Road Sign Type Classification Using Sensor Data and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation systems rely on limited probe vehicles for collecting and updating road sign data, which is not real-time and may not be accurate, especially considering hourly or weather-based changes, leading to potential navigation errors in automated driving applications.
Innovation Solution
A system that uses sensor-equipped user vehicles to collect and analyze data, employing machine learning techniques to predict and update map data, including road sign information, with features like exclusivity score, mean-of-highest, and estimated-vertical-offset, to differentiate between static and mechanical variable road signs, ensuring accurate navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If probe vehicles are used to collect road sign data, then map data can be obtained, but the data cannot be updated in real-time and may not be accurate
Solution Approach 1:
The patent transforms ordinary user vehicles into multi-functional data collection units by equipping them with sensors (cameras, LIDAR, GPS) that serve both their original navigation purpose and the additional function of collecting road sign data. This universal approach replaces the specialized probe vehicle system, enabling real-time data collection from a large population of vehicles simultaneously.
Solution Approach 2:
The system creates a virtual replica of the physical road sign environment by using sensors to capture images and data, then processes these copies through image recognition and machine learning algorithms to extract road sign information. This copying approach allows accurate reproduction of road sign data without requiring physical presence at every location.
2Measurement precision
If map data is updated frequently to reflect real-time changes, then navigation accuracy improves, but data collection and processing complexity increases
Solution Approach 1:
The patent divides the complex task of road sign data collection and processing into distinct segments: sensor data acquisition, image preprocessing, feature extraction using machine learning, road sign classification, and map data integration. This segmentation allows each component to be optimized independently and processed in parallel, reducing overall system complexity while maintaining high precision.
Solution Approach 2:
The system introduces an intermediary processing layer between raw sensor data and the final map database. This intermediary layer includes image recognition algorithms, machine learning models for feature extraction, and data validation mechanisms that filter and refine data before storage. This intermediary structure manages complexity by handling processing tasks in a standardized, modular fashion.
3Productivity
If sensors are installed on user vehicles to collect road sign data, then real-time data availability improves, but system implementation complexity increases
Solution Approach 1:
The patent merges the road sign data collection function with the existing user vehicle navigation systems. By integrating sensors and processing algorithms into vehicles that users already operate for navigation purposes, the system achieves data collection without requiring separate dedicated infrastructure. This merging approach leverages existing vehicle platforms, reducing deployment complexity while maintaining high productivity.
Solution Approach 2:
The system enables user vehicles to automatically perform data collection, processing, and reporting functions without requiring manual intervention or specialized operator training. The onboard sensors continuously capture data, machine learning algorithms automatically process and validate information, and results are transmitted to the map database autonomously. This self-service capability simplifies deployment by eliminating the need for specialized data collection teams.
Data Source
AI summary
A system, a method, and a computer program product for determining a sign type of a road sign are disclosed herein. The system comprises a memory configured to store computer-executable instructions and one or more processors configured to execute the instructions to obtain sensor data associated with the road sign, wherein the sensor data comprises data associated with counts of road sign observations, determine one or more features associated with the road sign, based on the obtained sensor data, and determine the sign type of the road sign, based on the one or more features.