Traffic Sign Recognition for Unfamiliar Signs in Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in responding to traffic signs, especially unfamiliar ones, as their perception systems may not accurately identify sign types without pre-mapped data, leading to potential safety issues and travel delays.
Innovation Solution
The method involves real-time detection of traffic signs using sensors, analyzing attributes like type, color, shape, and content, and using machine learning models to determine appropriate actions, even for signs not previously identified or stored in the vehicle's map information, allowing for dynamic response and updating of mapping data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the vehicle uses a sign type model to identify traffic signs, then the identification speed is improved, but the accuracy deteriorates for unfamiliar signs
Solution Approach 1:
The identification process is divided into two segments: first, the sign type model provides rapid initial classification; second, for signs with low confidence scores (unfamiliar signs), a more detailed attribute-based analysis is performed. This segmentation allows fast processing for common signs while applying more rigorous analysis only when needed.
Solution Approach 2:
A confidence score mechanism acts as an intermediary between the sign type model and the final identification result. When the confidence score falls below a threshold, the system activates additional verification processes including attribute extraction and comparison with known sign databases, thereby improving accuracy for uncertain cases without impacting speed for confident identifications.
2Measurement precision
If the vehicle relies on pre-mapped traffic sign data, then the identification accuracy is improved, but the adaptability to new or temporary signs deteriorates
Solution Approach 1:
The system dynamically adjusts its identification strategy based on the familiarity of the detected sign. For signs matching pre-mapped data, the system uses stored information for accurate identification. For unfamiliar signs, the system transitions to real-time attribute analysis and comparison with general traffic sign knowledge bases, enabling adaptation to new, temporary, or relocated signs while maintaining efficiency for known signs.
3Measurement precision
If the vehicle performs detailed attribute analysis for all traffic signs, then the identification accuracy is improved, but the processing time increases
Solution Approach 1:
The system applies partial action by performing detailed attribute analysis only for signs that fall into the unfamiliar category (low confidence scores). For familiar signs with high confidence scores, the system uses streamlined identification pathways. This selective application of detailed analysis reduces overall processing time while maintaining high accuracy for the critical subset of unfamiliar signs.
Data Source
AI summary
Aspects of the disclosure relate to determining a sign type of an unfamiliar sign. The system may include one or more processors. The one or more processors may be configured to receive an image and identify image data corresponding to a traffic sign in the image. The image data corresponding to the traffic sign may be input in a sign type model. The processors may determine that the sign type model was unable to identify a type of the traffic sign and determine one or more attributes of the traffic sign. The one or more attributes of the traffic sign may be compared to known attributes of other traffic signs and based on this comparison, a sign type of the traffic sign may be determined. The vehicle may be controlled in an autonomous driving mode based on the sign type of the traffic sign.