Yet another Virtual Litchi Mission (Chrome/Firefox Extension)

[joeruby]

​

Hi Bazuchan,

I think I found a bug in GCE. I also think it was there in VLM too. I have been trying to find what I did wrong, but can't. Here is the problem I see:

When flying between two waypoints with curved turns, the camera (in the simulation) turns well before the curved section of flight has arrived. However when I actually fly the mission, the camera looks TNW (Torward Next Waypoint) as it should, and waits until the AC has arrived at the curved section.
You can simulate this with a three waypoint mission that forms a 90 degree curved turn. I placed three waypoints, each a mile apart, to form an "L" shaped corner. When I simulate it in GCE, the camera starts to turn almost immediately at the start of the mission.

What have I done wrong?

Thank you for your help.
Joe
KC7GHT

 

[Bill Ludwig]

Do you have “Interpolate” turned on at your waypoints? If so, that will cause the drone heading to gradually transition from one waypoint to the other.

 

[joeruby]

Do you have “Interpolate” turned on at your waypoints? If so, that will cause the drone heading to gradually transition from one waypoint to the other.

Yes I had "interpolate" turned on. I turned it off and saved it, and I still have the Yaw axis turning way too soon in the flight in the simulation. I thought the "interpolation" was for the pitch axis of the camera.

The problem I see (in the simulation) is the Yaw axis changing way before the AC arrives at the location where the curve starts - leaving the straight line section behind.

 

[Bill Ludwig]

Yes I had "interpolate" turned on. I turned it off and saved it, and I still have the Yaw axis turning way too soon in the flight in the simulation. I thought the "interpolation" was for the pitch axis of the camera.

The problem I see (in the simulation) is the Yaw axis changing way before the AC arrives at the location where the curve starts - leaving the straight line section behind.

If you make your mission public and post it here, we can take a look at it. I’ll even go fly it right now and let you know what I experience with the M2P.

 

[joeruby]

I have unchecked the mission in my mission hub, but I don't know how to post a link here. Sorry for being so dumb. Thanks

 

[Bill Ludwig]

No problem, just go back to the screen where you unchecked the box and look to the right at the column named 'Link". Just RIGHT click that and 'copy link address' and past that in your next post.

 

[joeruby]

Thanks. This is with the camera looking down (-90). What I see when I sumulate this, is the yaw turning almost immediately, way before the AC gets to the curved portion of the first turn. This is happening in all of my simulation - I think. Hope this link works for you.

Mission Hub - Litchi

Litchi's Mission Hub lets you plan waypoint flights and later execute them using the mobile Litchi app.

flylitchi.com

Thanks a lot for the help.
Joe

 

[bazuchan]

​ ​

I think I found a bug in GCE. I also think it was there in VLM too. I have been trying to find what I did wrong, but can't. Here is the problem I see:

When flying between two waypoints with curved turns, the camera (in the simulation) turns well before the curved section of flight has arrived. However when I actually fly the mission, the camera looks TNW (Torward Next Waypoint) as it should, and waits until the AC has arrived at the curved section.

Nothing wrong on your side, from your description I think I know where the bug is. I'll look into it next week.

 

[joeruby]

Hey, Thanks a lot!!!! I have been driving myself crazy trying to understand it.

 

[Bill Ludwig]

Hmmm...I ran your mission through GE Chrome, and it looks just like I would expect it to run. It starts out due North, and gradually changes to 270 degrees. Is that not what you are seeing?"

 

[Facherty]

My way of running this the way you seem to want to do would be to add a Waypoint between 1 and 2, and position it at the beginning of the first curve (blue line). This waypoint would continue the original heading. Repeat for each of the corners. (See Mission Hub - Litchi)

 

[joeruby]

Yes, I have done that in the past, and I can always do that. I seems that I don't understand Toward Next Waypoint (TNW) and I thought that the AC Yaw should follow the shape of the blue path with the AC Yaw angle always tangent to the blue (calculated, rounded) path. Maybe that is my misunderstanding.

However, I have a DJI P3P and the flight video indicates that the actual AC Yaw angle is tangent to the blue flight path. Let me know what you think. Thanks again for the help.

 

[joeruby]

Hmmm...I ran your mission through GE Chrome, and it looks just like I would expect it to run. It starts out due North, and gradually changes to 270 degrees. Is that not what you are seeing?"

Yes that is what I am seeing. The simulation AC Yaw starts turning immediately after start, instead of waiting for the AC to get to the curved portion which is about 0.8 miles away. My actual flight video shows that the AC Yaw angle stays constant until about the 0.8 mile point then it starts changing to the new waypoint direction. Thus the simulation and the actual flight don't match. I hope I am making sense.

 

[Chas2379]

Joe, I'm glad you pointed that out. I just checked some actual footage against the virtual mission. The virtual mission the ac starts turning gradually from one waypoint to the next waypoint. My video follows the curved lines. I'm going to fly a complicated mission today. I'll post the results.

 

[Chas2379]

Do you have “Interpolate” turned on at your waypoints? If so, that will cause the drone heading to gradually transition from one waypoint to the other.

Interpolate affects gimbal pitch only and not heading. Be aware VLM show the heading gradually changing from a waypoint to the next waypoint with a different heading. The aircraft will actually start to change heading where the curve begins on the next waypoint.

 

[bazuchan]

I have published new version 0.2.0 (you need to update extension in chrome). Changes are following:

• Added options page to specify drone model for correct FOV calculation.
• Fixed Waypoint Views in Diagnostics folder.
• Fixed TNW mode (no need to specify it in settings), @joeruby please test.

 

[marklyn]

Thanks so much. Any chance you could make this for a firefox add-on? I'm falling out of love with Chrome and jumping over to the FF camp but completely understand if you stick with one platform only.

 

[bazuchan]

Thanks so much. Any chance you could make this for a firefox add-on? I'm falling out of love with Chrome and jumping over to the FF camp but completely understand if you stick with one platform only.

It was planned originally, but I was afraid of not being able to implement same set of features for both plugins, so decided to postpone it. Maybe it is time now, will look into it.

 

[marklyn]

I appreciate you considering it anyway. FF is turning out to be a better browser for me so I for one am very happy you'd even consider adding another platform. Either way, I love the fact that you can do everything from the browser instead of a separate app. Great work!

 

[Facherty]

Many thanks for this. It has improved my flying experience and reduced my blood pressure.

If anyone wants to know how to change the aircraft Field of View:
1. open a new Chrome tab, and enter chrome://extensions/
2. find the Chrome Litchi Virtual Mission 0.2.0
3. click Details
4. click Extension Options
5. click on Drone model

6. choose from the options