Aerial Vehicle Control Software Testing With Recorded Flight Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for testing aerial vehicle flight software rely on synthetic sensor data, which may not accurately represent actual environmental conditions, leading to unreliable verification of flight software performance.
Innovation Solution
Utilize actual flight data recorded by aerial vehicles during previous flights as input to a UAV test bed, optionally augmented with simulated conditions using machine learning models, to create realistic testing scenarios for flight software.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synthetic sensor data is used for testing flight software, then testing can be performed without actual flights, but the accuracy of environmental condition representation deteriorates
Solution Approach 1:
The patent uses actual flight data as a template to create test flight data copies that replicate real environmental conditions. Instead of generating synthetic data from scratch, the system copies authentic sensor readings, flight parameters, and environmental conditions from actual flights, thereby preserving the accuracy and realism of the testing data while enabling repeated testing without additional actual flights.
2Measurement precision
If actual flight data is used for testing, then environmental condition representation accuracy improves, but testing versatility deteriorates due to limited flight scenarios
Solution Approach 1:
The patent dynamically adjusts and modifies test flight data by introducing simulated environmental variations such as different weather conditions, sensor degradations, and operational anomalies. The system takes actual flight data as a base and dynamically generates multiple variant scenarios by modifying parameters like sensor noise levels, environmental conditions, and flight states, thereby achieving both accuracy and versatility.
Solution Approach 2:
The patent systematically changes key parameters of the actual flight data to create diverse testing scenarios. This includes modifying sensor data parameters (adding noise, simulating failures), environmental parameters (weather conditions, lighting), and operational parameters (flight phases, control inputs). By controlled parameter changes, the system generates a wide range of test scenarios while maintaining the core authenticity of the base flight data.
Data Source
AI summary
Example embodiments may include determining test flight data based on actual flight data that has been captured by a sensor of an aerial vehicle during a previous flight performed by the aerial vehicle in a physical environment. The test flight data may be processed using a software component that forms part of an aerial vehicle control system. An observed performance of the software component may be determined based on processing the test flight data using the software component. A performance metric may be determined for the software component based on comparing (i) the observed performance of the software component to (ii) an expected performance of the software component. The performance metric may be output.