Unknown Traffic Object Detection Using Vehicle Speed Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic sign detection systems are unable to accurately identify and categorize a broad variety of traffic objects that do not explicitly depict speed limits, as they lack a globally agreed-upon standard and are influenced by diverse symbols, languages, and expected traffic behaviors.
Innovation Solution
A data-driven approach using machine learning algorithms that analyze sensor data, including vehicle speed changes and geographical positioning, to identify unknown traffic objects by correlating consistent speed adjustments with the presence of such objects, enabling the creation of training data sets for accurate identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional traffic sign detection systems are used, then known traffic signs can be identified, but unknown traffic objects with implicit speed limits cannot be accurately detected and categorized
Solution Approach 1:
The system changes the detection parameters by shifting from explicit visual recognition of known traffic signs to detecting behavioral parameters (speed changes) of vehicles in response to unknown traffic objects. This allows the system to identify implicit speed limit indicators without requiring prior knowledge of specific sign appearances or meanings.
Solution Approach 2:
The system introduces an intermediary detection approach by using vehicle speed changes as a mediator to infer the presence and type of unknown traffic objects. Instead of directly recognizing diverse traffic objects, the system detects the behavioral response (speed adjustment) of vehicles, which serves as an indirect but reliable indicator of traffic object characteristics.
2Adaptability or versatility
If a system attempts to cover all varieties of traffic objects from different languages and regions, then comprehensive detection is achieved, but system complexity increases substantially
Solution Approach 1:
The system achieves universality by detecting a common behavioral pattern (speed change) that applies to all types of traffic objects regardless of their visual appearance, language, or regional variations. This single detection mechanism serves multiple functions: identifying presence, inferring type, and determining relevance, eliminating the need for separate detection systems for each traffic object category.
Solution Approach 2:
The system extracts the essential functional characteristic (speed limit indication) from diverse traffic objects by focusing on vehicle behavioral responses rather than object appearance. This extraction approach isolates the critical information needed for autonomous driving while discarding irrelevant variations in language, symbols, and visual design.
3Reliability
If rule-based systems are used for traffic sign detection, then explicit speed limit signs can be identified, but maintenance becomes difficult for implicit indicators
Solution Approach 1:
The system implements self-service by automatically learning and adapting to new traffic object types through continuous detection of vehicle speed changes. The system improves itself over time by accumulating data on emerging traffic objects and their associated speed patterns, eliminating the need for manual updates and maintaining high reliability without increasing maintenance burden.
Data Source
AI summary
A method for generating training data for a machine learning (ML) algorithm configured for identification of an unknown traffic object present on a road is disclosed. The method includes obtaining sensor data from a sensor system of an ego vehicle travelling on the road, the sensor data including one or more images of a surrounding environment of the vehicle and speed information of the ego vehicle and/or speed of at least one external vehicle. The method further includes determining a presence of the unknown traffic object in the surrounding environment of the ego vehicle and determining a change of speed of the ego vehicle and/or of the at least one external vehicle. In an instance of a co-occurrence of the determined change of speed and the determined presence of the unknown traffic object, the method includes selecting one or more images of the at least one unknown traffic object.