Road Centerline Estimation via Vehicle Telemetry Clustering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for generating high-definition or medium-definition maps for semi-autonomous and autonomous vehicles are resource-intensive, particularly when road centerlines are faded or occluded, making it difficult to determine the delineation between two-way traffic for navigation.

Innovation Solution

A method that aggregates and clusters telemetry data from multiple vehicles to estimate the road centerline by identifying a separator based on heading, using spatial smoothing and machine learning techniques such as logistic regression or SVM, to determine the centerline and indicate it on a map for autonomous navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sensor-based and survey-based mapping approaches are used to determine road centerlines, then the accuracy of navigation maps is improved, but the resource consumption and cost increase significantly

Engineering Contradiction:

Improveroad centerline detection accuracyVSAvoidresource consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent uses vehicle telemetry data (GPS positions, headings, speeds) as a proxy copy of physical road centerline information. Instead of directly measuring road features with sensors and surveys, the system infers centerline locations from aggregated vehicle trajectory patterns, creating a virtual representation of road geometry from operational data

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system leverages the existing operational data collected by vehicles for their primary navigation purpose. The telemetry data is already being generated for vehicle control and monitoring, and this patent repurposes that same data for map generation, eliminating the need for separate dedicated mapping resources

Inventive Principle:

Principle #25Self-service

2Loss of information

If multiple aerial surveys are conducted to determine traffic flow and road features, then the completeness of navigation maps is improved, but the time and resource requirements increase

Engineering Contradiction:

Improvecompleteness of road feature dataVSAvoidmapping time

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The system continuously aggregates telemetry data from vehicles as they traverse the road network during normal operations. This continuous data collection replaces discrete, time-consuming aerial surveys, allowing map information to be progressively refined over time without interrupting vehicle operations or requiring dedicated survey missions

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The patent combines telemetry data from multiple vehicles traversing the same road segments into a unified dataset. By merging trajectories, headings, and positional information from numerous vehicles, the system reconstructs road centerlines and traffic flow patterns that would traditionally require multiple separate aerial survey passes

Inventive Principle:

Principle #5Merging (Combining)

3Reliability

If traditional mapping methods are used for roads with faded or occluded centerlines, then the reliability of navigation data is maintained, but the complexity and cost of the mapping process increase

Engineering Contradiction:

Improvenavigation data reliabilityVSAvoidmapping process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

Instead of directly observing road centerlines with sensors and surveys (top-down approach), the system inverts the approach by inferring centerline locations from vehicle trajectory patterns (bottom-up approach). The centerlines are not directly measured but derived from the aggregate behavior of vehicles that naturally follow road boundaries

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent introduces vehicle telemetry data as an intermediary between the physical road infrastructure and the navigation map. The telemetry data serves as a mediator that translates physical vehicle movements into digital road geometry information, bypassing the need for direct sensor-based road feature detection

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11654918B2Estimation of road centerline based on vehicle telemetry

Publication Date: 2023.05.23 GM GLOBAL TECHNOLOGY OPERATIONS LLC

AI summary

System and methods for estimating a centerline of a road that separates traffic moving in opposite directions include aggregating a data set from each of a plurality of vehicles traversing the road over a period of time as telemetry data. Each data set of the telemetry data indicates a location and a heading. The method includes clustering the data sets of the telemetry data based on the heading indicated by each data set, and identifying a separator to separate the data sets indicating a first heading from the data sets indicating a second heading, opposite to the first heading. The centerline is estimated based on applying a spatial smoothing to the separator.