Unmanned vehicle semantic map modeling and application constructing method based on perception and location monitoring

Abstract

The invention discloses an unmanned vehicle semantic map modeling and application constructing method based on perception and location monitoring and relates to the technical field of unmanned driving. According to the method in the invention, two concept modules, namely entity and attribute, are included in unmanned vehicle semantic map modeling, wherein the entity comprises a vehicle entity, a road network entity and an obstacle entity; the road network entity comprises region entity and point entity; the attribute comprises point coordinate, area coverage and constraint; and the semantic relation is divided into two parts, namely object attribute and data attribute. According to the method disclosed by the invention, a set of map element data hierarchy structure applicable to the unmanned vehicle is constructed, and by designing the sufficient semantic relationship among map elements, the semantic map is conveniently generated; and semantic reasoning is performed on the semantic map, global planning path, the current posture of the unmanned vehicle and surrounding real-time obstacle information so as to obtain local scene information of the unmanned vehicle, the unmanned vehicleis assisted in performing behavior decision, understanding of the unmanned vehicle on driving scene elements is efficiently completed, and the associated search efficiency of the map elements is improved.

Description

technical field [0001] The invention relates to the field of unmanned driving technology, in particular to an application method for modeling and constructing semantic maps of unmanned vehicles based on perception positioning monitoring. Background technique [0002] In recent years, unmanned driving has received extensive attention from domestic and foreign academic and industrial circles, and its related supporting technologies have developed rapidly. From the perspective of unmanned driving application products, they can generally be divided into two categories: unmanned driving products for industrial production applications and unmanned driving products for personal consumption applications. In terms of the composition of the unmanned driving technology system, the unmanned driving system can generally be divided into sub-modules such as environmental perception, decision planning, and motion control. Environmental perception is to obtain real-time scene information of ...

AI Technical Summary

Problems solved by technology

[0005] Existing Chinese patent: Publication No. CN104535070A (Application No. 20141083873.5), which provides a high-precision map data structure, acquisition and processing system and method, divides the map data structure into four layers: road network, lane network, Lane line information and special information data, although database-level associations are defined between several levels, due to the lack of semantic information, it is difficult for unmanned vehicles to establish various map elements in this map data structure and perfect communication between traffic participants. Semantic relationship, distinguish real-time scene information of unmanned vehicles, and realize scene understanding

At the same time, information such as intersections and U-turns is difficult to reflect in its data structure, and the relationship between lane lines and lanes is not accurate enough. For example, a certain section of road may change from two lanes to three lanes. In this case, the lane in the middle of the three lanes and the lane It is difficult to express the relationship between the lines

[0006] Existing Chinese patent: Publication No. CN104089619A (Application No. 201410202876.4), which provides an accurate GPS navigation map matching system for unmanned vehicles and its operation method, by obtaining road information, determining the starting point, and obtaining vehicle positioning information , the process of information matching and screening completes the precise matching of the navigation map, but the matching method is mainly to search through discrete points, without utilizing the correlation between map elements, which will lead to the problem of low matching efficiency

[0007] The main problem in the existing technology is that the data structure of the high-precision map is not convenient for unmanned vehicles to understand the scene, and at the same time, the correlation of map elements is not well utilized, resulting in a relatively low map search efficiency.

Images