Design and Test of an Autonomy Monitoring Service to Detect Divergent Behaviors on Unmanned Aerial Systems
Technical Report,01 Oct 2019,01 Jul 2020
AIR FORCE INSTITUTE OF TECHNOLOGY WRIGHT-PATTERSON AFB OH WRIGHT-PATTERSON AFB United States
Pagination or Media Count:
Operation of Unmanned Aerial Vehicles UAV support many critical missions in the United State Air ForceUSAF. Monitoring abnormal behavior is one of many responsibilities of the operator during a mission. Somebehaviors are hard to be detect by an operator, especially when flying one or more autonomous vehicles as such,detections require a high level of attention and focus to flight parameters. In this research, a monitoring system andits algorithm are designed and tested for a target fixed-wing UAV. The Autonomy Monitoring Service AMScompares the real vehicle or simulated Vehicle with a similar simulated vehicle using Software in the Loop SITL.It is hypothesized that the resulting design has the potential to reduce monotonous monitoring, reduce risk of losingvehicles, and increase mission effectiveness. Performance of the prototyped AMS model was examined by severalmeasures, including divergence detection rate, synchronization time, and Upper Control Limit UCL of aircraftlocation variability in different scenarios. Results showed 100 rate of divergence detection out of all divergentevents occurred. The weighted mean of AMS synchronization time was 4.02 seconds, and the weighted mean foraircraft location variability was 44.8 meters. The overarching AMS functionality was achieved. AMS supports theconcept that humans and machines should be designed to complement each other by sharing responsibilities andbehaviors effectively, making final system safer and more reliable.
- Pilotless Aircraft
- Human Factors Engineering and Man Machine Systems