How Realistic is VLM?

[Steve_Amerson]

First off I love Virtual Litchi Mission and find the planning and design process easy and fun to do. Sitting back and watching the flight is like being on this giant scenic monorail and makes it easy to spot areas where speed, altitude or camera angle need to be modified. My Question is just how reliable is the overall Litchi system in the real world when dealing with position and altitude? Power lines, trees and other obstacles need to be scouted first and the GPS signal strength is critical but after that will the drone be where it is supposed to be or is there a Variability of a few feet in all directions? Are there other things that can alter a drones preplanned course that we should be aware of? Thanks!

[wesbarris]

@Steve_Amerson

My Question is just how reliable is the overall Litchi system in the real world when dealing with position and altitude?

VLM is very reliable and realistic with the following caveats:

1. Neither VLM nor Litchi are aware of what your take-off (home) location will be. Since the drone’s altitude is measured from its take-off location, VLM and Litchi must assume that your take-off location is at the same altitude as waypoint 1.
2. Consumer GPS is only accurate up to plus or minus one meter. This is neither a fault of VLM nor Litchi. It is just a limitation of the technology.
3. Trees grow. Since maps can be years old, be careful when planning missions near them. They may be much higher and wider that what is shown in Google Earth Pro.
4. Each model of drone has a different Field of View (FOV). Unless the FOV of Google Earth Pro is set up to match that of your drone, your view will look different.

Keep in mind that VLM and Google Earth Pro are separate products from Litchi. Each has their own limitations. After planning a mission at your desktop, your first flight of that mission needs to be watched carefully as it is flown for the first time. Each time you make subsequent flights of that same mission, there will be slight variations in the position and height of your drone due to GPS accuracy and changes in take-off location).

[Steve_Amerson]

Thank you wesbarris for the help on this issue! This detailed answer is very helpful to me and hopefully other pilots as well.

[paxxa]

@Steve_Amerson

My Question is just how reliable is the overall Litchi system in the real world when dealing with position and altitude? Power lines, trees and other obstacles need to be scouted first and the GPS signal strength is critical but after that will the drone be where it is supposed to be or is there a Variability of a few feet in all directions?

You can count on an accuracy of about 10 meters.

GPS position failure

Here the same - with a Mini 2. I tried the waypoint mode some times, and the error is around 10m from the points of the map - this is crucial to solve, as drones can be lost to nearby obstacles, and our accuracys are bad for any type of job and map. If doing that with other app like Rainbow, did not find the same error on basically the same map. With so many here confirming the issue, there is no doubt it’s concrete and sound - not good at all. So, Litchi must address this problem ASAP - almost …

On this forum, a question about gps accuracy was very often asked, for some it ended with a broken drone

[Steve_Amerson]

Wow, 5-10 meters of unpredictability in all directions is dangerous unless you are well above all obstacles! I had been planning on doing some tight missions up and down rivers and streets as well as through building corridors but now I am not so sure what is safe. Insurance companies might even refuse coverage for certain missions. Thanks paxxa!!

[paxxa]

You can fly the drone during the mission, just be prepared to use the sticks

[Steve_Amerson]

wesbarris, have you had any problems with Litchi positioning recently? I was referred to a Forum post where a number of Litchi users were having their positions off by between 5-10 meters which is way off the norm of 1m as you mentioned. Thanks.

[paxxa]

Orbit mode centerpoint

The Litchi app can do an orbit with a minimum 15’ (3M) radius. The DJI Fly app can’t do a circle that small. BUT, the centerpoint of the orbit when marked on the map (using a table iPad) isn’t exactly the actual center. For example, an orbit around a flagpole or church spire may not have the object in the center (almost hit what I thought was the centerpoint structure). Am I not using the marker on the map properly? Is this just an approximate position based on GPS accuracy limits? Can the map…

Google Maps or Apple Maps

I’ve been flying a mission that requires fairly close GPS accuracy to get through a gate at the start. I need to observe the DJI Mini 2 closely during that section. Sometimes it flys through the middle and other times it wants to fly through a bush or tree! Once past that section it has lots of space to run and I can relax for its return. Consistency of its GPS is my question? I launch from the same spot, wait for a clear day, and no wind. How can I get better accuracy for its PERCEPTION of cur…

[wesbarris]

@paxxa

You can count on an accuracy of about 10 meters.

Nonsense. Anyone who has planned and flown missions will know that it is much more accurate than ± 10 meters (30 feet). You generally will see better than ± 2 meters. Usually ± 1 meter. If you are only seeing an accuracy of ± 10 meters something is wrong and you need to do some troubleshooting.

[paxxa]

These forum threads were not created by me.

[Steve_Amerson]

Thanks wesbarris and paxxa, you are both helping to hash out a very important issue and I appreciate all the effort and time taken. I can see that this is going to take time and practice for me to be a safe and informed pilot. Unless there are differing viewpoints, there can be no debate.

[wesbarris]

@paxxa

These forum threads were not created by me.

No, but the 10 meter accuracy statement was.

Anyone with any experience with flying missions will know that the accuracy is better than 10 meters. If the accuracy was really that poor, we would all be crashing our drones. You should know better than that.

[paxxa]

@wesbarris

No, but the 10 meter accuracy statement was

==>

GPS position failure

Here the same - with a Mini 2. I tried the waypoint mode some times, and the error is around 10m from the points of the map - this is crucial to solve, as drones can be lost to nearby obstacles, and our accuracys are bad for any type of job and map. If doing that with other app like Rainbow, did not find the same error on basically the same map. With so many here confirming the issue, there is no doubt it’s concrete and sound - not good at all. So, Litchi must address this problem ASAP - almost …

[paxxa]

I can only find a few more threads on the forum where this problem was mentioned.

I remember there was a case when mini crashed into the wall of the house, it was quite recently.

[wesbarris]

@paxxa

I can only find a few more threads on the forum where this problem was mentioned.

So you found one post from about 6 months ago from someone with an unknown GPS problem and you automatically assume that is the norm? Doesn’t your own experience tell you anything different? Wouldn’t you say that is a bit irresponsible of yourself?

[paxxa]

@wesbarris

Doesn’t your own experience tell you anything different?

I see a gps error of about ±5 meters.

[paxxa]

In any case, this is not enough to fly into the gate, the drone will definitely crash

[paxxa]

Seems like a similar thread 2 months ago

Accuracy of waypoints in windy conditions

On more windy days I noticed that my Mini 2 is struggling to get through the waypoints. It’s not windy enough that it’d be unsafe and I can easily fly “close enough” manually and take the photos I need. However the automatic hovering system struggles with gusts of wind, overshooting the waypoints then overcorrecting for that and eventually gives up. Is there a way of “reducing the precision” for waypoints so that the mission can still be flown automatically? For my use case, I don’t need much …

[wesbarris]

@paxxa

Seems like a similar thread 2 months ago

You are confusing wind or possibly signal strength issues with GPS accuracy.

[Steve_Amerson]

Thanks to you both paxxa and wesbarris, I feel stronger going into this Litchi drone world. While there is plenty to read, some of the learning is finding out what is real and which is rumor or odd glitches caused by who knows what. While guarding against rumor or false statements is vital, I now feel more prepared since I am more confident that the majority of the time the drone will behave properly but also to always expect the unexpected. I will keep you both informed as I venture out and gain more confidence as time goes on. This is a great community and I am proud to be part of it, so keep up the good work!!