Drone Manufacturer: DJI
Drone Model: Inspire 1 Pro
Country: United States of America
Pilot Qualifications: Licensed or Certificated by Aviation Authority Approved by Public Safety Agency or Company
Pilot Flight Experience: 10 Hours
Link to External Information About This submission:
File Uploaded: None
On August 6, 2020 our unit engaged in a joint SWAT/Unmanned Aerial System (UAS) training in Spokane, Washington. The Unit was tasked with supporting SWAT with UAS capabilities. The requests were to fly indoors with our Mavic Mini (Knight 4) as well as to fly exterior with the Inspire 1 Pro (Knight 3). The training with the Inspire was set up to use the Stork System to carry a small SWAT robot (Throwbot/Gopher) up to a structure and deploy the robot onto a raised balcony where the robot may not be able to deploy without transport by the UAS. The robot weighs 1.3 pounds. The Stork System is rated to carry 1.76 pounds and I have observed the Inspire 1 Pro carry upwards of 3.25 pounds.
Current weather for the operation as well as future weather was observed. Winds were going to increase throughout the exercise to around 16 knot sustained winds with gusts to 22 knots around 1800 hours. A Risk Assessment was completed, and the risk was low.
Training began around 1300 hours and the training environment was assessed. The weather was clear, dry and warm. The winds were 5-10 mph. The Inspire 1 Pro was flown several times testing the weight and control of the UAS while carrying the robot. The maneuver was successful in transporting the robot by UAS onto a 2nd floor balcony in the current wind conditions. After testing the ability of the Inspire and noting the wind turbulence created in and around the buildings the PIC involved was advised to continue to assess the potentially increasing winds. The PIC had never flown the Inspire using the Stork System to carry objects prior to this date but has flown the Inspire on multiple occasions.
At about 1800 hours the Inspire 1 Pro crashed while attempting to carry the robot onto a 2nd floor balcony. There were no injuries to persons and there was no damage to any other property other than the UAS. Post-crash the UAS would not shut down by use of the controller. The battery had to be removed to shut it down. There was damage to two motors, all four rotors, the landing gear and shell. At one point one of the motors was smoking. It was estimated the UAS fell from about 20 feet into a set of exterior basement stairs from the ground level and inside a narrow corridor.
Further review of AirData found the following: The maximum distance traveled from the home point was 12 feet and the maximum altitude reached was 7.2 feet above the home point (-16.1 feet landing). The maximum speed was 1.34 mph. The flight lasted 25 seconds. The PIC attempted to change the mode from Program to Attitude mode in an attempt to regain control prior to crash.
The Power (Efficiency/Battery Cells/Cells Graph/Volt and Amps) for this flight and the batteries appeared to be operating normally. There were no notable voltage drops. All sensors (Signal Map/Signal Score/GPS/Compass) appeared to be operating normally other than the GPS sensor. The GPS satellite count was extremely low and listed as dangerous. This could be an explanation as to why the Inspire was not holding position. The Controls (Rudder Response/Rudder Map) were reviewed and it was noted there was not enough data to display any reliable results due to the flight being so short.
The Weather (Ground Weather/Kp Index/InFlight Wind/Wind Map/Altitude Profile) for this flight were reviewed as reported in AirData. The wind at the ground was listed at 10.2 mph at the time of flight. The Kp Index was quiet. The InFlight Wind and Wind Map show the maximum wind gust at 98.4 mph at 35 seconds into the flight. However, the UAS had already crashed at 26 seconds into the flight and the wind measurement at this stage was an error reading. Spokane’s weather on this date at the time of crash (1800 hours) showed wind speeds at 25 mph with gusts to 40 mph. Since the training took place in and around large buildings its possible the PIC did not recognize the true wind issues as they stated in an interview that they did not appear to be an issue at the time, but the Inspire seemed to lose control in the narrow corridor between the brick walls on either side of the UAS probably from the rotor wash bouncing off the walls.
Based on the PIC’s statement, the known weather data and reviewing the flight data it is our opinion the UAS crashed as a result of pilot error. The wind speeds recorded at the time of crash, the turbulence created by the wind and rotor systems from the UAS around the buildings inside the narrow corridor as well as the added weight of the SWAT robot caused the Inspire to crash due to a loss of control by the pilot. Based on the location of the crash the narrow corridor itself created issues that would make it difficult to leave the PIC with any room for error. This error may have been exacerbated by the low satellite count. With a low satellite count the UAS may not hold its hover as expected when no flight control input is given. Included in this assessment is a lack of experience in carrying objects with a UAS.