Cleaning concept for solar panels

Solving the problem of dust adhesion on the surface of solar panels in desert or sandy environment! Mueataz Hamed came up with an innovative idea for his Master student project at the Technical University of Braunschweig, Mueataz developed a modern cleaning concept using Unmanned Air Vehicle (UAV). Thanks to the financial support of Sheikh Dr. Abdullah Bugshan on behalf Europe Hadharem Forum and with a strong cooperation of the Institute of Flight Guidance and Copting GmbH, which made building the prototype possible.

He built a prototype based on the Frame Tarot 960 Hexacopter which is bigger than usual copters. Hence it has lots of power using a battery system with 22000 mAh. It has a maximum load of 9 kg and a hover flight time with estimated 24 minutes. The hexacopter is build with two cameras, thermal camera to analyze dust quantity on the surface of solar panels and normal camera. Moreover the 340 KV motors down-wash is used to clean solar panels. The system is not only fully automated but also environmental friendly since there is not water included in the cleaning process. A great innovation for an ecologically worthwhile purpose as well!

Vertical Take-Off and Landing Changyucopter

Last year in September we started the project "Changyucopter". The head of the VTOL project is Yannic Beyer, an aerospace engineering student. His master thesis "Flight Control and Simulation of a Tandem Tilt Wing RPAS" is about the design of an appropriate flight control of the VTOL.
Meanwhile, the aerial system was built by AKAMAV and has succesfully passed the hardware test in February ‘17. The Changyucopter measures 1.40 x 1.40 m and has a maximum take-off weight of 5 kg including up to 1 kg payload. This first prototype reaches 45 min of flight time flying at 80 kmph. After some challenges the VTOL is now ready to take-off vertically! We wait eagerly for further flight tests and results!

Parcel Copter

According to the media, the delivery of packages via copter will start tomorrow. Sadly there are still some problems to solve. Firstly, the environment perception of today's drones is not good enough to avoid obstacles reliable enough. Also the accuracy of conventional GPS receiver is too poor. Here comes the creativity, ingenuity and modesty of Brunswick students into play: Our delivery Copter won the IMAV15 and is under further development this year:



LiDAR SLAM Visual Landing PPP GPS
Objects are detected using Laser Scanners and are entered into an environment map in real time. The created map is the foundation for following obsticle avoidance or adaptive path planning. Using on-board cameras visual markers (i.e. in the form of QR codes or infrared beacons) are detected. The detected object can then be used to guide the vehicle to land on it. Since the accuracy of common GPS receivers is in the range of several meters, the AKAMAV is corrently working on the implementation of PPP (Precise Point Positioning) GPS with centimeter accuracy.
(Student Thesis possible.) (Student Thesis possible.) (Student Thesis possible.)

Pocket-Size Quadcopter

Currently unmanned aerial systems are commonly implemented as quadcopter. In addition to the relatively simple mechanical setup, these systems benefit from a well tuned control loop. Acquiring the skills needed to be a PID magican can be costly if you destroy multible quadcopters during the process. To learn these skills in an easy, quick and painless manner the AKAMAV build 5 palm-sized micro training quadcopter. The control of the brushed motors is implemented via a STM32F1 Microprocessor, a MEMS accelerometer and a gyroscope. The processor has more than enough power reserves to experiment with more complex control algorithms.


A FPV (control via live video feed) version is currently under construction.

Autonomous Aerial Mapping

Currently the most common task for unmanned aircraft systems is the creation of low-cost aerial photographs. In order to assign or compare the data created correctly, it is meaningfull to perform a photogrammetric processing. The objectives are to create a georeferencedphoto maps, as well as the calculation of 3D models. While both is commonly done in post-processing, the AKAMAV is also working on procedures that run in real time on the aerial system.


A goal constructed in many contests is the autonomous localization of victims after a natural disaster. These images are therefore also examined by computer vision to identify specific characteristics  for example to identify people.

Here description of our mapping mission during IMAV15.

Carolo 15 (Airship)

The current Airship was designed to carry significantly higher payloads. Additionally the attitude control via control flaps were tested. The 2.20m long skin was made out of rescue film. The required flap deflection is calculated and commanded by an onboard Arduino Pro Mini (2grams). Both 5 inch propeller are commanded directy by via RC. 

During the project, the vehicle will be equiped with an autopilot including a GPS module for autonomous flight.

Outlook 2017

And again one year has passed and the winner of the IMAV is: Team AKAMAV! Okay well, this time we were only able to score a lot of points in the outdoor challenge. However the competitors in Beijing were really strong, so we are extra proud of this victory. In quite a few aspects we realised, that we remain behind our Goals. For 2017 that means: Sit together, discuss and learn from it!

The hot topics for next year will be precise navigation with RTK-GPS, a more advanced live mapping, tracking based Solutions of all kind and our own VTOL project. Does not only sound like it's going to be fun, it will be!