Driver Assistance System Risk Estimation Using Prototypical Situations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing driver assistance systems only consider risk factors derived from the current driving environment, failing to incorporate prior knowledge about dangerous locations or types of intersections, which limits their effectiveness in predicting and mitigating accidents.

Innovation Solution

A method and system that senses the vehicle's environment using various sensors, determines relevant parameters, and compares them to prototypical driving situations stored in a database to calculate an expected risk, incorporating accident statistics and road geometry information to provide more comprehensive risk assessment and driver assistance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If driver assistance systems use only current environment sensing data for risk estimation, then the system complexity remains manageable, but the risk estimation accuracy is insufficient because prior knowledge about dangerous locations is not incorporated

Engineering Contradiction:
Improverisk estimation accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system pre-stores prototypical traffic situations and associated risk information in a database before actual driving scenarios occur. When a current situation is detected, the system retrieves pre-computed risk data from the database that matches the detected situation type, avoiding the need to compute risk from scratch and incorporating prior knowledge about dangerous locations and scenarios.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates simplified representations (copies) of complex traffic situations by categorizing them into prototypical situations. Instead of processing all raw sensor data in real-time, the system copies relevant features into standardized situation templates and retrieves pre-analyzed risk information associated with these templates, significantly reducing computational complexity while maintaining estimation accuracy.

Inventive Principle:
Principle #26Copying

2Measurement precision

If the system incorporates comprehensive accident statistics and historical data for all possible situations, then the risk estimation becomes more accurate, but the database size and information processing requirements become unmanageably large

Engineering Contradiction:
Improverisk estimation accuracyVSAvoiddata volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The system segments the vast space of possible traffic situations into a finite set of prototypical situation categories based on common driving scenarios. By dividing the continuous environment into discrete situation types (e.g., intersection crossings, merging lanes, pedestrian zones), the system only needs to store risk data for these representative categories rather than every possible scenario, dramatically reducing data volume while maintaining comprehensive coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transforms detailed, high-dimensional sensor data into reduced-parameter prototypical situation representations. By changing the parameter space from raw sensor measurements to categorized situation types with key features, the system retains essential risk information while eliminating redundant data, achieving efficient storage and retrieval without sacrificing estimation accuracy.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If the system processes and compares current situation with all stored prototypical situations in real-time, then the most accurate risk classification is achieved, but the processing time exceeds acceptable limits for real-time driver assistance

Engineering Contradiction:
Improvesituation classification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system pre-computes and stores matching criteria for each prototypical situation in the database. When a current situation is detected, the system retrieves and compares only the relevant pre-computed criteria rather than comparing against all possible prototypical situations. This preliminary preparation of matching rules enables fast real-time classification while maintaining high accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system extracts only the essential matching features from the full set of situation parameters that are relevant for classification. Instead of comparing all available sensor data against all prototypical situations, the system identifies and extracts the critical features that distinguish different situation types, reducing the comparison space and enabling real-time processing without sacrificing classification accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3159235B1Method and system for assisting a driver of a vehicle in driving the vehicle and computer program
Publication Date: 2021.08.04 HONDA RES INST EUROPE
  • EP3159235B1 patent drawingFigure 1
  • EP3159235B1 patent drawingFigure 2~3

AI summary

The invention regards an advanced driver assistant system and a method for assisting a driver of a vehicle in driving the vehicle. The system comprises sensing means (2) for sensing an environment of the vehicle (1), a storing unit (6) for storing a database for storing prototypical situations associated with risk information, a processing unit (3) and an output unit (9) to output an assistance signal on the basis of which driver information can be output and/or vehicle operators can be actuated. The processing unit (3) is configured to determine parameters of the vehicle (1) and/or objects in the environment of the vehicle (1) on the basis of the sensing means output, to classify a traffic situation on the basis of the determined parameters by comparing the determined parameters with parameters of the prototypical traffic situations, to extract risk information from the database and to calculate an expected risk for the encountered traffic situation on the basis of the extracted risk information.