HD Map Special Vehicle Detection With Targeted AV Warnings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems struggle to recognize and provide instructions for special vehicles, such as oversized loads, which require unique operating procedures and pose safety challenges due to their size and weight.
Innovation Solution
A system comprising connected devices, machine learning models, and a location cloud platform that acquires sensor data to detect special vehicles, determines their attributes, and generates warning messages for nearby connected vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing navigation systems are used, then general navigation functionality is provided, but they cannot recognize or provide instructions for special vehicles such as oversized loads
Solution Approach 1:
The patent introduces an intermediary component - a special vehicle detection module with machine learning models - that sits between the standard navigation system and the sensor data. This intermediary specifically detects special vehicles and generates appropriate navigation instructions, allowing the system to handle special cases without completely redesigning the entire navigation system.
Solution Approach 2:
The navigation system is segmented into standard navigation functionality and a specialized special vehicle detection module. This segmentation allows the system to maintain general navigation capabilities while adding specific functionality for special vehicles independently, reducing the complexity burden on the overall system.
2Measurement precision
If sensor data is collected from connected vehicles to detect special vehicles, then detection accuracy is improved, but data processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by continuously collecting and pre-processing sensor data from connected vehicles even when no special vehicle is currently detected. Machine learning models are pre-trained and ready, so when a special vehicle appears, detection can occur rapidly without requiring extensive real-time processing, thus reducing the perceived processing time.
Solution Approach 2:
The machine learning models automatically process and analyze sensor data without requiring manual intervention or complex real-time computational resources. The system serves itself by having the trained models independently identify special vehicles and generate instructions, reducing the computational burden on central processing systems.
3Reliability
If warning messages are generated and distributed to connected vehicles, then road safety is improved, but communication bandwidth and system load increase
Solution Approach 1:
Warning messages are generated with local quality by providing specific, targeted information only to connected vehicles that are in proximity to or potentially affected by the special vehicle. Rather than broadcasting to all connected vehicles system-wide, the notification is localized to relevant vehicles, reducing unnecessary communication overhead while maintaining safety.
Solution Approach 2:
The system applies partial action by generating warning messages only when special vehicles are detected and only for vehicles within a relevant distance threshold. This partial application of the warning function reduces the total quantity of communication data while still providing adequate safety coverage for affected vehicles.
Data Source
AI summary
System and methods for systems for special vehicle detection and notifications. Connected vehicle probe and sensor data is acquired and processed to detect the presence of a special vehicle using one or more machine learning models. Attributes are determined for the special vehicle based on additional information. An informational message/notification is composed and sent to vehicles in the vicinity of the special vehicle. Warnings may be provided for how to maneuver vehicles around the special vehicle.