Grid-Based Vehicle Route Prediction Database

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current vehicle mapping technologies are static and lack coverage in areas with incomplete or dynamic environments, such as new roads or construction zones, making it difficult for driving control units to accurately predict vehicle routes and estimate route lengths without up-to-date map data.

Innovation Solution

A method and control unit that establish a grid-based representation of the geographical landscape, determining the vehicle's position and direction at cell borders, and storing these directions in a database to enable dynamic route prediction and length estimation, using a self-learning map system that updates continuously.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If static maps are used for route prediction, then map data is available and consistent, but coverage is limited in areas with new roads or dynamic environments

Engineering Contradiction:
Improvemap coverageVSAvoidroute prediction accuracy
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent transforms static map data into a dynamic system by continuously updating the database with real-time vehicle position and direction information. The grid-based representation allows the map to adapt to new roads and changing environments while maintaining structural consistency, resolving the contradiction between map coverage adaptability and route prediction reliability.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system implements feedback by using detected vehicle position and direction to continuously update the database. This feedback loop allows the map to learn from actual vehicle trajectories and improve coverage in dynamic environments while maintaining reliable route predictions through accumulated statistical data.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If detailed map data is stored for all possible routes, then route prediction accuracy improves, but memory usage increases

Engineering Contradiction:
Improveroute prediction accuracyVSAvoidmemory usage
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments the geographical landscape into a grid-based representation where only relevant cell border information is stored. Instead of storing complete route data for all possible paths, the system divides the environment into discrete cells and stores only the directional information at cell borders, dramatically reducing memory requirements while maintaining prediction accuracy through statistical aggregation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the parameter representation from storing complete route geometries to storing directional statistics at grid cell borders. This parameter transformation reduces data quantity while preserving essential route information, allowing accurate predictions with minimal memory usage through efficient data compression.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If comprehensive route data is collected and stored, then route prediction becomes more accurate, but system complexity increases

Engineering Contradiction:
Improveroute prediction accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent simplifies system complexity by segmenting the geographical space into a regular grid structure. This segmentation provides a systematic framework for data collection and processing, reducing the complexity of managing comprehensive route data while maintaining prediction accuracy through structured organization of directional information at cell borders.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3335008B1Methods and control units for building a database and for predicting a route of a vehicle
Publication Date: 2021.08.18 SCANIA CV AB
  • EP3335008B1 patent drawingFigure 1
  • EP3335008B1 patent drawingFigure 2A
  • EP3335008B1 patent drawingFigure 2B~2C

AI summary

Methods (400, 600) and control unit (300) for building a database (320) and for predicting a route of a vehicle (100), and estimating length of the predicted route. The method (600) comprises determining (601 ) geographical position of the vehicle (100); detecting (602) a cell border (222A) of a cell (222) in a grid-based representation (200) of a landscape, in a database (320), corresponding to the geographical position; determining (603) that the vehicle (100) is entering the cell (222) at the cell border (222A); extracting (604) a stored driving direction at the cell border (222A) from the database (320); detecting (605) a cell border (232B) of a neighbour cell (232), in the driving direction at the cell border (222A); repeating (606) step (604) and (605); predicting (607) the route of the vehicle (100); and estimating (608) the length of the predicted (607) route by adding an estimated distance through each cell (211, 212, 244) of the predicted (607) route.