Raytheon Company COBRA incident near Colorado Springs, Colorado, July 28, 2008

The NTSB's narrative final · quoted from the NTSB record

Note: This case was reclassified from an accident to an incident as a result of applicable revisions to 49 Code of Federal Regulations Part 830.2, as amended at 75 FR 51955, Aug. 24, 2010. The case was previously identified under accident number DEN08FA130. The unmanned aircraft system (UAS) was being demonstrated to the Air Force Academy for use in its aeronautics program. The unmanned aerial vehicle (UAV) was preflighted and taxied into position on a road for takeoff. The command for automatic takeoff was entered. The UAV then flew an automatic flight profile. After several demonstrated maneuvers, the UAV was directed to enter a rectangular traffic pattern for landing. During the first landing attempt, the UAV undershot the final turn waypoint. The internal pilot, who was in the control van on the ground, manually sent the UAV to the upwind waypoint to re-enter the traffic pattern for another landing attempt. During the second landing attempt, the UAV undershot the final turn again. The internal pilot again sent the aircraft to the upwind waypoint. On the third attempt, the airplane flew the base leg and overshot the final approach by about 89 feet before turning and descending sharply toward the landing point on the road. The UAV flew into the top of an 80-foot-tall stadium light that was displaced to the side of the road on which the aircraft was supposed to land. The UAV was destroyed in the impacts with the pole and the ground. The UAS had been flight tested to determine if the UAV could be operated in the confined airspace at the demonstration site. Subsequently, programmed bank angle and descent angle values were increased based on flight testing before the flight demonstration. The autopilot recorded flight parameter data. An analysis of the recorded data showed that a strong tailwind, the high density altitude, and the increased bank angle contributed to the airplane’s two undershoots. Upon examining the recorded data for the incident flight and previous flight demonstrations, an anomaly was discovered where the UAV would enter a sideslip and momentarily lose control. The UAV would “swing wide” and fly relatively straight for a short period of time, then resume the programmed flight. The operator believes that this anomaly occurred on the third approach. The manufacturer had told the operator that the large tail wind on downwind, combined with the excessive speed due to the steep approach could have resulted in the vehicle to switch early to the base leg. They also said that the approach was too aggressive (too steep and too short) and the vehicle would have been constantly pulling max bank angle and then try to correct upright. The vehilce could not adjust fast enough to satisfy the aggressive approach which resulted in the aircraft being unable to fly the proper ground track.

The complete narrative as the NTSB published it. The NTSB's docket holds the report as a PDF and any photographs, statements and other documents from the investigation.