UAV Fleet Maintenance Planning With Malfunction Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current maintenance strategies for fleets of unmanned aerial vehicles (UAVs) lack efficiency in predicting optimal maintenance times and optimizing vehicle health, leading to potential malfunctions and reduced utility.
Innovation Solution
A computer-implemented method using logistic regression and generative adversarial networks (GANs) to analyze sensor data from UAVs, predicting malfunction probabilities and generating additional cases for a combined data set, which helps in determining optimal maintenance schedules and improving fleet management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional maintenance strategies are used for UAV fleets, then operational simplicity is maintained, but prediction accuracy of maintenance needs deteriorates
Solution Approach 1:
The system performs preliminary actions by continuously collecting and analyzing sensor data before actual malfunctions occur. Logistic regression models predict malfunction probabilities in advance, enabling preventive maintenance scheduling that prevents failures before they happen, thus improving prediction accuracy while maintaining manageable complexity through proactive rather than reactive maintenance.
Solution Approach 2:
The patent uses generative adversarial networks to create synthetic malfunction data that copies and extends real malfunction patterns. This synthetic data augmentation technique improves prediction accuracy by providing more training examples without requiring additional physical sensors or complex hardware modifications, thereby resolving the contradiction between accuracy and complexity.
2Reliability
If preventive maintenance is scheduled based on predictions, then reliability of UAV fleet is improved, but loss of operational time increases
Solution Approach 1:
The maintenance scheduling system dynamically adjusts maintenance timing based on real-time sensor data and predicted malfunction probabilities. Instead of fixed schedules, the system adapts maintenance intervals to actual UAV conditions, performing maintenance only when prediction thresholds are met. This dynamic approach improves fleet reliability by targeting high-risk vehicles while minimizing unnecessary downtime for low-risk vehicles.
Solution Approach 2:
The system changes the parameter of maintenance timing from static/scheduled to dynamic/prediction-based. By using logistic regression outputs and synthetic data insights to adjust maintenance parameters adaptively, the system achieves higher reliability without proportional increases in downtime, as maintenance is performed only when and where needed based on actual risk levels.
3Measurement precision
If more sensor data is collected from UAVs, then prediction accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent employs generative adversarial networks to copy and synthesize additional malfunction cases from existing sensor data patterns. This synthetic data generation approach improves prediction accuracy by augmenting the training dataset without requiring proportional increases in physical sensors or data collection infrastructure, thereby managing processing complexity while enhancing predictive capabilities.
Solution Approach 2:
The system introduces an intermediary layer of synthetic data generation between raw sensor data collection and final prediction modeling. The GAN-based synthetic data acts as a mediator that enriches the dataset without directly increasing the complexity of sensor hardware or requiring linear increases in data processing infrastructure, thus resolving the contradiction between accuracy and processing complexity.
Data Source
AI summary
A computer-implemented method includes: receiving, by a computer device, sensor data for a plurality of UAVs in a fleet of UAVs; applying, by the computer device, logistic regression to the sensor data; predicting, by the computer device, a probability of malfunction of each UAV in the fleet of UAVs based on the applying; combining, by the computer device, the probability of malfunction of each UAV with a pre-existing malfunction data set to produce an intermediate malfunction data set; generating, by the computer device, additional cases of predicted UAV malfunctions with a GAN, the GAN using the intermediate malfunction data set as initial training data for the GAN; combining, by the computer device, the additional cases with the intermediate malfunction data set to produce a combined malfunction data set; and comparing, by the computer device, the sensor data for a first UAV of the UAVs to the combined malfunction data set.