This drone flight planner turns the footprint of one photo into a mission: the spacing between flight lines and between shots for a chosen overlap, the number of lines and photos for a site of a given size, the distance flown and the time it takes. Set the camera and the height once, then read off what the job actually costs in battery. Free, in your browser, no login.
Lines run along the length axis, so the planner wants the site measured along and across the direction you intend to fly. Swap the two numbers if you would rather fly the other way. For a non-rectangular site, take the bounding rectangle of the block — the planner is a rectangle model, which over-estimates slightly on an irregular shape.
Check the camera — sensor, focal length and image pixels must all be positive, the site and the height must be greater than zero, and each overlap must be under 95%.
| Item | Value |
|---|
The planner counts whole lines and whole shots — the spacing always rounds up, because a partial line still has to be flown and a partial shot still has to be taken.
The footprint of one photo comes from the camera and the height — the same proportion the
GSD calculator uses:
footprint_width = H × S_w ÷ f across and footprint_height = H × S_h ÷ f along.
The overlap then says how much of that footprint each successive photo or line may share:
line spacing = footprint_width × (1 − side overlap)
shot spacing = footprint_height × (1 − forward overlap)
Note which footprint each spacing uses. The lines step sideways across the width of a photo, so the side overlap governs the spacing between lines; the shots step along the height of a photo, so the forward overlap governs the spacing between shots. Mixing these two up is the single most common planning error, and it produces a plan that is internally consistent but wrong.
Divide the site by the spacing and round up, because the aircraft cannot fly a fraction of a line:
lines = ceil(site width ÷ line spacing)
shots per line = ceil(site length ÷ shot spacing)
photos = lines × shots per line
The distance flown is the length of every line plus the width of every turn between them:
distance = lines × site length + line spacing × (lines − 1). Dividing by the ground
speed gives the time.
The rectangle model leaves the outermost half-footprint of the block under-covered, because the first line is on the edge and its photos extend only half a footprint outside — and not at all inside the second row. Real missions therefore fly one extra line outside each edge and start the first shot before the boundary. Add one line to the count and about half a footprint to the length for a working number; the overlap you actually get at the edges is what decides whether the orthomosaic can be cropped cleanly at the boundary.
Overlap is not a quality dial you turn up for safety — it costs photos and flight time linearly, and beyond a point it stops buying accuracy. These are the ranges that hold up in practice.
| Terrain | Forward | Side | Why |
|---|---|---|---|
| Flat, hard surface, good texture | 70% | 60% | The matching has plenty of features to lock onto; more frames is wasted battery |
| Ordinary ground, mixed cover | 75% | 65% | A safe default for a survey deliverable |
| Grass, crops, bare sand | 80% | 70% | Features move with the wind and match poorly between distant frames |
| Steep slope or tall buildings | 80% | 75% | Parallax between the top and the base of a feature is large; the spare frames cover the occlusion |
| Corridor, road or pipeline | 80% | — | Only one line each way, so all the redundancy has to come from the forward overlap |
Above about 85% the returns flatten and the photo count climbs steeply: going from 80/70 to 90/80 roughly doubles the mission. On a vegetated site it is usually cheaper to fly lower with 80/70 than higher with 90/80, because the GSD improves as well.
Take the defaults: a Phantom 4 RTK at 120 m, a 500 × 400 m block, 70% forward and 70% side overlap, 8 m/s ground speed. At 120 m one photo covers 180 × 120 m (2.16 ha), which the GSD calculator confirms.
Spacing: across, 180 × (1 − 0.70) = 54 m between lines; along,
120 × (1 − 0.70) = 36 m between shots.
Counts: lines = ceil(500 ÷ 54) = 10, shots per line = ceil(400 ÷ 36) = 12,
so 10 × 12 = 120 photos.
Distance: 10 × 400 + 54 × 9 = 4 000 + 486 = 4 486 m. At 8 m/s that is
561 seconds, or 9.3 minutes of straight-line flying. Add a quarter for the turns and the
climb and the mission is about 12 minutes — comfortable on one battery.
Now halve the GSD to 1.6 cm/px, which needs 60 m. The footprint halves to 90 × 60 m, so the spacing
halves to 27 m and 18 m and the counts double in each direction:
ceil(500 ÷ 27) = 19 lines of ceil(400 ÷ 18) = 23 shots, which is
437 photos and roughly 3.6 times the flight time. That is the real cost of doubling the
resolution, and it is why the deliverable specification matters before the flight, not after.
At 70 percent forward and 70 percent side overlap with a Phantom 4 RTK at 120 m, one photo covers 2.16 ha and the spacing is 54 m between lines and 36 m between shots, so a hectare needs roughly 5 photos on a pure area basis. Real sites need more because the spacing is rounded up to whole lines and whole shots, and the edges of the block need a buffer line outside the boundary.
70 to 80 percent forward and 60 to 70 percent side for ordinary flat terrain. Go to 80/70 on vegetated ground, steep slopes or where the surface has few texture features, because parallax moves the apparent position of a point between shots and the matching needs spare frames to survive it. Above about 85 percent you gain little and the photo count climbs steeply.
The straight-line distance is the sum of the line lengths plus the spacing between adjacent lines, which already accounts for the turn width. A real flight adds time for decelerating into each turn and accelerating out of it, plus the climb to altitude and the return to the home point. The rule of thumb is to add 15 to 25 percent to the computed flight time for a block of a few dozen lines.
Fewer, longer lines are more efficient than many short ones, because each turn costs time and each line loses a little at the ends. Running the lines along the long axis of the site minimises the number of turns. Some planners also rotate the block slightly so that the lines follow the dominant grain of the terrain or a road, which reduces the number of awkward partial lines.
The GSD fixes the height, the height fixes the footprint of one photo, and the footprint fixes the line spacing. Halving the GSD halves the height and therefore does not just halve the spacing but quarters it in area terms, so the photo count goes up about four times. Doubling the resolution is roughly a four times longer flight, not twice.
Take the practical endurance from the aircraft's manual, not the marketing number, and keep 30 percent in reserve for wind and the return leg. A Phantom 4 RTK flies about 25 minutes of usable hover time; a Mavic 3 Enterprise about 35. If the flight time exceeds two thirds of that, plan a battery swap and a second mission rather than pushing one aircraft to the limit.
Yes, free, no login, no account, no usage limit. Every calculation runs locally in your browser, so no site boundary or flight plan is uploaded anywhere.
Calculations run locally in your browser. Results are for reference — confirm against the governing specification and the aircraft's flight manual before you fly. About · Privacy & cookies · Terms of use · Developers