Close
0%
0%

Tennis AI

This is an implementation of a Reinforcement Learning based agent to simulate a robot that plays tennis.

Similar projects worth following
Training robots to play sports can be a challenging task. Here in this implementation, we trained Reinforcement Learning agents to play tennis. With the help of this implementation, we have a proof of concept model and architecture that can help an agent play tennis.

This simulation can then be used for the real-time implementation of robots that can play tennis. Such robots will have a lot of importance as they can assists sports players in their training, play with kids, etc. Due to the COVID-19 pandemic as kids are not able to play with each other these robots can become their buddies and play with them.

md - 4.09 kB - 06/18/2021 at 02:03

Download

md - 2.05 kB - 06/18/2021 at 02:03

Download

pth - 286.90 kB - 06/18/2021 at 02:03

Download

pth - 294.87 kB - 06/18/2021 at 02:03

Download

pth - 48.76 kB - 06/18/2021 at 02:03

Download

View all 11 files

  • 1 × Mouse
  • 1 × Keyboard Development Kits, Boards and Systems / Development Kits and Boards
  • 1 × Panda PAU05 WiFi Adapter The Jetson nano does not have a Wi-Fi interface
  • 1 × 32 GB SD card
  • 1 × 5v Power Supply (Barrel Jack Type)

View all 8 components

  • Testing 2

    srimanthtenneti06/28/2021 at 15:18 0 comments

    Completed testing the Critic-Network and tuning

  • Testing 1

    srimanthtenneti06/28/2021 at 15:17 0 comments

    Completed initial actor-network testing and tuning

  • Completion log

    srimanthtenneti06/18/2021 at 02:48 0 comments

    1.  Finished defining the Actor and Critic models design.

    2. Completed the DDPG implementation with the OU noise process and a replay buffer.

    3. Finished setting up the soft-target updates for the agents

    4. Completed and successfully trained the RL agents. 

    5. Completed the software documentation.

  • Completion log

    srimanthtenneti06/18/2021 at 02:11 0 comments

    1.  Finished defining the Actor and Critic models design.

    2. Completed the DDPG implementation with the OU noise process and a replay buffer.

    3. Finished setting up the soft-target updates for the agents

    4. Completed and successfully trained the RL agents. 

    5. Completed the software documentation.

View all 4 project logs

  • 1
    Setup Instructions

    1. Download and setup Unity ML agents

    2. Install Anaconda and Pytorch

    3. Run the ipynb notebook. 

View all instructions

Enjoy this project?

Share

Discussions

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates