Offline Map Reconstruction with Batched 3D Road Image Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating offline maps for autonomous driving and parking are time-consuming and lack accuracy, which hinders the efficiency and safety of autonomous vehicle operations.
Innovation Solution
A method involving sequential environmental image collection by a vehicle's camera, followed by selective three-dimensional modeling in batches to obtain posture information and world coordinates, allowing for parallel processing to generate an offline map, ensuring accuracy and reducing the time-consuming process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If three-dimensional modeling is performed sequentially on all environmental images to ensure accuracy, then mapping precision is improved, but the time required for map generation increases significantly
Solution Approach 1:
The patent divides the environmental images into multiple batches and performs three-dimensional modeling on each batch separately. This segmentation allows parallel processing of different image batches while maintaining the precision requirements for each segment, thereby reducing total generation time without sacrificing mapping accuracy.
Solution Approach 2:
The patent performs preliminary three-dimensional modeling on selected batches of environmental images before final map generation. This preliminary action creates intermediate results that can be combined later, enabling faster overall processing while ensuring accuracy through staged computation.
2Productivity
If parallel processing is used to speed up map generation, then productivity is improved, but coordination complexity between processing batches increases
Solution Approach 1:
By segmenting the environmental images into distinct batches that can be processed in parallel, the system achieves faster map generation while managing complexity through clear separation of processing tasks. Each batch is handled independently with defined input-output relationships.
Solution Approach 2:
The patent uses intermediate results from preliminary three-dimensional modeling as mediators between parallel processing batches and the final map generation. These intermediates simplify coordination by providing standardized data structures that can be combined without complex real-time synchronization.
3Measurement precision
If all environmental images are processed to ensure complete coverage, then mapping precision is improved, but the complexity of the processing system increases
Solution Approach 1:
The patent segments the complete set of environmental images into multiple batches that can be processed independently through parallel three-dimensional modeling. This maintains comprehensive coverage and precision while reducing system complexity by breaking down the monolithic processing task into manageable, independent units.
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.