Arithmetic Device Target Grouping for Sensor Fusion Load Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In-vehicle driving assistance systems face increased processing loads as the number of sensors and targets to be recognized increases, leading to inefficiencies in sensor information fusion processing.
Innovation Solution
An arithmetic device with a reception unit, processing-unit allocation unit, association unit, and state fusion unit that allocates targets to groups, associates detection targets with tracking targets, and fuses their states to produce a tracking target, enabling parallel processing and accelerating recognition processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of sensors and targets to be recognized is increased, then the recognition capability of external situations is improved, but the processing load of the sensor information fusion processing unit increases
Solution Approach 1:
The patent divides the processing of multiple targets into multiple groups, where each processing unit handles a specific group. This segmentation allows parallel processing of different target groups, reducing the overall processing load on the sensor information fusion processing unit while maintaining the ability to recognize all targets.
Solution Approach 2:
The patent introduces a new dimension of processing by organizing targets into multiple groups and assigning different processing units to handle different groups. This dimensional organization transforms the processing architecture from sequential to parallel, enabling simultaneous handling of multiple targets without increasing the processing load on a single unit.
2Measurement precision
If the number of sensors is increased, then the recognition accuracy of external situations is improved, but the processing complexity increases
Solution Approach 1:
The patent segments the complex processing task into smaller sub-tasks by dividing targets into groups and assigning them to different processing units. Each processing unit handles a subset of targets independently, reducing the complexity that any single unit must manage while collectively maintaining high recognition accuracy across all sensors.
Solution Approach 2:
The patent performs preliminary grouping of targets before detailed processing. By pre-organizing targets into groups and assigning them to specific processing units, the system reduces the complexity of real-time processing while maintaining accurate recognition of all sensor data.
3Speed
If the processing load is reduced, then the processing speed is improved, but the recognition completeness of targets may be compromised
Solution Approach 1:
The patent segments the target set into multiple groups that are processed in parallel by different processing units. This segmentation enables faster processing through parallel execution while ensuring recognition completeness because each target belongs to exactly one group and is processed by a dedicated unit, preventing any target from being overlooked.
Solution Approach 2:
The patent creates a universal processing framework where multiple processing units can simultaneously handle different target groups using the same processing logic. This multi-functional approach maintains consistent and complete recognition across all targets while achieving high processing speed through parallel execution.
Data Source
AI summary
An arithmetic device includes: an input unit to which information pertaining to a detection target that is a target recognized by a sensor is input from the sensor; a processing-unit allocation unit that is configured to allocate a plurality of the targets to any one of a plurality of groups; an association unit that is configured to retrieve a second one of the targets to be associated with a first one of the targets from a part of the plurality of groups included in the plurality of groups; and a state fusion unit that is configured to fuse the first target and the second target that have been associated with each other by the association unit to produce a tracking target that is a target being tracked. The first target is any one of the detection target and the tracking target. The second target is any one of the detection target and the tracking target.