基于网格细胞模型的类脑大尺度空间矢量导航方法

动物和人类可以使用感官中不完整的空间信息来快速定位其当前位置并导航到目标, 为未知环境下的矢 量导航提供了生物模型. 本文针对基于连续吸引子模型和余数系统的大尺度空间矢量导航方法所存在的鲁棒性问 题, 提出了一种基于振荡干扰模型和逐级模糊度确定法的大尺度空间矢量导航方法. 仿真结果表明, 在2%的测量噪 声条件下, 该方法可以在245 m*245 m*sin 60° 的大尺度环境下准确解算出位置矢量, 并且每个维度中位置的解算 精度可以达到1 cm以内, 有效提高了大尺度空间内矢量导航的鲁棒性.

Brain-like large-scale spatial vector navigation method based on grid cell model

Animals and humans can quickly use the incomplete spatial information in the senses to locate their current position and navigate to the target, providing a biological model for vector navigation in an unknown environment. Aiming at the robustness problems of the large-scale spatial vector navigation method based on the continuous attractor model and the remainder system, a large-scale spatial vector navigation method is proposed based on the oscillatory interference model and the stepwise ambiguity determination method. The simulation results show that under the condition of 2% measurement noise, this method can accurately calculate the position vector in a large-scale environment of 245 m*245 m* sin 60°, and the calculation accuracy of the position in each dimension can reach within 1 cm, which effectively improves the robustness of vector navigation in a large-scale space.