🤖 Pickasso Unified Robotic System

📦 Project Overview

A fully autonomous trash-picking robot with integrated:

• 📸 Vision Inference (YOLO + MediaPipe)
• 📖 Ethical Decision-Making Engine
• 📡 Navigation (ROS)
• 🤖 Manipulation Simulation
• 🎛️ HMI Dashboard (React + Convex)
• 🗣️ Speech Interface (Recognition + TTS)

---

🚀 Deployment Steps

1. Clone the Repository

git clone https://github.com/AapseMatlb/pickasso_unified.git
cd pickasso_unified

2. Install Dependencies

 pip install -r requirements.txt
 cd hmi/frontend && npm install

3. Start the Complete System

chmod +x startup/start_simulation.sh
./startup/start_simulation.sh

🧩 Subsystem Details

Subsystem	Access/Command
HMI Dashboard	http://localhost:3000
Vision Inference	Runs Automatically
Decision Engine	Runs Automatically
Navigation & Manipulation	Simulated using ROS
Speech Interface	Accepts Simple Voice Commands

---

📖 For detailed configuration and subsystem usage, refer to the respective directories.

---

👨‍💻 Developed by: Yashashwani Kashyap

📚 Supporting Repositories

• Pickasso Trash Picking Simulation
• Pickasso Speech
• Pickasso Navigation Stack
• Pickasso HMI
• Pickasso Vision Ethics
• Pickasso Manipulation
• Pickasso Navigation Stack Simulation
• Pickasso Vision Inference

---

🛠️ Hardware Requirements (For Real Robot Deployment)

• Intel RealSense D435i Camera
• OpenManipulator-X Arm
• TurtleBot3 Waffle Pi Base

---

⚙️ Environment Setup

• Python 3.8+
• ROS Noetic (Navigation & Manipulation)
• Node.js 16+ (HMI Dashboard)
• Convex Backend Account

---

📦 Pre-trained Models

Ensure YOLOv8 models are downloaded and placed under : vision_inference/models/

---

🌐 Convex Backend Configuration

Update CONVEX_API_URL in all relevant scripts with your deployed Convex backend URL.

To Run HMI Frontend:

1. Navigate to frontend and run:

npm install
npm start

To Run Convex Backend:

1. Navigate to convex directory and run:

npx convex dev