Autonomous Navigation System (ANS)

A 3-wheeled omnidirectional robot capable of mapping unknown environments and navigating autonomously using ROS 2. Built with a Kiwi drive mechanism, this bot can travel in any direction on a 2D plane.

---

📸 Overview

• Electronics

  • Raspberry Pi 5 (ROS 2 master, sensor interface)
  • SLLIDAR (LiDAR sensor)
  • 3 × Arduino Nano (one per motor with encoder)
  • BNO055 IMU (orientation + acceleration data)
  • 3 × 12V geared DC motors with encoders
  • 4-cell 14.8V battery (motors power)
  • Buck converter (14.8V → 5V) (encoder + logic power)

  

• Mechanics

  • Triangular laser-cut chassis
  • 3 omni-wheels, one at each corner
  • Kiwi drive configuration

• Software

  • OS: Ubuntu 24.04
  • ROS 2 Jazzy Jalisco
  • SLAM Toolbox (mapping)
  • Nav2 (navigation + path planning)

---

🛠 Code Structure

├── ans_ws/                         # Main ROS 2 workspace
│   ├── bno055_driver/              # IMU driver
│   │   └── bno055_node             # Publishes /imu/data, /imu/euler, /imu/quaternion
│   │
│   ├── encoder_publish/            # Motor control + encoder handling
│   │   ├── forward.py              # Test: move forward
│   │   ├── backward.py             # Test: move backward
│   │   ├── calc_rpm.py             # Converts raw counts → /rpm{A,B,C}
│   │   ├── joystick.py             # Control with joystick
│   │   ├── omni_teleop.py          # Control with keyboard
│   │   ├── publish.py              # Serial comms with Arduinos (encoder + motor commands)
│   │   └── rotate.py               # Test: rotate bot
│   │
│   ├── pos_calc/                   # Odometry
│   │   └── pos_calc                # Combines IMU + RPM → /odom
│   │
│   └── ...
│
├── ws_lidar/                       # Lidar workspace
│   └── sllidar_ros2/               # Publishes /scan
│
├── arduino_code.cpp                # Runs on each Nano (encoder + motor control)
│
└── ...

---

⚙️ System Flow

1. On the Bot (Raspberry Pi 5):

   • Collects data from LiDAR, IMU, and encoders.
   • Publishes sensor topics via ROS 2.

2. On the PC (Remote workstation):

   • Subscribes to shared topics over WiFi.

   • Processes sensor data:

     • Compute RPM & odometry.
     • Map environment using SLAM Toolbox.
     • Plan paths with Nav2.

   • Sends velocity/motor commands back to Pi → Arduinos → motors.

---

🚀 Setup & Usage

1. Network Configuration

• Setup a PC with Ubuntu 24.04 and ROS 2 Jazzy.

• Install slam_toolbox and Nav2 on the PC.

• Clone this repo to your PC and build ans_ws.

2. Network Configuration

• Connect Raspberry Pi and PC to the same WiFi.

• Set ROS domain ID (on both PC and Pi):

  export ROS_DOMAIN_ID=0

• SSH into Pi:

  ssh ans@<pi-ip-address>
  # password: ans

3. Run on Raspberry Pi

In separate terminals:

sudo ./start.sh
ros2 launch sllidar_ros2 sllidar_a1_launch.py
ros2 run bno055_driver bno055_node
ros2 run encoder_publish encoder_publish

4. Run on PC

In separate terminals:

ros2 run encoder_publish calc_rpm
ros2 run pos_calc pos_calc
ros2 launch slam_toolbox online_async_launch.py

5. Visualization

rviz2

• View map on /map.
• Monitor odometry on /odom.
• Send navigation goals via Nav2 RViz panel.

---

🧭 Features

• Omni-directional Kiwi drive.
• Real-time mapping & localization with LiDAR + IMU + encoders.
• Autonomous navigation with obstacle avoidance.
• Teleop control (joystick + keyboard).

---