Traffic Control Feature Labeling Using Vehicle Sensor Patterns

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The generation of highly detailed maps for autonomous vehicles is resource-intensive and time-consuming, making it difficult to scale AV fleets into new regions, and human-generated labels can introduce errors.

Innovation Solution

Utilizing sensor data from mapping vehicles to identify and label traffic control features, including static and dynamic signage and lighting, by analyzing vehicle location, speed, and acceleration patterns, and processing image and point cloud data to determine the presence and type of traffic control features.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If human experts manually label traffic control features in map data, then labeling accuracy can be maintained, but the process becomes resource-intensive and time-consuming

Engineering Contradiction:

Improvelabeling accuracyVSAvoidmap generation efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system enables automated labeling by having the mapping system itself perform the labeling function through machine learning models, eliminating the need for external human experts. The AVs collect sensor data and the system automatically generates labels for traffic control features, making the system self-sufficient in the labeling process.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical process of manual human labeling with an automated computational system using machine learning models. The system processes sensor data from multiple AVs and automatically generates labels for traffic control features, substituting human manual work with automated algorithmic processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If more sensor data is collected from multiple autonomous vehicles to improve map data quality, then labeling accuracy improves, but data processing complexity and resources increase

Engineering Contradiction:

Improvemap data qualityVSAvoiddata processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system merges sensor data from multiple autonomous vehicles that are geofenced to the same location. By combining data from multiple sources, the system improves the quality and reliability of labels for traffic control features while distributing the data collection burden across the fleet.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The machine learning model serves multiple functions: it processes sensor data from different vehicle types, handles various traffic control feature types, and generates standardized labels. This multi-functional approach simplifies the overall system architecture despite handling diverse input data.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12511924B2Automated labeling of traffic control features

Publication Date: 2025.12.30 GM CRUISE HOLDINGS LLC

AI summary

Sensor data obtained from vehicles driving through a particular environment (e.g., a particular city or region being mapped) is used to identify and label traffic control features. Location, speed, and/or acceleration of mapping vehicles can be used to identify intersections that may have traffic control features. Environmental data, such as image data captured by one or more cameras, and point cloud data collected by lidar and/or radar sensors, is used to automatically detect traffic control features at the intersections.