星凸形多扩展目标跟踪中的传感器控制方法

Sensor control method for star-convex shape multiple extended target tracking

中文关键词  多扩展目标跟踪  随机超曲面模型  传感器控制  多伯努利滤波器

英文关键词  multiple extended target tracking  random hypersurface model  sensor control  multi-Bernoulli filter

中文摘要

      针对多扩展目标跟踪中的传感器控制问题, 本文基于有限集统计(FISST)理论与随机超曲面模型(RHM), 利 用多伯努利(MBer)滤波器提出有效的传感器控制策略. 首先, 文中给出多扩展目标跟踪中基于信息论联合目标形状 估计优化和目标运动状态估计优化的传感器控制方法的求解思路. 其次, 给出RHM容积卡尔曼高斯混合(GM)势均 衡多扩展目标多伯努利滤波算法的具体实现过程. 然后, 结合GM密度间的柯西施瓦兹(Cauchy-Schwarz)散度提出 相应的传感器控制决策方法. 此外, 详细推导了扩展目标势的后验期望(PENET)的GM实现, 并提出以GM–PENET 为评价函数的传感器控制方法. 最后, 通过构造随机星凸形多扩展目标的跟踪优化仿真实验验证了本文所提传感 器控制方法的有效性.

英文摘要

      Aiming at the sensor control in multiple extended target tracking, this paper proposes the effective sensor control strategies based on the finite set statistics (FISST) theory and random hypersurface model (RHM) by using multi- Bernoulli (MBer) filter. First, this paper presents the solution ideas of sensor control for joint target shape estimation optimization and target motion state estimation optimization based on the information theory in multiple extended target tracking. Subsequently, this paper gives the detailed implementation of the RHM cubature Kalman Gaussian mixture (GM) cardinality balanced multiple extended target multi-Bernoulli filter. Then, a sensor control decision is proposed through the Cauchy-Schwarz divergence between the GM densities. In addition, this paper derives the GM implementation of the posterior expected number of extended targets (PENET) in detail and proposes a sensor control method using GM–PENET as an evaluation function. Finally, the effectiveness of the proposed methods is verified by the tracking optimization simulations of multiple extended targets with random star-convex shape.