Flying Object Group Abnormality Detection From Orbit Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to detect coordinated abnormalities in a group of flying objects, such as uncoordinated operations, which are costly and inefficient, focusing instead on individual flying objects.
Innovation Solution
A detection device that generates feature values from orbit data of multiple flying objects, trains a model using these values, and detects abnormalities in the group based on the acquired state, utilizing deep neural networks and dimension reduction techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If monitoring is performed by focusing on a single flying object, then the detection of individual abnormal states is possible, but the cost of monitoring is high and it is difficult to detect uncoordinated operations in a flying object group
Solution Approach 1:
The patent combines monitoring of multiple flying objects into a unified system that detects abnormalities in the entire flying object group. Instead of individually monitoring each flying object, the system integrates data from multiple sources and performs collective analysis to detect uncoordinated operations and group-level abnormalities, thereby reducing overall monitoring costs while maintaining detection accuracy.
Solution Approach 2:
The monitoring system is designed to perform multiple functions: it can detect abnormalities in individual flying objects, identify uncoordinated operations within the group, and assess the overall state of the flying object group. This multi-functional approach allows a single system to replace multiple individual monitoring systems, reducing costs while comprehensively addressing different detection needs.
2Measurement precision
If monitoring focuses on individual flying objects, then individual abnormal states can be detected, but appropriate monitoring of the entire flying object group becomes difficult
Solution Approach 1:
The patent introduces a new dimension of monitoring by adding group-level analysis to the traditional individual object monitoring. The system operates at two levels: individual flying object monitoring and collective group monitoring. This dimensional expansion allows the system to simultaneously maintain precise individual detection while achieving comprehensive group coverage, detecting patterns and uncoordinated operations that span multiple objects.
Data Source
AI summary
A detection device includes a generation unit that generates feature values representing features of a plurality of pieces of orbit data, on the basis of the orbit data representing a state of each flying object included in a flying object group, an inference unit that acquires a value corresponding to a state of the flying object group on the basis of the feature values generated by the generation unit, and a detection unit that detects an abnormality having occurred in the flying object group on the basis of the value corresponding to the state of the flying object group acquired by the inference unit.