Unstructured road state parameter estimation method and system

Abstract

The invention discloses an unstructured road state parameter estimation method and system, and the method comprises the steps: 1, carrying out the time synchronization and space synchronization of collected point cloud data and image data, and obtaining a mapping relation between a point cloud in the point cloud data in a laser radar coordinate system and a pixel in a pixel coordinate system; 2, performing multi-sensor fusion ground segmentation, and outputting a point cloud ground segmentation result and an image ground segmentation result; step 3, enhancing ground point cloud data; 4, road adaptive block fitting is carried out, and road surface modeling is realized; step 5, estimating road parameters: based on a pavement plane model, calculating a normal vector included angle of two adjacent planes to obtain relative longitudinal / transverse gradients of the adjacent planes; the distance between the ground point cloud and the corresponding fitting plane is calculated, and the root mean square of the distance is used as an evaluation index of the pavement roughness; and performing road curvature estimation by fitting the road boundary of each block plane.

Description

A method and system for estimating state parameters of unstructured road technical field The present invention relates to the technical field of intelligent vehicle environment perception, especially about a kind of unstructured road state parameter Estimation methods and systems. Background technique Automotive intelligence is an important trend in the development of the automotive industry, and environmental perception technology is the core technology of intelligent vehicle technology. one of the techniques. Environmental perception technology provides environmental information for decision-making, control and other technologies by acquiring and analyzing sensor data. Of course However, when driving on unstructured roads such as mining areas and off-road scenarios, intelligent vehicles need to face the changing road types and increasing slopes. This will undoubtedly bring great challenges to the motion control and safe driving of intelligent vehicles. Therefore...

AI Technical Summary

Problems solved by technology

[0004] (1) The method based on the depth camera has strict requirements on the lighting conditions, and is greatly affected by the estimated depth of the camera, and has poor robustness in complex environmental conditions;

[0005] (2) The method based on 3D point cloud only relies on spatial geometric information and cannot achieve accurate ground segmentation, and for distant point cloud sparse areas, the parameter estimation error is relatively large;

[0006] (3) The planar block strategy used in the existing road state parameter estimation methods does not consider the actual road conditions, and has poor robustness, which greatly reduces the accuracy of subsequent parameter estimation;

[0007] (4) The method based on vehicle driving parameters is a contact estimation method. Only when the vehicle is driving on the current road surface can the ground state be accurately estimated, and it is impossible to predict the ground state parameters of the driving area in front of the vehicle in advance, and it cannot achieve advance planning. the goal of

Images