Road Element Detection via Unsupervised Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems face inefficiencies in detecting road elements like roundabouts and junctions, as they rely on predefined characters that may not be present in all instances, limiting detection capabilities.
Innovation Solution
The system employs unsupervised learning to identify actual road elements using a remote computerized system and vehicle-mounted processors, processing a vast amount of road-related information from cheap sensors, and dynamically tracks changes, allowing for reliable detection even without predefined identifiers, with adjustable parameters for selecting relevant identifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If predefined identifiers are used for detecting road elements, then the detection process is simple and fast, but the detection capability is limited and many road elements cannot be detected
Solution Approach 1:
The system performs preliminary actions by collecting and storing road element data from multiple vehicles before detection is needed. This pre-collection of data including images, sensor readings, and location information enables the system to have a rich database of road element characteristics available when detection is required, thus improving detection capability without adding complexity to the real-time detection process
Solution Approach 2:
The patent introduces a remote server as an intermediary between vehicles and the detection system. The server consolidates data from multiple vehicles, performs complex analysis to identify actual road element identifiers, and returns results to vehicles. This intermediary handles the computational complexity centrally, allowing individual vehicles to maintain simple detection processes while benefiting from enhanced detection capabilities through the server's aggregated intelligence
2Reliability
If a vast amount of road related information is processed to identify actual identifiers, then detection reliability is improved, but data transmission requirements increase
Solution Approach 1:
The system extracts only the essential and relevant features from the vast amount of road-related information collected by sensors. Instead of transmitting or processing all raw data, the system identifies and extracts key characteristics such as geometric features, positional relationships, and distinctive patterns that are sufficient for reliable road element identification, thereby reducing data transmission requirements while maintaining detection reliability
Solution Approach 2:
The patent segments the data processing task into multiple stages: initial filtering at the vehicle level to remove obviously irrelevant data, selective transmission of promising candidates to the remote server, and final analysis only on this reduced dataset. This segmentation allows the system to process vast amounts of information reliably while minimizing the quantity of data that needs to be transmitted across the network
Data Source
AI summary
A method for detecting road elements that may include (a) detecting predefined identifiers of road elements, in road related information sensed by vehicles; (b) detecting potential identifiers of road elements that differ from the predefined identifiers of road elements, by processing road related information that was acquired by the vehicles during relevant time windows that are related to the detecting of the predefined identifiers; (c) finding actual identifiers of road elements out of the potential identifiers; wherein the findings is based, at least in part, on road related information that was acquired by the vehicles outside the relevant time windows; and (d) updating a database with the actual identifiers.