Cooperative Sensor Data Fusion for Priority Object Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cooperative sensor sharing in vehicles is bandwidth intensive and computationally demanding due to redundant information processing, which hampers efficient use of computational resources.
Innovation Solution
A method that prioritizes processing by identifying redundant sensor information, fusing data from multiple sources, and assigning priority levels based on object relevance and proximity to minimize computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If sensor information is shared from multiple remote vehicles, then information completeness is improved, but computational load increases
Solution Approach 1:
The patent combines sensor information from multiple remote vehicles into a unified dataset, merging redundant data while preserving unique information. This allows the host vehicle to access comprehensive environmental data without processing duplicate information from each source separately, thus improving information completeness while managing computational load.
Solution Approach 2:
The system identifies and discards redundant sensor information that is already available from other sources, while recovering and retaining unique valuable data. By filtering out duplicate object detections and sensor readings, the system reduces computational processing of redundant data while maintaining complete environmental awareness.
2Measurement precision
If all sensor data is processed, then detection accuracy is improved, but processing time increases
Solution Approach 1:
The patent segments sensor data processing by prioritizing objects based on their relevance to the host vehicle. Critical objects such as those in blind spots or close proximity are processed with higher priority and greater detail, while less critical objects receive reduced processing. This segmentation allows accurate detection of important targets while reducing overall processing time through selective attention.
Solution Approach 2:
The system applies partial processing to less critical objects and excessive (detailed) processing to critical objects. By adjusting the level of processing detail based on object priority, the system ensures high detection accuracy for important targets while avoiding unnecessary detailed processing of less relevant data, thus optimizing the balance between accuracy and processing time.
3Reliability
If redundant information is processed, then data reliability is improved, but energy consumption increases
Solution Approach 1:
The patent changes the processing parameters dynamically based on object priority and redundancy level. For redundant information from multiple sources, the system adjusts processing intensity by applying confidence thresholds and priority-based filtering. This allows the system to maintain data reliability through cross-validation of multiple sources while reducing computational energy consumption by lowering processing parameters for redundant data.
4Productivity
If priority-based processing is implemented, then computational efficiency is improved, but system complexity increases
Solution Approach 1:
The system performs preliminary prioritization of sensor objects before detailed processing. By pre-classifying objects based on their spatial relationship to the host vehicle, motion characteristics, and detection confidence, the system establishes a processing hierarchy in advance. This preliminary action reduces the complexity of real-time processing decisions while maintaining high computational efficiency through pre-established priority queues.
Data Source
AI summary
A method includes receiving object data, by a controller of a host vehicle, from a plurality of sources, the plurality of sources including remote objects and a sensor system of the host vehicle; identifying, by the controller of the host vehicle, a target object using the object data from the plurality of sources, the object data including a target-object data, and the target-object data is the object data that specifically pertains to the target object determining, by the controller, that the target-object data is available from more than one of the plurality of sources; and in response to determining that the target-object data is available from more than one of the plurality of sources, fusing, by the controller, the target-object data that is available from more than one of the plurality of sources to create a single dataset about the target object.