A global environment map generation method for driverless road level

Abstract

The invention discloses a method for generating a road-level global environment map for unmanned driving, which can obtain current data of unmanned vehicles, calculate the current motion state of unmanned vehicles, extract image feature matrices, screen similar visual templates and add new visual Steps such as template, experience point matching, lane line detection, and lane line and global experience map fusion finally generate an unmanned road-level global environment map. Unmanned vehicles use on-board sensors to perceive the surrounding environment of the vehicle, and according to the perceived road , vehicle position and obstacle information to achieve the purpose of unmanned driving, and overcome the traditional problems of being susceptible to terrain, weather conditions and inaccurate precision. A road-level global environment map generation for unmanned driving in the present invention Method, by continuously capturing environmental information during vehicle driving to generate a global experience map, combined with multi-lane line detection, a road-level global environment map with high accuracy is obtained.

Description

technical field [0001] The invention relates to the technical field of unmanned driving, in particular to a method for generating a road-level global environment map for unmanned driving. Background technique [0002] With the vigorous development of communication technology, machine intelligence technology, computer network technology and geographic information system technology, unmanned driving has gained wide attention in many aspects such as military, civilian and scientific research. Interdisciplinary research in many fields, such as electronics, artificial intelligence, etc., has a very broad application prospect. Driverless car is a kind of smart car, which can also be called a wheeled mobile robot. Since the 1970s, developed countries such as the United States, Britain, and Germany have begun to conduct research on driverless cars. Over the years, in the feasibility Significant breakthroughs have been made in terms of practicality and practicality. [0003] At pre...

AI Technical Summary

Problems solved by technology

Many traditional driverless cars use GPS for map drawing. However, as mentioned above, this existing technical solution has great problems and shortcomings, that is, it is vulnerable to terrain, weather conditions and lack of accuracy. exact question

Moreover, the production cost is not low. It is obviously very disadvantageous to use the map produced by this method to promote driverless cars.

Images