Sparse Map Model for Autonomous Vehicle Navigation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in navigating efficiently due to the vast amounts of data they need to process and store, particularly with traditional mapping technologies, which can limit their ability to accurately identify location, avoid obstacles, and respond to traffic signals and signs.

Innovation Solution

The system employs cameras to collect and analyze navigational information, determining a vehicle's location and transmitting selected data to a server, which requests specific information for updating maps, using a sparse map model that requires less data storage and transfer, allowing for efficient navigation with reduced data overhead.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional mapping technology is used for autonomous navigation, then comprehensive map data can be obtained, but data storage requirements and data transfer overhead increase significantly

Engineering Contradiction:
Improvenavigation accuracyVSAvoiddata storage requirements
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential navigational elements from traditional comprehensive maps, creating a sparse map model that contains only critical information needed for autonomous navigation such as road boundaries, intersections, and key landmarks, thereby reducing data storage requirements while maintaining navigation accuracy

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the map data into hierarchical levels, with sparse maps providing overall navigational structure and detailed maps providing localized information only when needed, allowing the system to load and process map data incrementally rather than all at once

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If comprehensive navigational data is collected and stored, then navigation accuracy is improved, but data transfer overhead and processing time increase

Engineering Contradiction:
Improvelocation identification accuracyVSAvoiddata processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements partial action by collecting and processing only the subset of navigational data that is currently needed for the vehicle's position and intended route, rather than processing all available data, thereby reducing processing time while maintaining sufficient accuracy for safe navigation

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent performs preliminary actions by pre-processing and pre-filtering navigational data to identify and store only relevant features before the vehicle reaches specific locations, so that when the vehicle needs navigation information, the data is already prepared and ready for rapid processing

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If frequent data transmission to servers is performed, then map updates and navigation improvements are achieved, but network bandwidth consumption and data transfer overhead increase

Engineering Contradiction:
Improvemap update capabilityVSAvoidnetwork bandwidth consumption
Core Design Contradiction:
Adaptability or versatilityVSLoss of energy

Solution Approach 1:

The patent implements feedback mechanisms where the autonomous vehicle transmits navigation performance data and observed environmental features to remote servers, which then generate updated sparse map data and transmit back only the necessary corrections and updates, creating an efficient closed-loop system that minimizes bidirectional data transmission while maintaining adaptability

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent discards redundant and duplicate navigational data that can be inferred from existing sparse maps, and only transmits novel or corrected information to servers, while recovering and reusing previously transmitted data patterns to reduce overall transmission requirements

Inventive Principle:
Principle #34Discarding and recovering

Data Source

PatentEP3998457A1Systems and methods for vehicle navigation
Publication Date: 2022.05.18 MOBILEYE VISION TECH LTD
  • EP3998457A1 patent drawingFigure 1
  • EP3998457A1 patent drawingFigure 2A
  • EP3998457A1 patent drawingFigure 2B

AI summary

Systems and methods may selectively collect information from a host vehicle. In one example, a method may include causing collection of navigational information associated with an environment traversed by the host vehicle; storing the collected navigational information; determining, based on an output of at least one navigational sensor, a location of the host vehicle; transmitting the determined location of the host vehicle to a server; receiving, from the server and in response to the transmitted determined location, a request for transmission of a selected subset of the navigational information collected by the host vehicle; and transmitting the selected subset of the navigational information to the server.