摘要: |
Unmanned Aerial Vehicle (UAV) formation flight and swarming are active areas of research within the Department of Defense (DoD). These areas of study cover multiple engineering disciplines; from mechanical and aeronautical to computer science and human factors. The current use of low cost commercial off the shelf (COTS) components to architect UAV formation flights results in insufficient position accuracy of the UAVs in the formation. Latency in communication between autonomous vehicles degrades formation cohesion. This research aims to demonstrate the position error of formation flights decreases by using onboard sonar sensors to accurately measure the distance the follower UAV is from the leader UAV. The sensor enables the follower UAV to appropriately and quickly respond to errors in position by adjusting the followers velocity. The UAV architecture, using onboard sensors, demonstrated tighter formation cohesion by measuring the average position error during multiple flight tests of UAVs and compare these results with previous formation flight tests that did not utilize onboard sensors. The previous flight tests, used the same guided position algorithm, the same X-8 airframes, but no onboard sensor for real time distance error measurement. Since the previous flight test had a similar configuration. This assessment shows how position error was effected by incorporating the sonar sensor. This research effort was able to reduce the Root Mean Square Deviation (RMSD) by 37.3% and the average position error by 70.9% when compared to the previous flight tests without sonar. |