Odometry Data Collection for Autonomous Driving Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for creating high-precision geometric maps for autonomous or automated driving require specialized vehicles and are costly, with deviations in odometry data leading to inaccuracies in vehicle localization.
Innovation Solution
A method and device that collect and transmit odometry data using customer or consumer vehicles, breaking routes into sections with specified maximum driving lengths based on sensor precision and reproducibility, allowing for cost-effective and timely map data provision, using reference sensors and actuators similar to those in standard vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specialized measuring vehicles with special sensor systems are used to collect data for high-precision geometric maps, then measurement precision and data accuracy are improved, but device complexity and cost increase
Solution Approach 1:
The patent makes standard vehicle sensor systems perform the specialized function of collecting high-precision odometry data for geometric map creation. Customer vehicles with conventional sensors are used instead of dedicated measuring vehicles, allowing the same sensor system to serve both normal driving functions and map data collection
Solution Approach 2:
Customer vehicles automatically collect and transmit odometry data during normal operation without requiring specialized equipment. The vehicles serve themselves by utilizing their existing sensor systems to contribute to map creation, eliminating the need for separate specialized measuring vehicles
2Productivity
If the maximum driving length of route sections is increased, then productivity of map data collection is improved, but measurement precision and reproducibility of odometry data deteriorate
Solution Approach 1:
The patent dynamically adjusts the maximum driving length parameter of route sections based on the detected precision of odometry data and reproducibility of vehicle actuators. When precision is high, longer sections are allowed; when precision decreases, section length is reduced to maintain localization accuracy within 10 cm deviation
3Loss of time
If odometry data are collected over long route sections, then loss of time for map data provision is reduced, but reliability of vehicle localization deteriorates due to accumulated deviations
Solution Approach 1:
The patent divides the territory into multiple route sections with specified maximum driving lengths. By segmenting the overall mapping task into smaller sections, the system achieves a balance between collection efficiency and localization accuracy, ensuring deviations remain within acceptable limits while maintaining productive data collection
Data Source
AI summary
A method for providing data for a geometric map for autonomous or automated driving of a vehicle includes collecting and/or determining odometry data of the vehicle when driving along a respective specified route section of a plurality of route sections using specified vehicle sensors. The respective specified route section has a maximum driving length that is dependent on at least one of a detection precision of odometry data using a specified vehicle sensor system, and on a reproducibility of provided odometry data using specified vehicle actuators and/or vehicle actuating drives. The method also includes providing the collected and/or determined odometry data to a specified vehicle interface, and transmitting the collected and/or determined odometry data by the specified vehicle interface to an evaluation device.