Any field photogrammetric campaign needs precise flight planning to achieve the desirable dataset required to later process the captured imagery, and GNSS data to obtain the required products like Orthomosaics, Digital Surface Models, Digital Terrain models, vectorial maps, point clouds, 3D model meshed surfaces, …

Failing to properly design the flight planning phase may lead to severe survey issues that may end up in insufficient or excessive imagery overlap, which in extreme cases may force you to repeat the field campaign.

What is flight planning?

When you fly your drone, you probably have a mission in mind or you want to meet a specific data goal. Having the right flight plan will allow you to not only achieve the mission goal but will also help you to save time, resources and avoid damaging your drone or threatening the safety of others.

Flight planning consists of determining the flight schedule, height, altitude, pattern, videos, or images specification, as well as any weather-related condition (irradiance limitations, light, temperature), to achieve the data goals of your job.

In the specific case of using your drone to obtain a high-quality end product, it is necessary to get first good-quality images and to get those images you should prevent a number of variables. Defining a good flight plan should help you to create the right conditions to manage them.

A drone flight should be consistent in every possible aspect, from the camera angle, passing through the elevation above the target, to airspeed.

Another excellent way to save time in the field and get accurate images for your photogrammetry projects is using JASON PPK cloud service, that allows you to reduce or eliminate the use of Ground Control Points (GPCs) in your operations, which translates into a reduction in fieldwork that leads to substantial savings in time and money. You can enjoy a 14-day of JASON unlimited service by clicking here!

Flight planning considerations

Let’s go over some considerations that you may want to include in your photogrammetric flight plan:

Speed: the speed is dependent primarily on two factors, flight altitude and light availability. The higher you fly, the faster the drone can go and also the more light there is available at the camera sensor, the faster you can fly. This means that on a sunny day you may be able to fly faster than on a cloudy/dark day. It should not be forgotten that the light at the camera sensor is also dependent on the camera settings.

Heading: the drone can move in one direction and face another, for typical photogrammetric flights there are two common setups, the drone flies with the frontal part pointing towards the direction of the movement or the drone points always towards a fixed direction, for instance, this can be used to obtain all the pictures oriented with the top part pointing to the North.

Camera settings: the faster you set the shutter speed, the less light that will reach the sensor, hence while there is a higher chance to obtain a “frozen” image without ground smear the higher is the likelihood to underexpose (dark) the image. The wider the camera shutter aperture, the more light will reach the camera sensor, but also the shallower will be the depth of field, therefore the higher the chance to obtain a blurry image, especially if the camera is not precisely focused. Note that there are drone cameras that do not allow adjusting all the parameters mentioned.

Camera shutter type: ideally we want the conversion of raw pixel values scanning of the whole camera pixel matrix to happen all at the same instant however, these kinds of camera models are expensive and are typically found in professional systems, in prosumer drones the cameras are typically of the rolling shutter type which means that pixels are scanned line by line, this means that the faster the drone is flying, the greater the image deformation due to rolling shutter pixel scanning method. Most photogrammetric softwares are able to compensate for rolling shutter deformation.

Lost link behavior: If the telemetry link is lost, the drone should continue flying and taking pictures until the flight plan is completed as otherwise, you may have picture gaps where the ground is poorly covered by the pictures, due to gaps during the flight.

Low battery behavior: If the drone reaches a battery level that is not sufficient to ensure a safe return to the take-off location, the flight plan should immediately be aborted and the drone should return to the take-off point at an altitude no greater than 120 meters, but higher enough to clear all the obstacles in the return.

Gimbal pitch: The gimbal is typically set to an orientation of -90° which is the same to say that is pointing at the nadir. If your intention is to create a 3D model of the object rather than just create orthomosaic / DSM then it may be wise to increase the pitch angle to something closer to the horizontal like -75°, so that in this way the walls of the object are better captured.

Flight pattern: the typical flight pattern for photogrammetric projects it’s the crosshatch, where the drone draws a series of parallel flight paths while taking pictures at regular intervals. If you want to create a 3D model rather than just flat products, it may be beneficial to use a grid pattern as in this way you ensure that all the sides of the object of interest are properly captured. If the area you want to map is predominantly longitudinal, like roads, railways, rivers, … then the advice is to use a linear flight pattern because in this way you will only gather the imagery required for the object of interest.

Terrain following: If there are significant terrain variations in your photogrammetric region of interest, the advice is to use a flight planner that allows you to create a mission that follows the terrain by using the area Digital Surface Model (SRTM for instance), so that in this way the distance between the drone camera and the ground is somewhat constant which implies that the final product scale will be constant too.

Flight altitude: The higher you fly, the lower the Ground Sampling Distance (resolution of the final products) and the fewer pictures you will need to cover your region of interest. Conversely, if you fly lower because you will have to take more pictures to cover your ROI the more computer-intensive the project will be.

Image overlap (as per Pix4D support web page): the general case suggested minimum overlap is 75% frontal and 60% side. 80% frontal and side overlap for agriculture fields and 85% frontal and side overlap for forests and dense vegetation. 85% frontal overlap for single-track corridor mapping.
Use 60% side overlap if the corridor is acquired using two flight lines.

If you want to improve your drone operation projects and save time in the field, additionally to creating an ad hoc flight planning that helps you to achieve your data goals, don’t forget as well to use JASON PPK for accurate and affordable geolocation results. This short video shows you how to use it.
Start enjoying a 14-day free trial of unlimited service by clicking here!