V2X-Assisted BEV Signaling for Low-Overhead Cooperative Perception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems, particularly in the context of V2X technologies, face challenges in efficiently managing and minimizing signaling overhead for vehicle-to-everything (V2X) messages related to bird's eye view (BEV) features, which are crucial for cooperative driving and autonomous vehicles.
Innovation Solution
The proposed solution involves transmitting V2X messages with associated BEV feature descriptors and object identifiers, allowing vehicles to report BEV features at a lower periodicity, leveraging V2X information for extrapolation and prediction, and using affine transformations for coordinate alignment, thereby reducing signaling overhead and enhancing cooperative driving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If V2X messages are transmitted frequently to report BEV features, then the accuracy and timeliness of cooperative perception is improved, but the signaling overhead increases
Solution Approach 1:
The patent extracts and transmits only the essential BEV feature descriptors and object identifiers from the complete BEV representation, rather than transmitting all BEV data. This selective extraction reduces the quantity of transmitted information (signaling overhead) while maintaining the accuracy needed for cooperative perception by focusing on the most critical features.
Solution Approach 2:
The patent performs preliminary processing of BEV features at the transmitting vehicle, pre-computing and encoding the feature descriptors and object identifiers before transmission. This preliminary action reduces the complexity and volume of data that needs to be transmitted in real-time, thereby reducing signaling overhead while ensuring accurate perception data is available to receiving vehicles.
2Loss of information
If V2X messages include detailed BEV feature information, then the completeness of environmental perception is improved, but the message size and transmission complexity increase
Solution Approach 1:
The patent segments the BEV feature information into distinct components: BEV feature descriptors and object identifiers. This segmentation allows the system to transmit comprehensive environmental perception data by organizing it into structured, manageable segments that can be efficiently processed and transmitted without overwhelming complexity.
Solution Approach 2:
The patent creates a universal message structure that can convey multiple types of BEV information (features and object identifiers) in a standardized format. This universal structure enables the same message framework to handle diverse environmental perception data types, reducing transmission complexity while maintaining information completeness.
3Speed
If BEV features are reported at high periodicity, then the real-time performance of cooperative driving is improved, but the signaling overhead and energy consumption increase
Solution Approach 1:
The patent transmits partial BEV feature information (selected descriptors and object identifiers) rather than complete BEV data at every transmission interval. This partial action approach maintains real-time performance by providing sufficient information for cooperative driving decisions while reducing the frequency and volume of transmissions, thereby lowering energy consumption.
Data Source
AI summary
In some aspects, a vehicle-to-everything (V2X)-capable vehicle may determine an association between one or more features of a bird's eye view (BEV) space around the V2X-capable vehicle and one or more V2X messages, where the one or more features of the BEV space are associated with the one or more V2X messages based on one or more objects detected in the one or more features of the BEV space being reported in the one or more V2X messages. The V2X-capable vehicle may transmit one or more messages indicating the association and a set of BEV features representing the BEV space other than the one or more features of the BEV space associated with the one or more V2X messages.


