Quad Move

Code Authors: Advait Desai and Gargi Gupta
Program: SRA VJTI's Eklavya 2025 Program

---

Sample Demonstrations

Go2 Quadruped Walk	Lunar Lander

Blackjack	Frozen Lake

---

About the Project

We aim to train a PPO-based gait policy in MuJoCo and deploy it on a low-cost, tortoise-style quadruped robot.

Project Flow

1. Learn Reinforcement Learning basics (Monte Carlo, Q-Learning)
2. Learn Deep Learning (MNIST digit classifier)
3. Combine Deep Learning with RL (DQN, DDQN, TRPO, PPO)
4. Implement on environments (Brax, MuJoCo Menagerie, Bipedal, Go2)

---

Project Components

Foundation Learning

• Monte Carlo Methods - Basic RL policy evaluation
• Q-Learning - Value-based reinforcement learning
• Deep Learning - MNIST digit classification for neural network fundamentals

Advanced RL Implementation

• DQN/DDQN - Deep Q-Networks for discrete action spaces
• TRPO/PPO - Trust Region and Proximal Policy Optimization for continuous control

Environment Training

• Brax - High-performance physics simulation
• MuJoCo Menagerie - Diverse robotics environments
• Bipedal Walker - Humanoid locomotion
• Go2 Quadruped - Advanced quadruped robot simulation

---

File Structure

quad_move_eklavya/
├── 1_brax_training_viewer/
├── 2_mujoco_menagerie/
│   └── flybody/
├── 3_monte_carlo/
├── 4_Q_learning/
├── 5_number_classifier/
├── 6_DQN/
├── 7_DDQN/
├── 8_PPO_bipedal/
└── 9_PPO_go2_stable_baselines/

---

Getting Started

Clone the Repository

git clone https://github.com/Advait2211/quad_move_eklavya.git
cd quad_move_eklavya

Create a Virtual Environment

python3.10 -m venv venv

Activate the Virtual Environment

macOS/Linux:

source venv/bin/activate

Windows (PowerShell):

.\venv\Scripts\activate

Install Dependencies

pip install -r requirements.txt

---

Requirements

• Python 3.10+
• MuJoCo physics engine
• PyTorch/TensorFlow for deep learning
• Stable Baselines3 for RL algorithms
• Additional dependencies listed in requirements.txt