Detecting Barrier Crossings via Trajectory Density Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing digital map systems are ineffective in detecting barrier crossing locations and schedules, particularly for ferry boats and other convoyed objects, due to inaccurate or incomplete data, and lack of methods to exploit the repetitive nature of these crossings.

Innovation Solution

A method to determine barrier crossing information by analyzing historic trajectory data, bundling trajectories with similar geographical and directional properties, and using frequency analysis to identify convoyed objects and derive crossing schedules and frequencies from trajectory density changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional digital map systems are used to detect barrier crossings, then the system structure remains simple, but the detection accuracy and completeness of barrier crossing locations and schedules deteriorates

Engineering Contradiction:
Improvedetection accuracy of barrier crossing locations and schedulesVSAvoidsystem complexity for analyzing trajectory data
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the complex task of barrier crossing detection into multiple independent processing stages: trajectory data collection, trajectory bundling by spatial and temporal proximity, density analysis, frequency analysis, and barrier crossing identification. Each stage processes a specific subset of data with a specific algorithm, making the overall complex system manageable and modular.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms 2D spatial trajectory data into 3D space-time data by adding the time dimension. Trajectories are analyzed not just in their spatial path but also in their temporal pattern, creating a four-dimensional representation (x, y, z, t) that enables detection of repetitive crossing patterns that would be invisible in traditional 2D map analysis.

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

2Reliability

If trajectory data from multiple probes is collected and analyzed to improve map accuracy, then the completeness of barrier crossing information improves, but the computational complexity and data processing requirements worsen

Engineering Contradiction:
Improvecompleteness of barrier crossing informationVSAvoidcomputational complexity of trajectory analysis
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges multiple trajectory datasets from numerous probes into unified trajectory bundles by combining trajectories that are spatially close and temporally proximate. This consolidation reduces the number of individual data streams that must be processed separately, making the computational load manageable while maintaining comprehensive coverage through the aggregation of multiple data sources.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary processing of trajectory data by pre-bundling trajectories into spatial and temporal groups before conducting the complex frequency analysis. This preliminary organization of data reduces the computational complexity of subsequent analysis by working with pre-grouped data rather than raw individual trajectory records.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If existing digital maps are updated with inferred road geometries and features from trajectory data, then the accuracy of navigation information improves, but the time required for map updates and data processing increases

Engineering Contradiction:
Improveaccuracy of navigation informationVSAvoidtime required for map updates
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent enables the system to automatically detect barrier crossings, infer road geometries, and update digital maps without requiring manual intervention from map editors. The automated detection algorithms process trajectory data and directly generate map updates, eliminating the time-consuming manual annotation process while maintaining high accuracy through statistical analysis of trajectory patterns.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2583062B1Detecting location, timetable and travel time estimations for barrier crossings in a digital map
Publication Date: 2017.08.09 TOMTOM GLOBAL CONTENT
  • EP2583062B1 patent drawingFigure 1~2
  • EP2583062B1 patent drawingFigure 3~5
  • EP2583062B1 patent drawingFigure 6A~6B

AI summary

A method for determining barrier crossing information for convoyed objects (22) using historic trajectory data (28). Trajectories (28) having similar geographical and directional properties are bundled so that trajectory density can be measured as a function of position and time (s, t). Visual presentation of the trajectory information can be used to determine certain types of barrier crossing information useful in a digital map. Frequency analysis on a number of trajectory density time series may be performed to determine specific barrier crossing locations (26) through the detection of vehicle bursts. Such frequency analysis may also indicate barrier crossing times and schedules in the case of crossing patterns.