Driving Assistance System Using Time-Series Map Updates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing driving assistance systems cannot effectively capture and provide time-varying dynamic traffic environment information, such as vehicle movements and surrounding conditions, due to limitations in image recognition and statistical learning.

Innovation Solution

A driving assistance system that acquires and processes driving information as time-series patterns, associating it with vehicle location data to generate and update map information, allowing for real-time driving assistance based on statistical analysis and prediction of future traffic conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If image recognition processing is performed to collect traffic environment information, then information can be acquired from vehicle surrounding images, but time-varying dynamic traffic environment information cannot be statistically generated as map information

Engineering Contradiction:
Improvetime-varying dynamic traffic environment informationVSAvoidcapability to provide dynamic driving assistance
Core Design Contradiction:
Loss of informationVSAdaptability or versatility

Solution Approach 1:

The patent transforms static map information into dynamic map information by statistically processing time-series driving information. The system captures temporal variations in traffic environments (vehicle movements, pedestrian behaviors, traffic flow patterns) and generates probabilistic predictions that adapt to changing conditions, enabling the map to reflect dynamic characteristics rather than fixed spatial data only

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system implements feedback mechanisms by continuously acquiring new driving information, updating statistical models based on observed patterns, and using predicted future states to improve subsequent information collection. The statistical processing of time-series data creates a feedback loop where past observations inform future predictions, which then guide further information gathering and system adaptation

Inventive Principle:
Principle #23Feedback

2Measurement precision

If statistical learning is applied to suppress recognition position errors, then recognition accuracy is improved, but dynamic traffic environment information cannot be captured

Engineering Contradiction:
Improverecognition position accuracyVSAvoiddynamic environment information reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent extends the analysis from spatial dimensions only to include the temporal dimension. By processing driving information as time-series data and generating probabilistic predictions over time, the system adds a temporal dimension to the statistical analysis, enabling capture of dynamic patterns while maintaining spatial recognition accuracy through multi-dimensional statistical processing

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Ease of operation

If map information is updated based on static location data, then location-based information can be provided, but time-varying traffic patterns cannot be represented

Engineering Contradiction:
Improvelocation-based information provisionVSAvoiddriving assistance effectiveness
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent creates composite map information by combining traditional static spatial map data with dynamic statistical patterns derived from time-series driving information. This composite structure integrates location-based spatial relationships with temporal traffic patterns, producing a richer information framework that maintains ease of location-based querying while significantly enhancing driving assistance effectiveness through predictive capabilities

Inventive Principle:
Principle #40Composite materials

Data Source

PatentEP2876620B1Driving assistance system and driving assistance method
Publication Date: 2019.08.14 NISSAN MOTOR CO LTD
  • EP2876620B1 patent drawingFigure 1
  • EP2876620B1 patent drawingFigure 2
  • EP2876620B1 patent drawingFigure 3

AI summary

The invention includes: a driving information acquisition unit (1) for acquiring any of an amount of operations by the driver, a vehicle behavior, the state of the driver, and information regarding the environment surrounding the vehicle as driving information; a vehicle location information acquisition unit (2) for acquiring vehicle location information; a statistical information generation unit (3) for generating statistical information by statistically processing the driving information as a time-series pattern; a map information generation unit (4) for generating map information by associating the statistical information with the vehicle location information at the time of the acquisition of the driving information; a map information update unit (71) for updating existing map information to the generated map information; a map information reference unit (5) for referencing and reading the map information based on the vehicle location information; and a driving assistance unit (6) for performing driving assistance based on the read map information.