Partial Map Gap Parameter Calculation for Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map generation and own-position estimation techniques for automatic driving systems face challenges in accurately joining multiple maps without inconsistencies, especially as the region size expands, leading to errors in map accuracy.
Innovation Solution
A computing device and own-position estimating device that utilize sensor information, odometry information, and a reference map to generate a periphery map in a vehicle coordinate system, extract partial maps, and calculate gap parameters for coordinate conversion, allowing for the creation of continuous maps with reduced errors across different regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple maps are joined by matching position information of objects, then the map coverage area is expanded, but the map accuracy deteriorates due to inconsistencies and errors in joining
Solution Approach 1:
The patent divides the map into multiple partial maps, each with its own coordinate conversion parameters. By segmenting the map creation process into separate regions that are independently processed and then joined using calculated gap parameters, the system maintains high accuracy within each segment while expanding overall coverage. This resolves the contradiction by allowing area expansion without compromising the precision of individual map segments.
Solution Approach 2:
The patent replaces manual or simple automated map joining methods with a sophisticated coordinate conversion system that uses gap parameter calculations. Instead of directly matching position information of objects to join maps, the system calculates conversion parameters that account for discrepancies between adjacent partial maps, thereby maintaining accuracy during map expansion.
2Adaptability or versatility
If map region expands, then the application range of automatic driving system is extended, but the position estimation accuracy deteriorates due to accumulated errors
Solution Approach 1:
The patent implements a feedback mechanism where gap parameters are calculated based on the relationship between adjacent partial maps. This feedback loop allows the system to detect and correct coordinate conversion discrepancies, ensuring that position estimation accuracy is maintained even as the map coverage area expands to support broader application ranges.
Solution Approach 2:
The patent dynamically adjusts coordinate conversion parameters for each partial map based on calculated gap values. By changing these parameters according to the specific characteristics of each map region and its relationship to adjacent regions, the system maintains consistent position estimation accuracy across the entire expanded map area.
3Area of stationary object
If coordinate conversion is performed for each partial map, then the map can be expanded to larger regions, but the complexity of coordinate management increases
Solution Approach 1:
The patent introduces gap parameters as intermediary values that simplify the management of coordinate conversion parameters. These gap parameters serve as a bridge between adjacent partial maps, providing a systematic way to handle coordinate transformations without requiring complex direct relationships between all map regions. This intermediary approach makes the coordinate management system more tractable as map size expands.
Data Source
AI summary
A computing device includes: a sensor information acquiring unit which acquires sensor information output by a sensor which is mounted on a vehicle and measures a surrounding environment; a storage unit which stores a reference map which is created in advance and is a map of a reference coordinate system which is a predetermined world coordinate system; an odometry information acquiring unit which acquires movement information which is information regarding movement of the vehicle on the basis of an output of the sensor mounted on the vehicle; a periphery map generating unit which generates a periphery map, which is a map in a vehicle coordinate system which is a coordinate system different from the reference coordinate system, on the basis of the sensor information and the movement information; a partial map extracting unit which extracts partial maps, which are regions not overlapping each other, from the periphery map; and a gap parameter calculating unit which calculates, for a coordinate conversion parameter for converting each of the partial maps into a converted partial map which is a map in the reference coordinate system, a gap parameter which is information regarding at least a correlation of the coordinate conversion parameter between the partial maps adjacent to each other, and the partial map extracting unit generates a new partial map when an amount of offset from the reference map is equal to or larger than a predetermined threshold.