The project was divided into three stages. In the first stage, the focus was on precise UAV positioning in space using various techniques, particularly VSLAM techniques based on signals from stereoscopic cameras with a built-in IMU controller and RTK GPS. As part of this stage, a UAV platform with a multi-processor onboard computer was also designed and built. This allows for the execution of the VSLAM task in real-time.

In the second stage, the focus was on flight planning and monitoring, with particular emphasis on autonomous flight (without direct involvement) and algorithms for collision detection, prediction, and avoidance. One of the outcomes of this stage is a platform for emulating UAV operations – NeuroMatrix.

In the third stage, an extension of the hardware-software platform was developed, with an emphasis on its modularity. As a result, onboard electronics for the UAV drone were created, in which the flight control subsystem was separated from the data acquisition and analysis subsystem. Within the flight control subsystem, a cloud-based solution (Kubernetes) was implemented, which communicates with multiple UAV devices and allows for monitoring their behavior.

The project was implemented in cooperation with the Signal Processing Department (K-6) of the Faculty of Electronics, Wroclaw University of Technology.