Digital Localization Map Clustering for Scalable Production
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for creating digital localization maps face challenges in scalability and map density, particularly when dealing with large areas that contain both high-data regions like cities and low-data regions like rural areas, leading to inefficient map production processes.
Innovation Solution
A method that determines road sections, classifies them based on defined conditions and sensor data, combines them into clusters for joint computational processing, and adapts cluster sizes to available computing power, enabling efficient and balanced map production.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If conventional grid-based approach or road section approach is used for creating digital localization maps, then mapping on a large scale is enabled, but partitions with great deal of data (large cities) and partitions with very little data (rural regions) exist which makes map production process problematic
Solution Approach 1:
The patent applies local quality by classifying road sections into different complexity categories (simple, medium, complex) based on their specific characteristics such as intersection types, curvature, and signage density. This allows each region to be processed with appropriate computational resources - simple rural roads require minimal processing while complex urban intersections receive enhanced attention, optimizing overall map production efficiency across diverse geographic areas.
Solution Approach 2:
The patent segments the large-scale mapping area into individual road sections that are then grouped into clusters based on their complexity characteristics. This segmentation enables independent processing of each cluster, allowing parallel computation and efficient resource allocation. Complex urban clusters can be processed separately from simpler rural clusters, resolving the productivity issue caused by heterogeneous data densities across large mapping areas.
2Productivity
If data is processed independently in uniform rectangles (grid-based approach), then calculation is quick and inexpensive, but map density is not considered making the process problematic for large areas
Solution Approach 1:
Instead of uniform grid processing, the patent implements local quality by evaluating each road section's complexity and creating clusters with heterogeneous data densities. This allows computational resources to be concentrated where needed - complex urban areas receive more detailed processing while rural areas use streamlined processing - thereby maintaining both calculation efficiency and map density quality appropriate to each region's requirements.
Solution Approach 2:
The patent introduces dynamics by making the processing granularity adaptive rather than static. Road sections are dynamically grouped into clusters based on their complexity characteristics, allowing the processing units to expand or contract in size and detail based on local requirements. This dynamic clustering enables the system to adjust computational effort to match actual map production needs in different regions.
3Manufacturing precision
If road sections are classified and combined into clusters for joint computational processing, then highly accurate digital localization map is provided, but computing power requirements increase
Solution Approach 1:
The patent segments the computational task by dividing road sections into complexity-based clusters that can be processed independently and in parallel. This segmentation maintains high map accuracy through detailed local processing while reducing total computing power requirements by enabling distributed computation across multiple processors or time periods, avoiding the need for a single massive computational resource.
Solution Approach 2:
The patent applies parameter changes by adjusting the clustering granularity and processing depth based on road section complexity. Simple rural road clusters may use coarser processing parameters while complex urban clusters use finer parameters, optimizing the balance between map accuracy and computing power consumption. This adaptive parameter adjustment ensures high accuracy where needed without unnecessarily consuming computing resources in simpler areas.
Data Source
AI summary
A method for providing a digital localization map. The method includes: determining road sections by evaluating data of a road network to be mapped; classifying the road sections, taking defined conditions of the road network into account; classifying the road sections, taking driving-environment sensor data into account; combining the road sections to form clusters, taking the classifications into account; joint computational processing of each of the clusters; and transmitting the digital localization map created to a vehicle.