openpilot is an operating system for robotics. Currently, it upgrades the driver assistance system on 300+ supported cars.

Eye state localisation and detection for use in Advanced Driver Assistance System.

Multi-task learning using message passing graph neural network for radar based perception functions

Prediction of the intention of pedestrians to cross the street or not, using Graph Neural Networks and the coordinates of their skeleton that was previously generated using Openpose in the JAAD dataset.

🚸⛔Novel Deep Convolutional Network is proposed for traffic sign classification that achieves outstanding performance on GTSRB surpassing the best human performance of 98.84%.

Repository for the MSc project "To Err is Human: Designing Correct-by-Construction Driver Assistance Systems using Cognitive Modelling".

Using Blender and Python to generate synthetic data for training an 8-class YOLOv5 detector in a highway driving scenario (Sedan, SUV, Van, Truck, Bus, Warning Sign, Speed Limit Sign, Speed Cam Sign).

openpilot is an operating system for robotics. Currently, it upgrades the driver assistance system on 275+ supported cars.

🤖 Explore open-source robotics models and packages, including advanced vision-language-action systems for versatile applications and fine-tuning options.

This repository contains the code I wrote for an external industry representative from General Motors who works as an Active Safety Technical Expert. This project reads in data from GM Advanced Driving Assistance Systems and stores it in a memory-efficient linked list, and then visualizes that data.

Real-time ADAS using MiDaS depth estimation and YOLO object detection for collision alerts, lane departure warnings, and intuitive visual/audio feedback.

This repository contains the steps and processes followed in developing and evaluating an automated IR to RGB pedestrian detection pipeline

Advanced Multi-Modal Sensor Fusion System for Detecting Falling Humans: Quantitative Evaluation for Enhanced Vehicle Safety

🛠️ Build structured, JSON-serializable data for human-in-the-loop ML workflows with LabelingIR, enhancing collaboration across your labeling stack.