Radar Mapping and Vehicle Localization with Error-Corrected Data Fusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current autonomous and semi-autonomous vehicles rely heavily on outdated or inaccurate cartographic information, which hampers their ability to navigate accurately and efficiently.

Innovation Solution

A method and system that utilizes radar sensors to aggregate and optimize measurement data, reducing errors through clustering and comparison with existing data, enabling the creation or update of precise maps, and using machine learning to filter and align data for enhanced accuracy and storage efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If measurement data is continuously aggregated from multiple radar sensors and measurements, then map completeness and coverage are improved, but measurement errors and data deviations accumulate and increase

Engineering Contradiction:
Improvemap completenessVSAvoidmeasurement error
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The patent implements a feedback mechanism where newly acquired measurement data is continuously compared with existing aggregated measurement data and the created map. Deviations are detected and used to correct the aggregated measurement data, ensuring that error accumulation is prevented while maintaining map completeness through continuous data integration

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Before finalizing the map creation, the patent performs preliminary comparison and alignment of newly acquired measurement data with existing aggregated data. This preliminary action identifies and corrects deviations early in the process, preventing error accumulation before it significantly impacts measurement precision

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If raw measurement data from multiple radar sensors is processed without aggregation, then measurement precision is maintained, but data processing time and computational load increase significantly

Engineering Contradiction:
Improvedata accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent merges measurement data from multiple radar sensors and multiple measurements into aggregated measurement data representing the same spatial area. This combining process reduces the total data volume and processing requirements while maintaining measurement precision through subsequent comparison and correction against the created map

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent segments the processing task by first creating a map from aggregated measurement data, then using this map for subsequent comparisons. This segmentation allows efficient processing by separating map creation from data validation, reducing overall computational load while maintaining accuracy

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances the accuracy and efficiency of map creation and vehicle localization by minimizing measurement errors and optimizing data processing, ensuring vehicles have access to up-to-date and precise cartographic information for navigation.

Implementation Method 1

measurement data of a vehicle environment of at least one vehicle are determined by at least one radar sensor of the at least one vehicle

Methodology Applied
Scientific EffectRadar: Radar

Data Source

PatentEP3688487B1Method and system for mapping and locating a vehicle based on radar measurements
Publication Date: 2025.07.09 ROBERT BOSCH GMBH
  • EP3688487B1 patent drawingFigure 1

AI summary

The invention relates to a method for mapping the surroundings of at least one vehicle and for locating the at least one vehicle, in which: measuring data relating to the vehicle surroundings is determined by means of at least one radar sensor of the at least one vehicle; the measuring data of the at least one radar sensor is aggregated; the aggregated measuring data is compared with already available aggregated measuring data; the aggregated measuring data is optimised by reducing measuring errors on the basis of the comparison between the aggregated measuring data and the already available aggregated measuring data; a map is created or updated on the basis of the optimised aggregated measuring data; and the at least one vehicle is located on the created or updated map by comparing the determined measuring data with the created map. The invention also relates to a system.