Autonomous Vehicle Object Grids Across Edge Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in sharing accurate and timely object maps due to delays in wireless communication, format inconsistencies, and legal liabilities associated with inaccurate data, which hinders widespread acceptance and safety.
Innovation Solution
A method and system for updating and sharing autonomous system grids (ASGs) across edge nodes, using scene descriptions from autonomous vehicles, non-AV sensors, and cloud updates, with outlier detection and confidence scoring to ensure accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If object maps are shared between autonomous vehicles using wireless communication, then information availability is improved, but data staleness increases due to communication delays
Solution Approach 1:
The system segments the geographic area into multiple zones and creates separate object maps for each zone. Edge nodes maintain local object maps for their respective zones, allowing vehicles to access region-specific information without requiring complete data transmission across the entire network, thereby reducing communication delays and data staleness.
Solution Approach 2:
Edge nodes proactively generate and maintain object maps for their zones in advance, continuously updating them with local sensor data before vehicles need the information. This preliminary preparation ensures that when vehicles request object maps, they receive current data without experiencing communication delays.
2Adaptability or versatility
If object maps are shared across the network, then collaborative awareness is improved, but format inconsistencies and accuracy issues worsen
Solution Approach 1:
The system establishes a universal object map data structure and communication protocol that all edge nodes and vehicles adhere to. This standardized format enables different vehicles and edge nodes to exchange object map information efficiently while maintaining consistent data accuracy and interpretation across the entire network.
Solution Approach 2:
The system implements feedback mechanisms where vehicles report detection results and edge nodes validate object map accuracy. This continuous feedback loop allows the network to identify and correct format inconsistencies and accuracy issues, improving overall data quality while maintaining collaborative awareness.
3Reliability
If object maps are shared widely, then safety improvements are achieved, but legal liability uncertainties increase
Solution Approach 1:
Edge nodes serve as intermediary authorities that generate, validate, and sign object maps for their zones. This intermediary role creates a clear chain of responsibility where the edge node is legally accountable for the accuracy of object maps in its zone, eliminating uncertainty about liability when vehicles use shared object map data for safety decisions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments are disclosed for sharing classified objects perceived by autonomous vehicles. In an embodiment, a method comprises: obtaining, from an autonomous vehicle (AV), a first scene description, the first scene description including one or more classified objects detected in a first zone of a geographic area; updating, by a first edge node in the first zone and using a plurality of scene descriptions, an autonomous system grid (ASG) for the first zone; and sending, by the first edge node, the updated ASG to a second edge node located in the first zone or in a second zone of the geographic area.