During my Master's studies at the university, I developed a UAV landing system for quadcopters. This feature allows self-landing of your drone on a special symbol on the ground. It means that drone can catch the target via its optical vision system and then slowly go down until its full landing on the symbol.
As the results of this project, a test drone was programmed to fly to a defined place via GPS navigation, where it recognized a prepared symbol "H" on the ground. Then, the drone did self-landing on that symbol with 93% accuracy.

In this package I provide a description of different data structures and algorithms with my own implementation and usage cases.
In addition, I solved more than 230 programming exercises.

In this project, I tried to the attitude between compressive strength of concrete and its attributes. Keras and sklearn are in the use to complete this project. The dataset is about the compressive strength of different samples of concrete based on the volumes of the different ingredients that were used to make them.
The outcome: 94% accuracy 2-layer neural network was built in Python and Keras.

The 200 W PoL reference design is my own system solution for data centers and telecom DC-DC power modules, which operates across a wide range of input voltages (36V - 75V). The reference design introduces a single step digital controlled hybrid Point-of-Load solution with a synchronous buck and a half-bridge current-doubler delivering 12V/130W and 1V/70W output power accordingly.

The outcome: evaluated a Loan Indebtedness with people different categories and tried to predict if we should give a new loan to certain candidates. In this project I used K Nearest Neighbor (KNN), Decision Tree, Support Vector Machine and Logistic Regression techniques.

In this project, I built a collaborative filtering-based Recommendation System to predict users' movie preferences. To do that, I applied Restricted Boltzmann Machine in Python with TensorFlow.

In this project, I deployed a Convolutional Neural Network in Python with PyTorch and using MNIST data and achieved an 89% model accuracy with just 7 training iterations.

At my Bachelor I developed an optical vision system to track special programmed objects. For instance, it can be an orange ball. The main purpose of the system was to navigate a self-driven car to follow an object.
This system contains a camera, 2 piezo-motors for moving in vertical and horizontal axes, and a Raspberry Pi as a head microcomputer. I developed a mechanical part of this system and its firmware on Python.