average RMS of clock errors is better than 0.3 ns, and the average RMS value of 2-h predicted clock errors is better than 0.6 ns. The standard point positioning (SPP) results show that the estimated orbit and clock by autonomous navigation can meet the requirements of meter-level navigation and positioning for more than 60 days without the support of the ground OCS.