🤖 Deep Q-Network AI Visualization

DQN Pathfinding & Neural Monitor

Watch an AI agent learn to navigate obstacles in real-time. Experience the power of Deep Q-Learning with live neural network visualization and interactive controls.

⚡ Real-time Learning 🧠 Neural Visualization 🎮 Interactive UI

Environment Visualization

Simulation Controls

Logic steps per frame: 1

Neural Monitor

Simulation Stats

Episode
0
Steps
0
Reward
0.0
Epsilon
1.000
Collisions
0

Agent State

pos: (0, 0)

Q-Values Output

Layer Activations

Hidden Layer 1 0/64 active
Hidden Layer 2 0/64 active

💡 The network learns to navigate from start (blue) to goal (red) while avoiding obstacles (gray). Watch the neural activations change as it learns!

Legend & Guide

Agent

The learner navigating the grid

Goal

Target destination

Obstacles

Boundaries to avoid

Path

Agent's movement history