New to Litchie...I learned a couple important lessons on my second flight plan

[56_kruiser]

This post is re: my second waypoint flight with Litchie. I learned a couple important lesssons:

1. Be sure you know how to take over a flight going wrong, and remember it. After the fact, I remember that to take over I need to switch on Sport Mode...a mode I've never used yet. I had thought that pressing the pause button on my Mavic controller would pause it, but it did not. (I 'tested' that on my first flight...but I think the pause it did was circumstantial, as opposed to pausing when I told it to).
2. You need to test and understand where the drone will fly compared to where you placed the waypoints on the map. I would expect it to be off, but it was a lot fore off than I thought. On my first test flight I thought it flew them pretty much spot on...but it was not a flight requiring as precise placement as my second.

Below is the video showing the waypoints, and where the Mavic actually was, and the scary results.

 

[TimesTen]

This post is re: my second waypoint flight with Litchie. I learned a couple important lesssons:

1. Be sure you know how to take over a flight going wrong, and remember it. After the fact, I remember that to take over I need to switch on Sport Mode...a mode I've never used yet. I had thought that pressing the pause button on my Mavic controller would pause it, but it did not. (I 'tested' that on my first flight...but I think the pause it did was circumstantial, as opposed to pausing when I told it to).
2. You need to test and understand where the drone will fly compared to where you placed the waypoints on the map. I would expect it to be off, but it was a lot fore off than I thought. On my first test flight I thought it flew them pretty much spot on...but it was not a flight requiring as precise placement as my second.

Below is the video showing the waypoints, and where the Mavic actually was, and the scary results.

Tip # 1 never ever fly under 200’ GPS is + or - 3 feet accurate.



-G

 

[56_kruiser]

Tip # 1 never ever fly under 200’ GPS is + or - 3 feet accurate.



-G

Not sure I understand. Are you saying to not fly under 200' altitude?

After really thinking about it, I do expect the GPS to be off by a bit. It was off by about 10'. I flew the first flight at about 40' when I tested, and couldn't see that it was that far off...was a bit surprised at how accurate it 'seemed'.

 

[Pedro]

Am I missing something? You never even got into GPS mode so how would you expect Litchi to run a successful a mission? The most satellites acquired was 9 I think but most of the time it was less than that..

 

[53-63-6f-74-74]

You were flying in an extremely poor location for GPS to work. Low to the ground, around a structure (which not only obstructs a clear view of the sky for GPS reception, but bounces signals and creates location issues), and below trees. Your screen recording showed an average 5-7 satellites - you should have had at least double that to obtain acceptable accuracy, and even then I would not plan a Litchi mission without building in at least a 15 foot buffer (both vertically and horizontally) for safety.

 

[MadValley]

I found this pretty helpful: GPS.gov: GPS Accuracy. There are a couple of other factors to consider. Does your house have an active WiFi and is the phone/tablet connected to it when you're flying? Your GPS signal seems fairly weak like 6 - 9 satellites and varying. This may be influenced by the limited view of the sky you have flying that low and close to the house. Is there wire reinforcement in the concrete you're flying near. That can be problematic for signal propagation at that height. Which way are the antennas facing? I'd try running the same pattern out on an open field with 10 or 12 satellites available and see if you don't get better performance. Then you can bring the path back to the house and start controlling for some of those other variables and see if your results improve.

 

[56_kruiser]

All very good points, everyone. I appreciate the input.

I didn't even think about the number of GPS hooked up during the "GPS test". I was just focusing on the results. I wish I had recorded the screen during the actual flight.

I will say that the mission started out in open area at 98' and worked it's way down and closer to the house. Right before the video I showed you started it was in the driveway at about 8' altitude and back at the street edge. So if during the flight it lost the sats pretty quick, which of course is quite possible.

Just for grins I think I'll do another test starting out at the edge of the driveway and watch for it to lose sats. The test above and the next one is not flying...just carrying.

Overall this was a good learning experience for me. All points noted.

 

[Wirezfree]

I'm very new to drones... Only 15 flights.
Even though you have plotted the waypoints... Won't the Mavics inbuilt sensors/software try to avoid objects
You are very close to "things" won't that cause it to shift around...???

 

[56_kruiser]

I'm very new to drones... Only 15 flights.
Even though you have plotted the waypoints... Won't the Mavics inbuilt sensors/software try to avoid objects
You are very close to "things" won't that cause it to shift around...???

I've another thread on here where I was testing manually flying across the front of the house, out at the street, with active track focus on the front of the house. A tree it was crossing behind made it lose sight of "the subject"...and it fast forwarded straight into the tree, despite obstacle avoidance. That was with DJI Go 4. So I don't have a lot of confidence in the obstacle avoidance.

In the case of the video that is the topic of this thread, it was traveling sideways so the sensors were not head on and would be of no use.

 

[MadValley]

That's interesting. I've flown mine right at me (I figured I'd be better than the shed from a drone damage control perspective) and it stopped every time. Just an "is it plugged in? kind of question: Did you remove the protective stickers from the front sensors?

 

[56_kruiser]

That's interesting. I've flown mine right at me (I figured I'd be better than the shed from a drone damage control perspective) and it stopped every time. Just an "is it plugged in? kind of question: Did you remove the protective stickers from the front sensors?

Oh, yes. In fact I'm not saying it hasn't/doesn't work. I've had it work several times. I suspect the time it crashed into the tree had more to do with a failure relating to it losing the subject in active track. The tree completely obscured the view to the focus point, and it forwarded and crashed into the tree.

I may try to contact DJI and see what they think.

 

[dvarapala]

1. Be sure you know how to take over a flight going wrong, and remember it. After the fact, I remember that to take over I need to switch on Sport Mode...a mode I've never used yet. I had thought that pressing the pause button on my Mavic controller would pause it, but it did not.

Does anyone know the reason why you can't just pause the waypoint mission and take full control? I have a feeling a lot of n00bs get freaked out the first time they hit the pause button in the Litchi app and find their control sticks do absolutely nothing. Switching into Sport Mode isn't exactly the most intuitive thing, especially when you're in a minor panic because your drone isn't working.