摘要
基于Jerk模型及当前统计模型,提出了一种多级修正的高机动Jerk模型(MJerk).在假设机动加速度变化率(即加加速度)为非0均值指数相关随机过程的条件下,通过对Jerk模型状态分量作Taylor级数展开,得到了各状态分量的Jerk修正方程,使得机动加加速度对系统各状态分量的作用得到反映,减小了模型误差.Monte Carlo数值仿真表明,在进行高机动目标跟踪时,MJerk模型的均方根误差比Jerk模型的约减小了20%,而二者的计算量相当.
Based on Jerk model and current statistical model, a multilayer modified high maneuvering Jerk model （MJerk） is derived for target tracking. In the situation that the rate of maneuvering acceleration variety （also named Jerk） is assumed to be an exponentially correlated random process with non-zero mean and the Taylor series expansion is performed on the components of state of the Jerk model, the modified differential equations of the components of state can be obtained, and the influence of the Jerk on the system state is taken into account reasonably, so the model error is reduced to some extent theoretically. The Monte Carlo simulations show that the root mean square （RMS） error of MJerk in maneuver target tracking is reduced by 20% compared with Jerk, meanwhile the computation complexity of the MJerk is nearly same as Jerk.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2006年第2期138-141,共4页
Journal of Xi'an Jiaotong University
基金
国家重点基础研究发展规划资助项目(2001CB309403)
