Skip to content

Training of a hexapod robot to walk using PPO algorithm, and applying the tripod walking method.

Notifications You must be signed in to change notification settings

Ammar500Issa/Hexapod

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 

Repository files navigation

Hexapod Robot Reinforcement Learning

Project Description

This project involves applying reinforcement learning to a hexapod robot to learn how to walk toward a target position. The robot was built using Dynamixel AX-12A servos and controlled with a BeagleBone Blue board. A PyBullet simulation environment was created to train the robot using the Proximal Policy Optimization (PPO) algorithm implemented in PyTorch.

Key Features

  • Reinforcement learning for a hexapod robot
  • PyBullet simulation environment
  • PPO algorithm implementation in PyTorch
  • Kalman Filter for state estimation
  • Fine-tuning neural networks to adapt to real-world scenarios

Technologies Used

  • Python (PyBullet, PyTorch)
  • BeagleBone Blue
  • Dynamixel AX-12A Servos
  • Reinforcement Learning (PPO)

About

Training of a hexapod robot to walk using PPO algorithm, and applying the tripod walking method.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published