Vehicle Map Version Selection for Real-Time Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Consumer vehicles generate high-volume, detailed road sensor data that is often discarded due to storage and privacy concerns, limiting map learning and updates.

Innovation Solution

An on-board vehicle apparatus selects a preferred map version from multiple versions based on sensor data and updates a map database in real-time, without requiring large data transmission, by analyzing responses from multiple vehicles to determine the most accurate map representation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high-volume detailed sensor data is transmitted for map learning, then map accuracy is improved, but data transmission burden and privacy issues worsen

Engineering Contradiction:

Improvemap accuracyVSAvoiddata transmission volume

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential selection information (preferred map version indicator) from the vehicle's sensor data processing, rather than transmitting the complete high-volume sensor datasets. The vehicle apparatus independently processes sensor data to determine map version preference, extracting and transmitting only the critical decision outcome.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The vehicle apparatus acts as an intermediary that receives multiple map versions, compares them against local sensor data, and selects the preferred version. This intermediary processing occurs at the vehicle level rather than requiring centralized analysis of all raw sensor data, reducing transmission requirements while maintaining map learning capability.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple map versions are distributed to vehicles for selection, then map learning capability is improved, but system complexity worsens

Engineering Contradiction:

Improvemap learning capabilityVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The map data is segmented into multiple versions representing different possible states of the same map tile. Each version encapsulates a specific interpretation of road features, allowing vehicles to select the most appropriate version based on their sensor observations without requiring the system to manage all possible map variations centrally.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Each vehicle apparatus performs self-service by independently comparing received map versions against its own sensor data and autonomously selecting the preferred version. This distributed decision-making reduces central system complexity while enhancing overall map learning capability across the fleet.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10502577B2Iterative map learning based on vehicle on-board sensor data

Publication Date: 2019.12.10 HERE GLOBAL BV

AI summary

Methods, apparatuses, systems, and computer program products are provided. An example method comprises receiving a change trigger; and providing two or more map versions to a plurality of vehicle apparatuses. The map versions may comprise a stable map version and a changed map version. The example method further comprises receiving two or more responses from at least two of the plurality of vehicle apparatuses. A response comprises an indicator of a preferred map version selected by a vehicle apparatus of the plurality of vehicle apparatuses from the two or more map versions. The example method further comprises analyzing the responses to determine a most preferred map version; and when it is determined that the most preferred map version is a changed map version, updating one or more map databases based at least in part on the changed map version.