一种基于加速度预估计的机动目标跟踪算法

Algorithm for Maneuvering Target Tracking with Acceleration Pre-Estimation

摘要:

研究分析了几种典型单机动目标模型的建模方法,针对现有单机动目标模型中机动参数需要先验假设,并且不能随目标机动情况的改变而自适应调整的问题,提出了一种加速度预估计模型(Acceleration Pre-estimation Model,APM)。该模型首先用位置量测对机动加速度进行预估计；然后,将加速度估计值作为系统的输入控制项建模；将估计误差看做系统的机动控制项,并作为系统的相关噪声建模。由于 APM模型中,加速度机动参数是通过位置量测实时估计得到的,不需要先验假设。与现有单机动目标模型相比,该模型的自适应能力得到了提高。

Abstract:

Some kinds of typical maneuvering model were analyzed. Due to the problem of un-adaptive adjustment of targetmodel, which needed a priori hypothesis with the change of target maneuver condition, a novel acceleration pre-estimationmodel was proposed. Firstly, the pre-estimation of acceleration was gained by position measurements. Then, the accelera.tion estimate was modeled as control input. At last, the estimate error was regarded as the maneuvering control input andmodeled as the correlative noise of system. In the proposed model, acceleration maneuvering parameter was estimated byreal-time position measurements and the self-adjusting performance was improved comparing with the existing models.