Maritime moving target tracking is a core technology for autonomous navigation and situational awareness of unmanned surface vessels. LiDAR-based tracking in maritime environments faces three key challenges: clustering failure caused by sparse point clouds at long range,false detections from dynamic near-shore clutters,and velocity jumps induced by sensor noise.To address these issues,this paper proposes a LiDAR-based maritime moving target tracking method integrating distance-adaptive clustering with multiple constraints.First,exploiting the inverse-square law of LiDAR point cloud density with respect to range,a distance-adaptive density-based spatial clustering of applications with noise (DBSCAN) algorithm is designed,in which the neighborhood radius grows linearly and the minimum core-point count decays exponentially with distance,resolving clustering failure under sparse long-range point clouds.Second,a multimodal false-detection suppression mechanism is introduced,integrating motion direction consistency,velocity fluctuation analysis,position jitter detection,and spatial isolation assessment to accurately filter dynamic near-shore interference.Third,a triple velocity constraint framework combined with a dynamic track-revocation mechanism monitors velocity rationality at instantaneous,raw,and average timescales,suppressing abnormal velocity estimates caused by sensor noise.Finally,a distance-adaptive confirmation strategy is proposed,applying differentiated confirmation conditions for near- and long-range targets to balance precision and detection sensitivity.Experiments conducted in two real-world scenarios—a near-shore harbor and an open-water environment—demonstrate that the proposed method achieves a multi-object tracking accuracy (MOTA) of 98.56%, precision of 100%,and multi-object tracking precision (MOTP) of 0.41 m with zero false detections in the near-shore scenario,and maintains a MOTA of 94.89% and Precision of 100% with zero ID switches in the open-water scenario under ego-vessel motion.Ablation studies show that the distance-adaptive DBSCAN reduces false detections by 653,and the spatial isolation module reduces false detections by 477,with Precision reaching 100% in both scenarios,which validates the stability and adaptability of the proposed method in complex maritime environments.