Drones these days come in various shapes and designs, directly corresponding to the purpose they are supposed to fulfill. The one thing they all have in common is: Drones are aircrafts.

As such, their design needs to meet the same main requirements and overcome the same constraints as regular aircrafts, like planes and helicopters (apart from the environmental control system for life-support). Main elements include the basic shape, the size and the weight of the drone, followed by the payload, the envisioned flight time, the available budget, and of course the regulations. All these features are interlinked, meaning they have to be in a sort of balance for the aircraft or drone to be able to take off. This makes prioritisation and compromises compulsory for the final design.

Let us give you a quick overview of the interconnections between different aspects of a drone:

To start with the basics, a small drone can carry a light payload, and similarly, a bigger drone can carry a heavier payload. Your project now might have requirements to use one of the newest broadcasting cameras with a high-quality lens system, sensors, and matching electronics to get the best production quality possible (which results in quite voluminous and heavy camera systems). Furthermore, it might require a high-performance onboard computer to make the drone as autonomous as possible.

As a result, the drone needs to be able to lift a heavy payload, which means a larger drone is needed. Consequently, there are the needs for more power to fly, resulting in the necessity of batteries with higher capacity (hence heavier) to keep the same flight duration, which makes the drone overall heavier and larger. As a result, the drone is probably bulkier and less transportable friendly.

On the other hand, a smaller drone may be agiler and easier to transport, but the cameras, sensors and onboard computers will not be as powerful and qualitative as in the previous case. Hence the need for compromises.

What’s the situation in MultiDrone?
For the MultiDrone project, Alerion will design and prepare a prototype according to the partners’ requirements. The prototype will then be replicated to have a swarm of drones, which can be tested within the MultiDrone project.

In terms of setup, the drone will be divided into five parts:
– the drone platform which is the main hardware structure of the drone,
– the drone core which consists of the main features to make the drone fly,
– the flight payload to add autonomy features,
– the audio-visual payload to have smooth and high-quality images and the batteries.

Overview of the MultiDrone drone’s architecture

This setup will allow the MultiDrone Consortium to cover sports events using a swarm of drones with an audio-visual camera system in broadcasting quality and applying autonomy features.

The specifications of the drone for the project are currently being established following the requirements from the partners. As described in the beginning, some requirements are in conflict with one another and others are hard to meet with today’s technologies.

As for example, from a media production perspective, it is important to have a lightweight drone, in order to be able to quickly move around, kick off the drone and start filming. In the meantime, this also requires a long flight time and a good camera with broadcasting quality.

Both these requests interfere with the requirements from the technical partners. Autonomy features such as intelligent tracking of individuals or sensing and avoiding of obstacles are based on sophisticated algorithms which require high-performance sensors, such as LIDARs and cameras, coupled with powerful onboard computers. These devices are typically heavy and expensive, which results in needing a larger drone.

Octocopter drone up in the air

How do we solve this issue within MultiDrone?
Several options are being investigated to find the right compromise, requiring a constant check and update of the requirements and solutions. A first test of the envisioned algorithms that will be installed on the drone, for example, required an increase of the computing power.

In parallel, the team is prioritising the different components (inflicting the payload) by deciding which ones are mandatory for the project, and which ones are not crucial. This meant, for example, using only a LIDAR, but dropping the idea of having an additional stereo-camera for visual analysis.

Other ways of finding a compromise are taking a look ahead, at technologies not yet available, but with an envisioned market exposure in the next months or splitting responsibilities between drones in the swarm. This means that not all drones might have the same setup, but each drone will fulfill a crucial part in the swarm for the rest, like mapping, while others are dedicated to filming.

The project is on the right track and the consortium confident to find a solution to bring together the first prototypes in 2018.