论文部分内容阅读
为实现无人作战飞机(UCAV,Unmanned Combat Aerial Vehicle)认知导航的空间方位自主推算,提出了一种基于多尺度网格细胞的路径整合方法.该方法模拟背侧内嗅皮层(dMEC,dorsal Medial Entorhinal Cortex)的相同区域网格细胞放电特征相同、不同区域放电特征递增变化的特点,构建尺度递增的仿生多尺度网格图组,在各层中引入突触样式(synaptic pattern)计算各细胞权值,通过细胞的活跃度变化表征各网格层中位置的变化,并在各层分别实现路径整合,进而利用低尺度整合结果调整高尺度整合,提高空间位置的推算精度.实验结果表明,所提方法在一定的速度误差与方向误差范围内能够精确推算方位,具有较高的空间位置推算精度,并且方向误差值随运动方向变化呈锯齿状分布.
In order to realize the autonomous estimation of the spatial orientation of UCAV (Unmanned Combat Aerial Vehicle) cognitive navigation, a multi-scale grid cell based path integration method is proposed to simulate the dorsal entorhinal cortex (Medial Entorhinal Cortex) cells in the same region, and the characteristics of discharge in different regions were changed gradually. The scale-increasing bionic multi-scale grid group was constructed, and synaptic patterns were introduced into each layer to calculate the number of cells Weight, the change of cell activity is used to characterize the change of location in each grid layer, and the path integration is achieved at each layer, and then the high-scale integration is adjusted by using the low-scale integration results to improve the accuracy of spatial location estimation.Experimental results show that, The proposed method can accurately calculate azimuths within a certain range of velocity errors and directional errors, and has a high accuracy of spatial position estimation. The direction error values are distributed in a zigzag pattern according to the movement direction.