Deep Reinforcement Learning in C#

Make autonomous landing rockets using Deep Reinforcement Learning (Unity ML-Agents)

UAV Logistics Environment for Multi-Agent Reinforcement Learning / Unity ML-Agents / Unity 3D

Target Strike game is an unity based compititive game. This game is created for CS662 - Mobile VR & AI course offered at IIT Mandi. Here the source files are given.

My master's degree project, titled Quadcopter Control with Deep Reinforcement Learning and PID Controllers in Simulated Environments

Autonomous Aerial Vehicle

Reinforcement learning tecnique applied to PONG on unity

Top 8 Finalist Hackathon project and Urban Track Winner at HackMIT 2020

A small ML-Agents project to become familiar with the basics of machine learning in Unity

Final task for my Reinforcement Learning class in Deusto. The research paper discuss examples of using ML-Agents toolkit of Unity. Paper is avaible at:

Reinforcement Learning excercise where the agent is in a maze looking to find a square containing gold

This repository features a series of VR games designed to enhance cognitive therapy and muscle movement for children and young adults with movement disorders by integrating Reinforcement Learning (RL) techniques to generate immediate feedback and better results.

This repository contains a University project on the Reinforcement Learning subject, in which we develop a bot able to use some of Brandon Sanderson's Mistborn fantasy book's abilities in order to reach a goal in a 2D platform environment.

A simple demo showing asteroids with ML-Agents

Pacman application coded in C# and tested with NUnit

Thesis project exploring an AI system that dynamically adapts game difficulty in real time by analyzing key player performance indicators (reaction times, accuracy, strategic choices). The system tailors the experience to each player's unique playstyle and counteracts it to encourage adaptation and skill evolution.

Implemented a reinforcement learning-based autonomous parking system using TD3 in a custom Unity environment with ray sensors, stochastic initialization, and optimized reward shaping for efficient learning. 🚗🤖