This application will convert a Google Earth kml file to a Litchi csv file.
In Google Earth Pro:
Use Google Earth Pro to navigate to the mission area.
From the ribbon select the "Ruler" tool.
In the Ruler tool, select the "Line" tab (normally selected by default).
Click once on the map at the desired, ending point of interest (POI). NOTE: This position will be the end of your dronception mission where the drone is high and the camera is pointing straight down at gimbal pitch angle of -90 degrees.
Move to (and click) where the first waypoint is to be located. NOTE: This position will be the start of your mission where the first photo will be taken relatively close to the ground.
In the Ruler tool click on "Save".
The New Path window will open. Click OK.
In the "Places" window pane (on the left), right-click on "Line Measure" and "Save Place As" a kml file.
In page one of Droneception Mission Maker for Litchi:
Select the desired units.
Select the starting height. The starting height should be low, maybe 20 feet (7 meters).
Select the ending height. The ending height should be high enough to capture some sort of point of interest which will end up at the top of the droneception image, maybe 200 feet (70 meters).
Select the number of photos to capture. Some capture as few as three images. There is no penalty in capturing more, and the more photos you have, the more you will have to work with. Plus, you don't have to use them all.
Use the file chooser to upload your kml file.
Select "Generate Droneception Mission Maker for Litchi".
In page two of Droneception Mission Maker for Litchi:
Review (and understand) the parameters and calculations shown.
Download the csv file to your computer.
In the Litchi Mission Hub or Virtual Litchi Mission (VLM is recommended to preview the flight path):
Use the Missions menu to import the csv file created by this web application.
Adjust Litchi mission parameters as desired (although no adjustments should be necessary).
IMPORTANT: Set the "Path Mode" to "Straight Lines". Litchi will not perform waypoint actions unless the "Path Mode" is set to "Straight Lines".
Set the cruising speed to something comfortable like 8 mph (13 km/h).
Save the mission.
Preview the flight in Google Earth Pro using VLM (optional)
It is recommended to preview and verify the flight using Virtual Litchi Mission.
Fly the mission
Fly to a position somewhere between the first and last waypoint while pointing the camera to the POI.
Set the camera to manual mode and adjust the exposure as desired.
Load the mission and fly.
The inputs to this application are:
A kml file containing two locations from Google Earth Pro
A point of interest (POI) which ends up being the same as the final waypoint and is where the final photo is captured
A starting location where the first photo is captured
A starting height
An ending height
The number of photos to be captured
With this information, one item is assumed.
Each photo will be captured at an equal horizontal distance from each other. The distance between the two points supplied to Google Earth Pro will be divided by the number of photos requested (minus one) to produce this horizontal distance.
From this information several quantities are calculated.
When progressing from one waypoint to the next, the waypoint height is computed by multiplying the remaining height differential (between this waypoint and the final waypoint) by a fraction. The fraction is determined by a formula developed empirically.
Once the location and height of each waypoint has been calculated, the gimbal pitch angle can be computed for each waypoint so that the camera points at the POI.
Finally, all waypoint and POI information is written to a .csv file along with camera actions to pause and take a photo at each waypoint.
A "droneception" image is a "mind-bending" photo montage where it looks like
the ground is bending upward as you look into the distance. The inspiration
of this type of image is from the movie Inception (see photo).
Creating a "droneception" image requires three primary steps, each presenting
a different type of challenge:
Identify a subject that will look good.
Take a series of photos with a drone at various heights and distances.
Use photo editing software to blend the individual photographs into a single, warped photo montage.
Identify a subject that will look good
From examples I have seen on the web (there are many), a good subject for a
"droneception" photo montage includes:
A rarely-travelled road going off into the distance
A point of interest located on or around the road in the distance
Not too many objects on either side of the road (which will make the final blending process more difficult)
Take a series of photos with a drone
This web application provides an easy, repeatable method for generating a
series of photos. The application not only determines the height and position
for each image but will also define camera actions in the file that will be
imported into Litchi. The result is a Litchi mission that automatically
captures all photos as the mission is flown. Be sure to set your camera to
manual mode. Auto mode will most likely cause the photos to be taken with
Blend the series of photos into one
This step is done using photo editing software such as Photoshop. The general
idea is to import each image into a different layer and then use warping
functions and masks to overlay and blend the layers into a seamless montage.
People with some Photoshop (or equivalent) experience shouldn't find this too
difficult. There are tutorials and instruction videos describing this process
on the web. It is easy to find them by searching.
New Path Computation Option: There is now a choice of two different methods for computing the waypoint heights and gimbal pitch angles.
Empirical: The waypoint distances from the POI are evenly spaced. The height of each waypoint is computed based on a fraction of the remaining height. The fraction is empirically calculated based on the number of waypoints. The gimbal pitch angles are then computed at each waypoint. This is the original algorithm.
Logarithmic: The waypoint distances from the POI are evenly spaced. The change in gimbal pitch angles between waypoints is constant and computed based on the number of waypoints. The height of each waypoint is computing using log(n) base(m) where n is the waypoint number and m is the number of waypoints. This is a new algorithm.
Improved Waypoint/Camera Actions: Despite having a
heading and gimbal pitch angle defined for each waypoint, the drone does not
strictly adhere to them at the first and last waypoints of a mission. To
remedy this, additional waypoint/camera actions were added to each waypoint.
The following actions are now performed at each waypoint:
Rotate aircraft to the correct heading
Tilt camera to the correct gimbal pitch angle
Stay for one second
Stay for one second
This results in more accurate camera positioning at each waypoint, especially the first and last waypoints of the mission.