/Flocking-RL

Reinforcement Learning Based Flocking

Primary LanguagePythonMIT LicenseMIT

Flocking-RL

An implementation of a 2D Flocking model based using RL.

Background

Flocking

Major Components

Alignment

Seperation

Cohesion

Result(So far)

Simulation.mp4

Prerequisites

Clone Repository

git clone github.com/Hamza-101/Flocking-RL

cd Flocking

Install Dependencies

pip install -r requirements.txt

Working

Program Structure

  1. Settings
  2. ConfigGenerator
  3. Simulation
  4. Analysis
  5. PlotAnimation

Supported Algorithms

  1. PPO (Proximal Policy Optimization) ✅
  2. TRPO
  3. DQN
  4. MADDPG

Links

Neural Simplex Architecture

Frameworks & Libraries

Open AI Gym , Stable Baselines3

Algorithms

PPO , TRPO DQN MADDPG