The Missions

Here you can find all about the missions of our small satellite.

The primary mission

This mission is required by the code of competition and can be broken down into the two following actions:

• computing the current altitude and rate of fall
• generating a temperature profile

The data for the primary mission is provided by our BMP280, a high precision digital pressure sensor.

The (secondary) main mission

The aim of this mission, being conceived by our team, is:

• an optical/visual determination of the position from our CanSat

This is very useful caused by these two advantages. The first advantage is that we could perform operations independently of the presence of a GPS-like system, Because we could use the information from the position determination for the operation of small control nozzles.
In addition, a lander could calculate its own position during the landing and it could spark it to earth after landing. So the search for its actual landplacement is not necessary.
The second and also far greater advahntage is that we can perform the evaluation of the images on board. In this way, we relieve our interconnection by analyzing the pictures there, which in reality can also lead the way for remote sensing in the universe as an idea.
The only prerequisite for our procidure is a digital recording of the surface of the respective planet or asteriod.

In order to reach this aim, our CanSat should take high-angle shots of the earth surface below during its drop. After touchdown the shots get analyzed by a an application on the Raspberry Pi 3.
We have evaluated two different methods to achieve this. The first method tries to determine the position by feature detection. Therefore we compare the meaningful image points from our taken picuture with a reference image, from which can assign a position to each point.
Finally, we can calculate the position from the same points. The second method tries to determine the position by using a neural network. We use it to get more information from the images. Each point is assigned to a category and thus image areas are for example buildings or a runway.
The prerequisite is that, we have to train the neural network with a large image database.

After the touchdown, we will be able to compare the coordinates our applications returned with the actual coordinates that our GPS module provided.