Vulnerable Road User Data for Vehicle Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in efficiently detecting and mapping vulnerable road users (VRUs) over wide geographic areas, which is resource-intensive and difficult due to dynamic movements and fleeting appearances, posing a challenge for real-time applications like autonomous driving.
Innovation Solution
A computer-implemented method and apparatus that determine road links with VRUs by querying a geographic database, providing notifications to activate or deactivate autonomous driving modes or vehicle sensors based on VRU attributes meeting threshold criteria, using a cloud-based platform to collect and process sensor data and VRU messages for generating and publishing VRU data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If VRU detection and mapping is performed over wide geographic areas using traditional methods, then comprehensive VRU coverage is achieved, but resource consumption increases significantly
Solution Approach 1:
The system segments the geographic area into discrete road links and assigns VRU attributes to each segment. Instead of continuously monitoring entire geographic areas, the system divides the space into manageable units (road links) and selectively queries VRU data for specific segments based on vehicle location and trajectory, thereby reducing overall resource consumption while maintaining comprehensive coverage capability.
Solution Approach 2:
VRU attributes are pre-computed and stored in the geographic database for each road link before vehicles need this information. The system performs preliminary detection and mapping efforts to populate the database with VRU presence indicators, risk levels, and historical data, so that vehicles can quickly query and use this pre-processed information without performing resource-intensive real-time detection across entire geographic areas.
2Reliability
If continuous VRU monitoring is implemented for autonomous driving safety, then collision risk is reduced, but system resource usage increases
Solution Approach 1:
The system dynamically adjusts VRU monitoring intensity based on vehicle context. Instead of continuous monitoring at full capacity, the system queries VRU attributes selectively based on vehicle location, intended trajectory, speed, and environmental factors. Monitoring intensity is modulated to match actual risk levels, maintaining collision avoidance capability while optimizing system efficiency through adaptive resource allocation.
Solution Approach 2:
The geographic database automatically updates and maintains VRU attributes for road links using data from multiple sources including vehicle reports, infrastructure sensors, and historical records. This self-service mechanism reduces the computational burden on individual autonomous vehicles, as the heavy lifting of VRU detection and database maintenance is performed collectively by the ecosystem rather than each vehicle independently.
3Loss of information
If detailed VRU data is collected and stored for all road segments, then situational awareness is improved, but data storage requirements increase
Solution Approach 1:
The system stores detailed VRU attributes selectively for different road link types based on their specific safety requirements. High-risk areas such as pedestrian zones, school areas, and intersections receive more granular VRU data including presence indicators, historical encounter frequencies, and risk level classifications. Lower-risk road segments receive simplified attribute sets, optimizing the balance between situational awareness and storage requirements through localized data density adjustment.
Data Source
AI summary
An approach is provided for operating a vehicle using vulnerable road user data. The approach, for example, involves determining a road link on which the vehicle is traveling or expects to travel. The approach also involves querying a geographic database for a vulnerable road user attribute of the road link. The approach further involves providing a notification to activate or deactivate an autonomous driving mode of the vehicle, a vehicle sensor of the vehicle, or a combination thereof while the vehicle travels on the road link based on determining that the vulnerable road user attribute meets a threshold criterion.