Litchi Autonomous Test Fail: Race Track

You were flying too low, I was scared all the time. You should always consider 10-15' margin for position accuracy.
 
With all due respect the app didn't fail. Your setting of waypoints and taking into account clearances did. Would have been a cool video.


Sent from my iPhone using PhantomPilots mobile app
 
I have no idea what caused the second crash but suspect it has something to do with the first crash. Perhaps the debris that was caught in the guards.

I've flown numerous Litchi waypoint missions. What really caught my eye was the wind on your coat when you were attempting to bring it home. I'm guessing 15 MPH? It's my experience that when flying Litchi missions in the wind the aircraft is challenged to maintain its course between waypoints. And that and flying near obstructions at 30 MPH.

The other thing is the elevation you were trying to fly. If you take off and hover at head-height and check the elevation that your machine thinks it at, then fly off a hundred yards with some elevation change, and then return and hover at head-height again, you'll see that the craft doesn't think it's at 6'.

Anytime I program an "iffy" waypoint mission in, I'll do the test flight in a conservative manner with my left index finger ready to abort. Keep in mind you're using a map that's created from a collage of photographs taken over a year ago and then stitched together with software (thus the coordinates can be a few meters off). And the vegetation is at least a year older. If fact, I don't see the white observation towers that your bird didn't hit in your mission map.

SB
 
Pilot understanding of Litchi: Fail

Pilot understanding of civilian GPS: Fail

Pilot understanding of inertia on AC: Fail

It appears maybe a rotor loosened after hitting the branches. Second crash looked like loss of a prop or engine.

Overall assessment: Pilot error, not failure of Litchi.

Don't feel too bad. We all screw up.
 
  • Like
Reactions: Sagebrush
Whatever you did - it is not the Litchi app that failed here. I have flown many complex waypoint missions with both Autopilot and mostly Litchi. I NEVER found fault in the app. It always turned out to be an issue with my planning or mission definition. The Phantom will reach a particular waypoint within the margin of GPS accuracy aided by barometric altimeter. This will almost always put the drone within a sphere of about 30 feet of diameter. If you cut it closer than that - you are taking a chance and it may go bad. Here is an example of a complex Litchi mission in difficult terrain and significant elevation changes. Check out this Youtube video of one of those missions:




Sent from my iPad using PhantomPilots mobile app
 
So, if Lichi isn't the right tool for the job, what would be? It's a cool idea. Would waypoints be better? Autopilot? It doesn't seem like this would be unreasonably difficult.
 
So, if Lichi isn't the right tool for the job, what would be? It's a cool idea. Would waypoints be better? Autopilot? It doesn't seem like this would be unreasonably difficult.
I don't think there is a product out there that that will do this at the accuracy required combined with the aircraft speed and altitude. There is a lot of combined error, between GPS error, barometer error, etc.

Perhaps multiple takes, done at a slower speed and then sped up in your editor of choice will allow for a finished product that conveys what he is trying to convey. But I seriously doubt that you are going to get it in one take, on a windy day, at full speed with today's technology at this price point. Perhaps a Matrice 600 is up to that challenge, but I don't think a Phantom is.
 

Recent Posts

Members online

No members online now.

Forum statistics

Threads
143,086
Messages
1,467,525
Members
104,965
Latest member
cokersean20