Matrice 200 Crash Data Analysis Help

[lhanania]

Hello,

I have a Matrice 200 (no gimbal attachment) that I am using DJI's GS Pro app to fly waypoint missions with. I have a mission set up and I flew the same mission with nothing changed 7 successful times. The mission is over a truck parking lot (stationary trucks) and I did not have a single issue throughout the 7 attempts. On the 8th attempt, the drone crashed into a truck. I have pulled the log file for the flight and I need some help analyze it. I cannot find a reason for the crash in the data and I am quite perplexed as I looked over number of satellites, navigation health, remote control input, etc... and cannot find a reason. The drone was autonomously flying throughout the whole mission except when it crashed I attempted to take over however it was too late. The log file for the flight is in the google drive link below, keep in mind that the mission was run 4 times total (including the crash run) with the drone turned on, therefore there is a small test run (manual) in the beginning, three successful runs (waypoint), and the crash data (waypoint) in the same log file.


Log File: Meet Google Drive – One place for all your files


Please let me know what you guys think. I really appreciate it.

 

[Fly Dawg]

Your link is not to a file.....

 

[lhanania]

Thats weird it takes me to the file when I click on it.

Heres another one, try that.

FLY071.DAT

 

[Fly Dawg]

Just a quick look, and bear with me, because this is the first Matrice Data I have seen. But if I had to take an educated guess. This looks like a VPS issue of some sort. If you notice in the graph below, the VPS signal is in and out and very erratic through all 4 flights. I did notice an Autoland command in the first one, and that's what led me to look at the VPS. If this data is accurate, consider your self lucky that it didn't crash on one of the first 3 in this file. I have seen the VPS do some bizarre things before, I will however ask @sar104 if he has time to take a look as well for you.

 

[sar104]

That was equipment failure of some kind. At 434 seconds the FC loses attitude control:

434.084 : 22135 [L-COMPASS][currt cali(0)] find_ref_index_flag = 0|
434.130 : 22137 [L-FDI][CTRL]: fault on , tilt_ctrl_fail
434.151 : 22138 [L-RC]craft ctrl failed!!!
434.297 : 22145 [L-FDI][CTRL]: fault off, tilt_ctrl_fail
434.485 : 22154 [L-FDI][CTRL]: fault on , tilt_ctrl_fail
434.811 : 22169 [L-FMU/FSM]near ground
435.047 : 22181 [L-FDI][CTRL]: fault on , impact_in_air
435.061 : 22181 [L-FMU/FSM]not near ground
435.146 : 22186 [L-COMPASS][currt cali(0)] find_ref_index_flag = 0|
435.172 : 22187 [L-FDI][CTRL]: fault off, impact_in_air

It can be difficult to determine cause and effect, but the attitude data show the loss of control, starting with an uncommanded backwards pitch at 431.8 seconds that appears to have been due to a rapid motor speed excursion. Then it just gets worse.

 

[sar104]

Just a quick look, and bear with me, because this is the first Matrice Data I have seen. But if I had to take an educated guess. This looks like a VPS issue of some sort. If you notice in the graph below, the VPS signal is in and out and very erratic through all 4 flights. I did notice an Autoland command in the first one, and that's what led me to look at the VPS. If this data is accurate, consider your self lucky that it didn't crash on one of the first 3 in this file. I have seen the VPS do some bizarre things before, I will however ask @sar104 if he has time to take a look as well for you.

View attachment 106494

The VPS excursions are likely just the vehicles under the aircraft.

 

[Fly Dawg]

The VPS excursions are likely just the vehicles under the aircraft.

Possibly so, I neglected to extract the log on this one.....That does explain more....

 

[sar104]

Possibly so, I neglected to extract the log on this one.....That does explain more....

It probably rules out some possibilities but I'm really not sure what caused the failure. The M200 has a more complex flight control system than the Phantoms and Mavics. There are a lot of data to sift through.

 

[clackey]

Hello,

I have a Matrice 200 (no gimbal attachment) that I am using DJI's GS Pro app to fly waypoint missions with. I have a mission set up and I flew the same mission with nothing changed 7 successful times. The mission is over a truck parking lot (stationary trucks) and I did not have a single issue throughout the 7 attempts. On the 8th attempt, the drone crashed into a truck. I have pulled the log file for the flight and I need some help analyze it. I cannot find a reason for the crash in the data and I am quite perplexed as I looked over number of satellites, navigation health, remote control input, etc... and cannot find a reason. The drone was autonomously flying throughout the whole mission except when it crashed I attempted to take over however it was too late. The log file for the flight is in the google drive link below, keep in mind that the mission was run 4 times total (including the crash run) with the drone turned on, therefore there is a small test run (manual) in the beginning, three successful runs (waypoint), and the crash data (waypoint) in the same log file.


Log File: Meet Google Drive – One place for all your files


Please let me know what you guys think. I really appreciate it.

Awesome. Just wanted to make sure you know we recently launched MatricePilots.com for all our Matrice owners.

Check it out here - DJI Matrice Drone Forum