Self-Learning Digital Map Creation Using Onboard Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems rely on pre-stored digital maps that may become outdated and are not adaptable to real-time changes in the vehicle's surroundings, limiting their effectiveness in safety-relevant applications like Advanced Driver Assistance Systems (ADAS).
Innovation Solution
An apparatus with a sensor unit to ascertain topographical data, a creation unit to generate digital maps, and a memory unit to store these maps, allowing for iterative updates and customization based on user input and sensor data, including detection of driving patterns and road conditions, enabling the creation of high-quality, safety-relevant digital maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If pre-stored digital maps are used in navigation systems, then the system structure is simple and easy to implement, but the map data becomes outdated and cannot adapt to real-time changes in the vehicle's surroundings
Solution Approach 1:
The system enables vehicles to autonomously create and update their own digital maps using onboard sensors (camera, radar, GPS, inertial sensors) without requiring external map providers or complex infrastructure. Each vehicle serves itself by capturing topographical data, processing it through creation units, and storing updated maps locally, thereby achieving real-time adaptability while maintaining relatively simple system architecture.
Solution Approach 2:
The system continuously compares sensor-captured environmental data with existing map data, identifies discrepancies and updates, and iteratively refines the digital maps. This feedback loop allows the system to adapt to real-time changes in road geometry, signage, and surroundings while maintaining system simplicity through automated processing algorithms.
2Reliability
If digital maps are updated in real-time using sensor data, then the map accuracy and safety relevance are improved, but the device complexity and processing requirements increase
Solution Approach 1:
The system divides the map creation and update process into distinct functional modules: sensor units for data capture, creation units for processing and map generation, and memory units for storage. This segmentation allows each component to be optimized independently and facilitates iterative updates without requiring complete system redesign, thereby improving reliability while managing complexity.
Solution Approach 2:
The system performs partial updates by focusing only on changed elements in the environment rather than completely regenerating entire maps. The creation unit processes only the portions of the environment that have changed since the last map update, reducing processing requirements and apparatus complexity while maintaining high safety relevance through continuous refinement of critical areas.
3Measurement precision
If iterative creation of digital maps is performed by traveling the same route multiple times, then the map quality and precision are improved, but the time required for map creation increases
Solution Approach 1:
The system performs preliminary map creation during the first traversal of a route, establishing a baseline map that provides immediate navigational value. Subsequent traversals then focus on refining and updating specific portions of the map, rather than creating maps from scratch each time. This preliminary action approach reduces the time investment required for iterative improvements while progressively enhancing map quality and precision.
Solution Approach 2:
The system maintains continuous useful action by ensuring that each route traversal contributes to map improvement, even if only incrementally. Rather than requiring complete re-mapping, the system continuously refines existing maps through repeated traversals, accumulating precision over time while minimizing additional time expenditure through intelligent update strategies that focus on changed or uncertain areas.
Data Source
AI summary
A self-learning map or a device for creating and storing a digital map for a transport unit on the basis of environmental sensors, vehicle-to-X communication and satellite navigation systems. The self-learning map and device create and store the digital map without the use of data from navigation maps. The obtained digital map is iteratively improved and can be used for the validity check of an existing digital map for a driver assistance system.