Map Generation Device Using Distance-Based Integration Areas
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map generation technologies face challenges in integrating data from multiple vehicles when they are packed densely, leading to complications in determining weighting and potential inaccuracies in positional information.
Innovation Solution
A map generation system that includes on-board devices and a map generation device, which acquire positional information from vehicles, calculate distances between them, set integration areas based on these distances, and integrate surrounding object information into a map, using methods like SfM and triangulation to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If vehicles are packed densely and images from a large number of vehicles are integrated, then the coverage area of the map increases, but the determination on weighting in integrating data becomes complicated
Solution Approach 1:
The patent segments the integration process by first performing local integration within detection areas of individual vehicles, then performing higher-level integration of these local results. This hierarchical segmentation reduces the complexity of integrating data from all vehicles simultaneously while maintaining comprehensive map coverage.
Solution Approach 2:
The patent introduces a hierarchical dimension to the integration process, moving from two-dimensional spatial integration to three-dimensional integration that includes the hierarchy level. This allows complex multi-vehicle data to be organized and processed in manageable layers, reducing integration complexity.
2Quantity of substance
If images of the same object are captured from a large number of vehicles at the same time, then the data quantity increases, but the weighting determination becomes complicated
Solution Approach 1:
The patent applies local quality by determining weighting based on local conditions such as the detection area of each vehicle and the distance to the target object. Instead of applying uniform weighting rules across all vehicles, the system adapts weighting to local geometric relationships, simplifying the overall determination process while utilizing data from multiple vehicles.
Solution Approach 2:
The patent changes the parameters used for weighting determination from simple vehicle count to geometric parameters such as detection area coverage and distance metrics. This parameter transformation simplifies the weighting calculation by using physically meaningful quantities that naturally reflect data quality and reliability.
Data Source
AI summary
A map generation device includes an information acquisition unit that acquires positional information on a specific vehicle and positional information on a surrounding vehicle that is positioned around the specific vehicle from on-board devices that are arranged respectively on the specific vehicle and the surrounding vehicle, respectively; a distance calculator that calculates a distance between the specific vehicle and the surrounding vehicle based on the positional information on the specific vehicle and the positional information on the surrounding vehicle; an integration area setting unit that sets an integration area of integration into a map based on a result of calculating the distance and an area of detection by the specific vehicle; and a map integration unit that integrates the positional information on the surrounding object that is acquired from the on-board device in the integration area into a map.