A modular autonomous driving simulation platform with computer vision, deep learning, and sensor fusion. Features lane detection, object recognition, and adaptive control in BeamNG.tech, with real-time visualization via Foxglove.

SimBEV is a configurable and scalable synthetic driving data generation tool based on the CARLA Simulator.

Carla Simulation + Model training. Model playground. ML Workflow Builder

PlanT 2.0: Exposing Biases and Structural Flaws in Closed-Loop Driving

Autonomous vehicle motion planning system with behavioral planning, local path optimization, and velocity profiling. Features collision avoidance, stop sign compliance, and lead vehicle following. Implemented in Python with CARLA simulator integration.

[CoRL'22] PlanT: Explainable Planning Transformers via Object-Level Representations

Blender add-on for creating OpenDRIVE and OpenSCENARIO based automotive driving scenarios including 3D models

A PyTorch implementation of end-to-end learning for self-driving cars, inspired by NVIDIA's "End to End Learning for Self-Driving Cars" paper on CARLA simulator for data collection and testing with custom implementation

Digital twin of the University of Michigan’s Mcity Test Facility to support research and development of transportation technology, both academic and commercial.

Deep Reinforcement Learning (PPO) in Autonomous Driving (Carla) [from scratch]

Traffic Scene Generation from Natural Language Description for Autonomous Vehicles with Large Language Models

The aim of this work is to build an end-to-end pipeline for autonomous driving where the trained agent controls the vehicle to follow routes and minimize accidents. The project is divided into three main components:

Autonomous driving with lane detection using CARLA simulator, OpenCV, and Python - Real-time lane tracking with sliding window algorithm

Reinforcement Learning-based project using PPO algorithm to enable a car to drive autonomously and efficiently in the CARLA simulation environment, optimizing for both speed and safety.

Race a Tesla Model 3 with model predictive control.

[WACV 2025] Official code of "SEED4D: A Synthetic Ego-Exo Dynamic 4D Data Generator, Driving Dataset and Benchmark"

[ICCV'23] Hidden Biases of End-to-End Driving Models & A starter kit for the CARLA leaderboard 2.0.

BSc (Hons) Computing - Dissertation project on Ethical Autonomous Vehicle Decision-Making using CARLA simulation, machine learning, and a Streamlit app for teacher–student decision comparison

Drive a Tesla Model 3 in CARLA Simulator on Windows using Python and keyboard controls (WASD/Arrow keys). No pygame required.

🚘⚠️ RAAS is a modular driver assistance system developed in the CARLA simulator for integration into Russian vehicles such as Lada, Aurus, and others. It includes a wide range of ADAS features and proposes a concept and prototype of how such a system could be built for domestic cars, focusing on adaptability, open architecture, and real-world use.