Started drafting "investigation reports" for my incidents/crashes

[slup]

...Just wondering if DJI drones are designed to right themselves after hitting wires, and continue to fly, if they're not damaged...

All drones have that ability... it's the basic foundation when it comes to drone systems & how they are managed.

Drones use a so called PID controller (proportional integral derivative) & a control loop feedback mechanism to control process variables. The P is the "pusher" which starts the corrective action, D is the "damper" & prevents overshoot & I is kind a long term pusher/damper preventing attitude drift.

The main mission for the PID controller is to make sure the drone do what you command with the sticks... when it comes to "photo drones" with GPS, other commands besides stick moves come in play... GPS for positional hold, barometric sensors for height hold... and code commands coming from stored waypoint missions.

So if the drone suddenly moves or change attitude uncommanded, various sensors will tell the flight controller that the drones deviates from the "setpoint" (sticks, GPS, barometric sensor, code) & will try to command the motors to get back to the "setpoint".

The thing that decides if the drone will "get back on track" is usually either possible damage in the propulsion system or distance to ground/objects as it will take some time to correct a drone that just have hit a guy wire in full speed.

 

[rjwmorrell]

I had three incidents where I hit wires. The first was close to home, and the mavic 3 hit a 75 ohm video cable that was crossing the street. Fortunately, it didn't crash and recovered by itself. It did have one propeller damaged that I replaced.. The second I'm not sure about, but I think it hit a guy wire and I never found it. The third it hit a guy wire at 240 feet and crashed into a tree. I was able to recover it, and have it repaired. Just wondering if DJI drones are designed to right themselves after hitting wires, and continue to fly, if they're not damaged. Probably. When you say "but control was regained", was that by something you did, or something the drone did?

The flight was unstable, and then it wasn't. No prop damage. I did nothing to intervene other than try to gently steer it away from the wires.

I can only guess that the drone's computer corrected itself and was able to continue.

I've read that even with one wonky prop or motor, a quadcopter can often maintain stability, with the modern ESC's making compensatory adjustments. I think the issue is if the drone tips over too far getting hung up or snagged on a wire (or branch, etc) that it goes more than 90 degrees roll or tilt, that could shut down the motors (because they are no longer providing lift at that point), sort of like when you hand catch and hold them upside down to trip the motors off.

 

[rjwmorrell]

All drones have that ability... it's the basic foundation when it comes to drone systems & how they are managed.

Drones use a so called PID controller (proportional integral derivative) & a control loop feedback mechanism to control process variables. The P is the "pusher" which starts the corrective action, D is the "damper" & prevents overshoot & I is kind a long term pusher/damper preventing attitude drift.

The main mission for the PID controller is to make sure the drone do what you command with the sticks... when it comes to "photo drones" with GPS, other commands besides stick moves come in play... GPS for positional hold, barometric sensors for height hold... and code commands coming from stored waypoint missions.

So if the drone suddenly moves or change attitude uncommanded, various sensors will tell the flight controller that the drones deviates from the "setpoint" (sticks, GPS, barometric sensor, code) & will try to command the motors to get back to the "setpoint".

The thing that decides if the drone will "get back on track" is usually either possible damage in the propulsion system or distance to ground/objects as it will take some time to correct a drone that just have hit a guy wire in full speed.

This makes sense. I was slowly lowering my Mini SE last year to line up a shot I was planning, and I got it too close to a road sign.

The sign was aft of the drone, and at least one aft prop made contact. I couldn't see, but I believe the drone tipped over forwards, so nose down attitude, because the rear rotors were blocked by the sign, and then the motors shut down. It fell into a flower bed, luckily, and the only damage were props that were easy to replace.

The drone tried to compensate for the loss of lift to the aft, probably by trying to increase the rotor speed aft and reduce the rotor speed fore, but that just resulted in a motor shutdown when it went nose down.

I'm glad they shut down like that, I was maneuvering between live lanes of traffic and had flashes of fear that it would fly uncontrolled into a windshield of a car.

 

[RodPad]

I think the best analogy is "constructive criticism". What worked, what didn't, and suggestions for improvement.

Thanks for reply!
Yes I would have caught that, I didn't know that word was not same. I see many miss spells and other language slang's.

Yes my wife does all the legal paper work.

Rod ..

 

[Robert Prior]

I think the issue is if the drone tips over too far getting hung up or snagged on a wire (or branch, etc) that it goes more than 90 degrees roll or tilt, that could shut down the motors

I nearly lost my new pre-ordered Mavic 2 Pro on my first trip with it.

I was landing in a constrained area. My fault, I'd taken off from there and hadn't thought much about landing. Did what I always did to land my Phantom: positioned it over the landing spot and held full down on the left stick. It came down, and then hovered for 1-2 seconds about half a metre off the ground. I wasn't expecting that, and while it was hovering a gust of wind moved it sideway and a rotor hit a bush, which stopped the rotor and the drone just flipped 180 degrees and over a cliff above rapids before the motors cut out.

Unfortunately I'd foolishly not got DJI insurance for it. Fortunately I could see it resting on a ledge. Unfortunately it was out of reach. Foolishly I climbed down to get it. Alone. Above rapids. One of the most boneheaded decisions I've made in my life. I was really tired (only 3-4 hours sleep the night before, 2+ hours driving to reach the site) and somehow got focused on the task of retrieving my new expensive drone above all else.

I think DJI programmed the pause to rotate the gimbal to face forward so the camera doesn't get damaged (being so close to the ground). It makes sense, but it was a change I wasn't expecting.

Lessons learned:

• adequate rest is a must
• hours practice on type is important
• landing is harder than taking off
• have alternative landing sites scouted before taking off
• be prepared to lose the drone on every flight (Stoic philosophy is useful here)

 

[rjwmorrell]

I nearly lost my new pre-ordered Mavic 2 Pro on my first trip with it.

I was landing in a constrained area. My fault, I'd taken off from there and hadn't thought much about landing. Did what I always did to land my Phantom: positioned it over the landing spot and held full down on the left stick. It came down, and then hovered for 1-2 seconds about half a metre off the ground. I wasn't expecting that, and while it was hovering a gust of wind moved it sideway and a rotor hit a bush, which stopped the rotor and the drone just flipped 180 degrees and over a cliff above rapids before the motors cut out.

Unfortunately I'd foolishly not got DJI insurance for it. Fortunately I could see it resting on a ledge. Unfortunately it was out of reach. Foolishly I climbed down to get it. Alone. Above rapids. One of the most boneheaded decisions I've made in my life. I was really tired (only 3-4 hours sleep the night before, 2+ hours driving to reach the site) and somehow got focused on the task of retrieving my new expensive drone above all else.

I think DJI programmed the pause to rotate the gimbal to face forward so the camera doesn't get damaged (being so close to the ground). It makes sense, but it was a change I wasn't expecting.

Lessons learned:

• adequate rest is a must
• hours practice on type is important
• landing is harder than taking off
• have alternative landing sites scouted before taking off
• be prepared to lose the drone on every flight (Stoic philosophy is useful here)

Fatigue is mentioned numerous times in many, many TSB reports. I just got a copy of a police service SOP and it includes a warning about working hours and fatigue.

Glad it all worked out!

 

[jaja6009]

I fly for public safety and for my own side gig.
We have a similar accidetn report that must be filed anytime there is and accident, incident or mishap.
We have since moved onto using the NASA ASRS form (Not officially for accidents) , but do not submit it to NASA if its used for an accident.

Further an After Action report will be written for any accident, incident or mishap that caused property damage, injury or close call.

OP, I like it and as others mentioned a lot of large drone service providers and public safety entities use something similar.