A Robot Navigation Method Based on the Neural Mechanism of Visual Perception and Spatial Cognition

Abstract

The invention relates to a robot navigation method based on the neural mechanism of visual perception and spatial cognition, which converts the collected visual images into visual nodes representing robot position and orientation angle information through a neural network to form visual cells; The encoding is transformed into a spatial description of the environment, and a cognitive map similar to that formed in the brain when mammals move freely is constructed; the robot's positioning and navigation are realized according to the cognitive map. Based on the neural computing mechanism of environmental perception and spatial memory, the robot has completed a series of tasks such as visual processing, spatial representation, self-positioning, map update, etc., and realized robot navigation with high bionicity and strong autonomy in unknown environments; compared with traditional SLAM technology, the invention avoids a series of complex calculations such as manual design of visual features and feature point matching, and greatly improves the robustness of the system to factors such as illumination changes, viewing angle changes, and object movement in natural environments.

Description

technical field [0001] The invention relates to a brain-like navigation method. Specifically, it is a system for robots to navigate in unknown environments using the neural computing mechanism of visual perception and spatial cognition. Background technique [0002] Robot autonomous navigation research mainly focuses on: "where" (positioning); "where should go" (path planning). Although the existing navigation technology has solved these two problems to a certain extent, there are still great defects, such as the low positioning accuracy of GPS technology, and it cannot normally provide navigation information in special or concealed environments such as indoors, underwater, and after disasters. ; The traditional robot localization and mapping technology (SLAM) relies on expensive sensors such as odometers and lasers, and the spatial perception information is single. Visual navigation has become a research hotspot in recent years due to its rich source of perceptual informa...

AI Technical Summary

Problems solved by technology

Although this research field has great impetus for the further development of robot navigation technology, due to a series of factors such as the difficulty of brain science exploration, there are still few breakthroughs in this research.

Images