Offline Map Generation Using Parallel 3D Modeling Batches
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for generating offline maps for autonomous driving and parking are inefficient, as they require extensive time and resources, and often compromise on accuracy due to the complexity of three-dimensional modeling processes.
Innovation Solution
A method and apparatus that utilize sequential environmental images collected by a vehicle's camera to perform batch three-dimensional modeling, dividing the image set into segments for parallel processing, thereby optimizing posture and world coordinate calculations to generate an accurate offline map quickly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sequential three-dimensional modeling is performed on all environmental images, then mapping accuracy is improved, but processing time increases significantly
Solution Approach 1:
The patent divides the sequence of environmental images into multiple batches, where the first batch is processed sequentially to establish initial mapping relationships, and subsequent batches are processed in parallel using the established relationships as references. This segmentation allows accurate three-dimensional modeling while significantly reducing total processing time by enabling concurrent processing of multiple image batches.
2Productivity
If parallel processing is used for three-dimensional modeling, then processing speed is improved, but mapping accuracy deteriorates
Solution Approach 1:
The patent performs preliminary sequential processing on the first batch of environmental images to establish accurate initial mapping relationships and coordinate systems before initiating parallel processing of subsequent batches. This preliminary action ensures that the reference framework is accurate, allowing parallel processing to maintain mapping accuracy while achieving improved processing speed.
3Area of stationary object
If all environmental images are processed together, then comprehensive coverage is improved, but computational complexity increases
Solution Approach 1:
The patent segments the comprehensive set of environmental images into multiple batches processed in different stages. The first batch establishes the foundational mapping for the entire coverage area, while subsequent batches expand and refine the mapping in parallel. This segmentation maintains comprehensive coverage while reducing computational complexity at each processing stage compared to processing all images simultaneously.
Data Source
AI summary
A method and an apparatus for generating an offline map are disclosed. The method includes: obtaining 1st to nth environmental images sequentially collected by a camera of a vehicle when the vehicle is traveling on a target road section; selecting ith to jth environmental images from the 1st to Nth environmental images, performing a first batch of three-dimensional modeling on the ith to jth environmental images to obtain first posture information of the camera and a first world coordinate; performing a second batch of three-dimensional modeling in parallel on the 1st to (i−1)th environmental images and (j+1)th to Nth environmental images, based on the first posture information and the first world coordinate, to obtain a second world coordinate and to obtain a third world coordinate; and generating an offline map of the target road section according to the world coordinate.