|
I recently traveled from the east coast of the USA to ZZZ to perform a large-scale drone mapping mission. In this ASRS report, I want to focus on the importance of "Fatigue" in IMSAFE (Illness, Medication, Stress, Alcohol, Fatigue, Emotion) so other drone pilots can learn from my outright mistakes.
The hastily put together plan was for me was to connect with a local / resident drone team in ZZZ and arrange to fly several 400' AGL missions in this otherwise 100' LAANC class Delta. The local team was in possession of a standing LAANC NOTAM / waiver that allowed flights up to 500' AGL. The plan was that they would extend this waiver for me to use under their supervision. Given the size of the region we wanted to map (an entire college campus) a 400' AGL flight plan would make this mission a relative walk-in-the-park. At least that was the plan.
Upon arriving, things started to fall apart quickly. The word from my corporate office was that there were complications with connecting with the local team. While corporate would continue to work on the problem, I should begin the data collection myself. AKA: start collecting data on a structure-by-structure basis at 100' AGL… which, in fairness, was the backup plan.
At this point, the correct action on my part should have been to either ask for additional resources to help with the mission execution (Visual Observers) or for me to wait in my hotel room until the necessary connections were made. While that sounds fine on paper, the reality is that between the jet lag induced fatigue and the stress of the severe blackeye the drone program could suffer, I decided to be a good soldier and press on.
In retrospect, what amazes me is that, despite my training as a Part 61 pilot and extensive work with the FAASTeam related to flight safety, at the time it never occurred to me that the timezone induced (F)atigue and the corporate (S)tress are core elements in IMSAFE. It just never crossed my mind at the time.
Independent of IMSAFE, an additional complication was the local terrain. Being from the East Coast of the USA, I am basically a Flat Land pilot. Mountain flying is not something we need to worry about in my region. The area I was to fly, the campus, was extremely mountainous. While I had done several walk throughs of the campus via Google Earth, the sheer elevation changes were daunting in person.
Regardless, I pressed forward flying mission after mission at or about 100' AGL - the "about" is what I will be focusing on. Complicating matters was the fact that the elevation change between my take-off point and the structure of interest was not always the same - something I had not thought about up front. Therefore, I resorted to guestimating the elevation differences. For example, if it looked as if the ground around the building was 50' higher than my take off point, I would program the drone to fly at 150' AGL. The thought being that I would be 100' AGL in the vicinity of the building.
Clearly this approach was flawed from the start. Not only could I not guarantee that my flights were at or below 100' at all times, the process was subject to errors. For example: I realized I had flown one mission at 150' AGL relative to my takeoff location (and in theory 100' AGL over the structure). The next mission I intended to fly at 100' AGL, but I forgot to change the hard-coded altitude. That means that I flew that next mission at 150' AGL - 50' above the LAANC limit.
Eventually, I resolved my fatigue issue via natural means (aka sleep) and came to grips with my faulty decision making. As a result, I stopped flying missions using guestimates of altitude. Ultimately, I was able to connect with the local team and complete the remainder of the data collection at 400' under their waiver. They also taught me how to exploit the autonomous AGL based flight planning in the drone. Ironically, I had known about the AGL feature, but being from a relatively flat region of the world, I had never used it and simply forgot about this feature. Had I used this option, I could have completely avoided my guessing the AGL flight levels around the structures at elevation.
Take aways:
1) We talk about the fatigue aspect in safety meetings, but its insidious nature was a wake-up call. Do not assume that you will be able to detect truly excessive fatigue (especially jet lag induced) while in that state - instead plan for it.
2) Do not underestimate the impact that fatigue and stress can play in your mission planning. The next time I am in such a position, I will explicitly bake in time for the on-the-ground team to accumulate to the new environment. The thought of "hitting the ground running" when extensive multi-time-zone travel is involved is wrong headed.
3) Sending one-person (aka one brain) out on a large mission was a mistake. This was somewhat understandable given that we were supposed to connect with an onsite team, but when that failed to materialize, that left one sleep deprived individual with no one to help with mission planning or as a general CRM (Crew Resource Management) voice.
4) By factoring in up front decompression time, you can better set expectations back in the corporate office and thus reduce stress on the in-field team.
5) If the plan is to connect with a team that has a waiver, it is mandatory that we connect with the team before departing. Launching blindly in the hopes that all the pieces would come together in the field was a recipe for failure.
6) Finally, just like in an aircraft where the pilot is expected to know how to use every piece of installed equipment, I should not have departed without doing my homework regarding the automated AGL flight plan support in the drones.
In conclusion I am disappointed in myself in how this mission unfolded. Yet, at the same time, I feel I am walking away with a significantly new and deep appreciation for the impact that [fatigue] can have both on my drone and crewed flight operations.
| 
