Two AI Fight for the same Parking Spot

After training a car to park (watch the training here:    • AI Learns to Park - Deep Reinforcemen...  ), I decided to put another agent into the environment and let them fight for the same parking spot.
I thought the result was quite fun to watch, so here is a video of it!

The AI consists of a deep Neural Network with 3 hidden layers of 128 neurons each. It is trained with the Proximal Policy Optimization (PPO) algorithm, which is a Reinforcement Learning approach.

Basically, the input of the Neural Network are the readings of eight depth sensors, the cars current speed and position, as well as its relative position to the target.

Subscribe for more content like this:
   / @samuelarzt  

Follow me on Twitter for more frequent updates on my projects:
  / samuelarzt  

Also check out my other videos related to this Project:
Neural Networks Explained in a Minute:
   • Explained In A Minute: Neural Networks  

Cars learn to maneuver Parcour with Genetic Algorithm:
   • Deep Learning Cars  


Music from Bensound.com:
Outro: "All That"

Background Music:
Scott Joplin: Maple Leaf Rag

#AI #MachineLearning #AIFight #ReinforcementLearning #ArtificialIntelligence #NeuralNetworks