On Thursday, we, together with the teacher Misha Dratovany of our SAKMIG VNTU department, who is also a 3rd year postgraduate student of specialty 124 – Systems Analysis of our department, conducted a practical lesson on the topic “Computerized data monitoring using drones” for VNTU students of specialty 126 – Information Systems and Technologies (educational program “Applied Information Technologies”) from their main discipline “Applied Information Technologies”. As part of the lesson, we found a deserted place and there we studied various modes of collecting information from a quadcopter, including “Follow me”, “Circle flight”, etc. This was preceded by my lecture explaining the modes, capabilities, operating rules and safety precautions using drones. And during the practical lesson, I showed myself how everything should work, and let them apply these modes themselves, simply controlling this process. A deserted place, a wind speed of 1.5 m/s, and the requirement to keep the drone at a height and away from people provide maximum safety for the operator and others present, which is important during the lesson.
The HUBSAN H501M X4 quadcopter (https://hubsan.in.ua/product/hubsan-h501m-x4-waypoints-fpv/) with an additional battery was kindly purchased for our department last year by one of the members of the group providing specialty 124 of our same department, and we often use it in classes for specialties 124 and 126 in the 2nd-3rd years and in the master's degree with 126.
Last semester, in the block of disciplines “Information Technologies for Monitoring and Analysis” (ITMA), students of specialties 126 (discipline “ITMAD”) and 124 (“ITMASSS”) used this drone to collect photo and video data, which we then processed using deep learning methods and pattern recognition in Python using convolutional autoencoders, and also performed clustering of image fragments using PCA and t-SNE methods using projector.tensorflow. In the discipline “PIT” we will continue to use these technologies, but in more applied areas. We will talk about this in the following posts.
We will also fly this drone later, with further image processing using deep learning (DL) methods, with two newly recruited groups of masters in specialty 126, where I teach ITMASSS for the entire stream. But first, we need to brush up on their Python and knowledge of the necessary libraries (tensorflow, keras, etc.).