Seyed Ahmad Hosseini Miangoleh

Seyed Ahmad Hosseini Miangoleh

Electrical Engineer | AI & Robotics Enthusiast | Control Systems Specialist

I’m an Electrical Engineering student at Amirkabir University of Technology, specializing in Control Systems and Artificial Intelligence. My work bridges classical control theory with cutting-edge machine learning, deep learning, and reinforcement learning to build intelligent, autonomous systems. From robotics and autonomous driving to speech and image-based emotion recognition, I focus on creating real-world AI-powered solutions through research and development.

Featured Projects

IL-RL-ObstacleAvoidance

Autonomous driving system using Imitation Learning and Reinforcement Learning for obstacle avoidance, powered by computer vision and LiDAR in Webots simulator.

Python Webots OpenCV Reinforcement Learning

Speech-Emotion-Recognition-using-Wav2Vec2

Deep learning-based Speech Emotion Recognition system using Wav2Vec2 to detect emotions like happy, sad, angry from raw audio data.

Python Wav2Vec2 Jupyter Notebook Deep Learning

Twitter-Emotion-Classifier-using-Transformer-Encoder

Transformer-based model for emotion detection in tweets, using GloVe embeddings and custom text preprocessing for high-accuracy classification.

Python Transformers GloVe Jupyter Notebook

ResNetInception-CNN-Classifier-For-TinyImageNetDataset

CNN-based image classification using a ResNet-Inception hybrid model on TinyImageNet, with extensive hyperparameter tuning and performance analysis.

Python ResNet Inception Jupyter Notebook

MNIST-Deep-Learning-Saliency-Maps-and-FGSM-Attacks

Deep learning project for MNIST digit classification, featuring saliency map visualization and robustness evaluation using FGSM adversarial attacks.

Python TensorFlow Jupyter Notebook Deep Learning

RISC-V-Single-Cycle-Processor

VHDL single-cycle RISC-V processor supporting R-type (ADD, SUB, AND, OR) and I-type (ADDI, ANDI, ORI, LW, SW) instructions, with modular design.

VHDL RISC-V FPGA

Bachelor Thesis

BLIP-FusePPO Framework

A vision-language deep reinforcement learning framework for autonomous lane keeping using BLIP (Bootstrapped Language-Image Pretraining) and Proximal Policy Optimization (PPO). The system learns to drive in simulation by understanding visual inputs and high-level commands.

Python PyTorch BLIP PPO Webots Computer Vision NLP

Let's Connect!

I'm always interested in new opportunities and collaborations. Feel free to reach out!