Navigation Information Fusion Using Sparse Maps for Autonomous Vehicles

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in navigating effectively due to the vast amounts of data they need to process and store for navigation, including visual information, GPS data, and sensor data, which can lead to inefficiencies and safety concerns.

Innovation Solution

The use of cameras and image processing systems to provide navigational responses based on image analysis, combined with GPS and sensor data, and the implementation of a crowdsourced sparse map for optimized navigation, ensuring safety and comfort constraints are met.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional mapping technology is used to navigate, then navigation accuracy is improved, but data storage requirements and system complexity increase significantly

Engineering Contradiction:
Improvenavigation accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential navigational elements from complete maps, creating sparse maps that contain only critical information needed for navigation decisions. This reduces data storage requirements while maintaining navigation accuracy by focusing on key features rather than storing entire map datasets.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The navigation system segments map data into hierarchical levels, using detailed maps only when needed and sparse maps for general navigation. This segmentation allows the system to manage complexity by dividing map representation into manageable portions based on navigation requirements.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If detailed map data is stored and updated continuously, then navigation accuracy is improved, but data transmission and processing time increase

Engineering Contradiction:
Improvenavigation accuracyVSAvoiddata transmission time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system extracts only essential navigational information from detailed maps for transmission and processing, reducing data volume while maintaining navigation accuracy. Sparse maps contain only critical path information, eliminating unnecessary data transmission overhead.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If vast volumes of sensor data are collected and analyzed, then navigation reliability is improved, but processing time and computational resources increase

Engineering Contradiction:
Improvenavigation reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system extracts only the most relevant features from sensor data that are critical for navigation decisions, rather than processing entire datasets. This selective extraction maintains navigation reliability by focusing on key indicators while reducing computational burden and processing time.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial processing to sensor data, analyzing only the portions necessary for current navigation decisions rather than comprehensively processing all available data. This approach achieves sufficient reliability for safe navigation while minimizing processing time and resource consumption.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3764060B1Fusion framework of navigation information for autonomous navigation
Publication Date: 2024.07.24 MOBILEYE VISION TECH LTD
  • EP3764060B1 patent drawingFigure 1
  • EP3764060B1 patent drawingFigure 2A
  • EP3764060B1 patent drawingFigure 2B

AI summary

The present disclosure relates to apparatuses, systems and methods for navigating vehicles. In one implementation, at least one processing device receive a first output from a first data source and a second output from a second data source; identify a representation of a target object in the first output; determine whether a characteristic of the target object triggers at least one navigational constraint; if the at least one navigational constraint is not triggered by the characteristic of the target object, verify the identification of the representation of the target object based on a combination of the first output and the second output; if the at least one navigational constraint is triggered by the characteristic of the target object, verify the identification of the representation of the target object based on the first output; and in response to the verification, cause at least one navigational change to the vehicle.