Trajectory Tracking of Inland Ships Based on Improved Kalman Filter

DOI: https://doi.org/10.62517/jbdc.202601328

Author(s)

Yiyang Wang, Jiapei Deng, Jian Wang

Affiliation(s)

Three Gorges Navigation Authority, Yichang, Hubei, China

Abstract

Aiming at trajectory jitter, target loss under occlusion and frequent ID switching of clustered ships in inland lock monitoring with fused AIS and vision data, this paper proposes an improved Kalman filter (IAKF) integrated with inertial damping motion constraints and speed-environment dual adaptive noise updating. Considering the low-speed large-inertia characteristic of inland vessels, a state transition matrix with velocity attenuation damping coefficient is reconstructed. The process and observation noise covariance matrices are dynamically tuned by real-time ship speed and water interference levels such as strong light, rain and fog. Comparative tests on open inland multi-source datasets cover four typical scenarios: clear weather, night glare, rain-fog and ship occlusion, comparing standard KF, traditional AKF and the proposed IAKF. Results show the IAKF cuts position RMSE by 26.8%, boosts trajectory smoothness by 41.2%, and lengthens occluded continuous tracking time by 43.7% versus standard KF. Featuring low computation and strong robustness in harsh navigation environments, the algorithm supports direct deployment for real-time trajectory monitoring of inland lock waterways.

Keywords

Inland Vessels; Trajectory Tracking; Kalman Filter; Inertial Damping; Adaptive Noise

References

[1] Zhang B W, Yu W G, Jia Y F, et al. Predicting vehicle trajectory via combination of model-based and data-driven methods using Kalman filter. Proceedings of the Institution of Mechanical Engineers, 2024, 238 (8): 2437-2450 [2] Liu X, Chen M, Zhang L, et al. Research on improving the ranging accuracy of ships with stereo vision through Kalman filter optimization. PLOS ONE, 2024, 19 (11): e0305714. [3] Zhang Y, Li X, Wang H, et al. Ship detection and tracking in inland waterways using improved YOLOv3 and Deep SORT. Symmetry, 2021, 13 (2): 308. [4] Li X, Wang Y. Robust inland waterway ship tracking via hybrid TLD and Kalman filter. Advanced Materials Research, 2014, 1037:373-377. [5] Sun Bei. Research and system design of vehicle trajectory prediction method based on Kalman filter. Transportation Science and Technology and Management, 2025, 6 (10): 19-21. [6] Ding Bingchao, Wang Liyong, Su Qinghua, et al. Trajectory tracking control algorithm of MPC vehicle based on adaptive feedback. Sensors and Microsystems, 2024, 43 (12): 150154. [7] Jiang Baichen, Key, Zhou Wei, et al. Ship trajectory prediction algorithm based on polynomial Kalman filter. Signal Processing, 2019, 35 (5): 741-746. [8] Wang Qiang, Song Wei, Hua Zhongwei, et al. Research on ship trajectory prediction based on IMM Kalman filter. Sensors and Microsystems, 2025, 44 (10): 48-51. [9] Wei Zhaokun, Zhou Kang, Wei Ming, et al. Ship motion pattern recognition and application based on AIS data. Journal of Shanghai Maritime University, 2016, 37 (2): 17-22 + 71. [10]Qiao Shaojie, Han Nan, Zhu Xinwen, et al. Dynamic trajectory prediction algorithm based on Kalman filter. Chinese Journal of Electronics, 2018, 46 (2): 418-423. [11]Wang Yuanxue, Zhang Qianbo, Zhou Yuanyuan, et al. Indoor positioning system based on unscented Kalman filter. Internet of Things Technology, 2022, 12 (7): 18-19. [12]Gao Xi. Trajectory anomaly detection method based on Kalman filter prediction. Ship Science and Technology, 2022, 44 (14): 73-76.