Confidence-Based Road Event Detection System

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Service providers face challenges in efficiently processing vehicle-reported sensor data to accurately detect road events, such as slippery roads, due to variability in vehicle sensors and lack of real-time confidence assessment, leading to potential false positives.

Innovation Solution

A system that aggregates vehicle sensor data and fuses it with external data sources like weather and vehicle specifications to calculate a confidence level for road events, considering factors like data freshness, number of reporting vehicles, and sensor quality, to enhance detection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If vehicle sensor data is aggregated and processed to detect road events, then the quantity of detected road events increases, but the reliability of detection decreases due to variability in vehicle sensors and lack of confidence assessment

Engineering Contradiction:
Improvequantity of vehicle sensor dataVSAvoidreliability of road event detection
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The system implements feedback by calculating a confidence level for each road event detection based on multiple factors including data freshness, number of reporting vehicles, and sensor quality. This confidence level feeds back into the decision-making process to determine whether to report the road event, thereby improving reliability while processing large quantities of sensor data from variable sources

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes parameters by introducing a confidence level metric that combines multiple factors (data freshness, number of vehicles, sensor quality) into a single reliability indicator. This parameter transformation allows the system to filter and prioritize road event detections, converting raw sensor data quantities into reliable actionable information

Inventive Principle:
Principle #35Parameter changes

2Reliability

If confidence level calculation incorporates multiple factors like data freshness and number of vehicles, then the reliability of road event detection improves, but the device complexity increases

Engineering Contradiction:
Improvereliability of road event detectionVSAvoidcomplexity of data processing system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system applies universality by creating a multi-functional confidence level calculation that simultaneously evaluates data freshness, number of reporting vehicles, and sensor quality. This single confidence metric serves multiple purposes: filtering false positives, prioritizing alerts, and guiding reporting decisions, thereby achieving high reliability without proportionally increasing system complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Speed

If all vehicle sensor data is processed in real-time, then the speed of road event detection improves, but the loss of energy increases due to significant computational requirements

Engineering Contradiction:
Improvespeed of road event detectionVSAvoidenergy consumption for data processing
Core Design Contradiction:
SpeedVSLoss of energy

Solution Approach 1:

The system applies partial action by processing vehicle sensor data selectively rather than exhaustively. The confidence level calculation determines which road event detections warrant further processing and reporting. By focusing computational resources only on high-confidence events rather than processing all sensor data equally, the system achieves fast detection speed while significantly reducing energy consumption

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3451312B1Providing a confidence-based road event message
Publication Date: 2023.05.10 HERE GLOBAL BV
  • EP3451312B1 patent drawingFigure 1
  • EP3451312B1 patent drawingFigure 2
  • EP3451312B1 patent drawingFigure 3

AI summary

An approach is provided for a confidence-based road event message. For example, the approach involves aggregating road event reports (e.g., slippery road event reports) from vehicles traveling in an area of interest. The approach also involves retrieving weather data records for the area of interest for a time period corresponding to the reports. The approach may further involve determining a data freshness parameter based on an age of the reports, and a number of vehicle generating the reports. The approach further involves calculating a confidence level for the road event based on the weather data records, data freshness parameter, number of the one or more vehicles, or a combination thereof.