仿信鸽脑-海马的无人机同时定位与建图导航系统及方法

本发明公开一种仿信鸽脑-海马的无人机同时定位与建图导航系统及方法,该系统包括7个模块,分别为图像采集与预处理模块、局部视场细胞模块、里程计模块、位姿细胞模块、头朝向细胞模块、网格细胞模块、经验地图模块;本发明方法模拟鸽脑海马具有两种独立导航机制的特点,基于连续吸引子神经网络模型,分别建立了两种类脑导航路径积分算法,并结合局部视图细胞、位姿细胞和经验地图等模块,实现冗余、鲁棒的定位和导航,弥补了原始RatSLAM算法不具备冗余导航的缺点,可以辅助无人机实现导航与定位,提高无人机智能水平。 The invention discloses a carrier pigeon brain-hippoca...

Full description

Saved in:
Bibliographic Details
Format Patent
LanguageChinese
Published 10.06.2022
Subjects
Online AccessGet full text

Cover

More Information
Summary:本发明公开一种仿信鸽脑-海马的无人机同时定位与建图导航系统及方法,该系统包括7个模块,分别为图像采集与预处理模块、局部视场细胞模块、里程计模块、位姿细胞模块、头朝向细胞模块、网格细胞模块、经验地图模块;本发明方法模拟鸽脑海马具有两种独立导航机制的特点,基于连续吸引子神经网络模型,分别建立了两种类脑导航路径积分算法,并结合局部视图细胞、位姿细胞和经验地图等模块,实现冗余、鲁棒的定位和导航,弥补了原始RatSLAM算法不具备冗余导航的缺点,可以辅助无人机实现导航与定位,提高无人机智能水平。 The invention discloses a carrier pigeon brain-hippocampus imitating unmanned aerial vehicle simultaneous positioning and mapping navigation system and method. The system comprises seven modules: an image collection and preprocessing module, a local view field cell module, a speedometer module, a pose cell module, a head orientation cell module, a grid cell module, and an experience map module. The method simulates the characteristic that a pigeon brain hippocampus has two independent navigation mechanisms, two kinds of brain navigation path integral algorithms are established based on a continuous attractor neural network model, and redundant and robust positioning and navigation are realized in combination with local view cells, pose cells, an empirical map and othe
Bibliography:Application Number: CN202010776277