bnspopi/openenv-scaler

Openenv Scaler

Smart Grid Navigation Reinforcement Learning (RL) Environment built using FastAPI. This project simulates an agent navigating a grid with rewards and goals.

🚀 Features

• —Custom RL environment
• —Grid-based navigation
• —Reward system
• —API endpoints for interaction
• —Docker-based deployment
• —Live testing via Swagger UI

🧠 How It Works

• —Agent starts at position [0, 0]
• —Goal is at [4, 4]
• —Each step gives reward -1
• —Reaching goal gives reward +100
• —Environment resets after completion

📡 API Endpoints

• —/reset → Reset environment
• —/step → Take action
• —/state → Get current state
• —/health → Check API status

Direction Mapping

• —0 = UP
• —1 = DOWN
• —2 = LEFT
• —3 = RIGHT

🧪 Usage

Open Swagger UI at: /docs

Then:

1. 1.Call /reset
2. 2.Use /step with:

{
  "direction": 1
}