Photon-statistics-based IMM Kalman filtering method for moving targets detection using single-photon LiDAR

Single-photon LiDAR has the advantages of small size, weight, and power, high ranging accuracy, and long measurement distance for detecting and tracking moving targets, such as satellite laser ranging, rendezvous and docking, and so on. A photon-statistics-based interacting multi-model (IMM) Kalman filtering method is proposed to detect a moving target and estimate the distance and velocity of the target. Photon statistics is used to make the photon ranging error more convergent to a Gaussian distribution, and to obtain light intensity information used to compensate for walk error. IMM Kalman filtering method is used to further improve the estimation accuracy of ranging, and to estimate the velocity of the target. Simulations and experiments were designed to verify the robustness and accuracy of the proposed method. In simulations, the initial distance and moving velocity of the target were set as 3000 m and 10 m/s. Simulation results show that the walking error compensation method can effectively improve the ranging accuracy, which is better than 0.05 m The mean absolute error (MAE) and root mean square error (RMSE) of the distance were reduced by 25% and 22% after the IMM Kalman filtering. The MAE of the estimated velocity of the target is reduced by 69%, and the RMSE of it is reduced by 47% using the IMM Kalman filtering. In the experiment, a single-photon LiDAR was developed and the proposed method was used to detect a target performing reciprocating movements driven by a motorized translation stage. Experiments show that the method has excellent accuracy and robustness.