Autonomous Drone Light Simulator

A Python-based drone simulation platform for learning autonomous flight control — from basic FSM logic to real physics with MuJoCo.

Projects

Folder	Description	Complexity
maria1/	Kinematic simulator with FSM brain, fake sensors, and 3D visualization	Beginner
maria_physics/	MuJoCo-based physics with PID cascade controller	Intermediate

Quick Start

# Clone the repo
git clone https://github.com/OGODEVO/Automous-drone-light-sim.git
cd Automous-drone-light-sim

# Run kinematic sim (no physics, just logic)
cd maria1
pip install matplotlib numpy
python main.py

# Run physics sim (real forces and torques)
cd ../maria_physics
pip install mujoco
python main.py              # Headless
mjpython main.py --gui      # With 3D viewer (macOS)

Architecture Overview

┌────────────────────────────────────────────────────────────┐
│                        BRAIN (FSM)                         │
│              SEARCH → TRACK → AVOID → RECOVER              │
└────────────────────────────────────────────────────────────┘
                              ↓
┌────────────────────────────────────────────────────────────┐
│                    CONTROLLER (PID)                        │
│         Velocity → Attitude → Rate → Motor Mixing          │
└────────────────────────────────────────────────────────────┘
                              ↓
┌───────────────────────┬───────────────────────────────────┐
│      maria1 (Fake)    │       maria_physics (Real)        │
│   Instant velocity    │   MuJoCo forces & torques         │
└───────────────────────┴───────────────────────────────────┘

Learning Path

1. Start with maria1 — Understand FSM states, perception, and gate tracking
2. Move to maria_physics — Learn PID control and motor mixing
3. Tune the gains — Experience why real drones are hard to fly
4. Port the brain — Connect FSM to physics controller

Requirements

• Python 3.10+
• numpy, matplotlib (maria1)
• mujoco (maria_physics)

License

MIT