一种间断观测下的无人机地面目标跟踪方法

提出了一种基于贝叶斯滤波器和适航地图的跟踪算法(BF-Hmap),解决间断观测下的无人机地面目标跟踪问题.采用基于中心线和梯度图的快速道路提取算法生成目标适航地图,该图描述了地面目标与当前地形相关的运动能力.采用贝叶斯滤波器分两步在线更新目标状态的后验概率分布,即在预测阶段考虑适航地图对目标运动的约束,以及在估计阶段充分利用目标的所有观测信息.仿真结果表明,在间断观测下该算法对地面机动目标有着良好的跟踪效果.

A Ground Target Tracking Approach for UAVs with Intermittent Measurements

A tracking algorithm using Bayesian filter and hospitability map (BF-HMap) is proposed to deal with the ground target tracking problem for unmanned aerial vehicles (UAVs) with intermittent measurements. The hospitability map, which describes the terrain-dependent motion capability of ground targets, is built through a centerline-and-gradient-based road detection method. The posterior probability density function (PDF) of the target state is maintained on-line through a Bayesian filter in two stages: the prediction stage takes into account the motion constraint provided by hospitability map and the estimation stage considers of all detection information that both the target is detected or not. Simulations show that the proposed approach provides a feasible solution to tracking maneuvering targets with intermittent measurements.