Autonomous Mobile Robot Perception & Navigation

Production-grade ROS 2 navigation stack for Husarion ROSbot XL with Ouster OS1 3D LiDAR. Features high-performance C++ perception nodes, 2D/3D SLAM integration, and automated stability validation.

---

Key Features

Feature	Implementation
Ouster OS1 Time Sync	C++ timestamp_fixer_node - Zero-copy PointCloud2 processing
2D/3D Mapping	SLAM Toolbox (2D) + Octomap Server (3D) running in parallel
Velocity Control	C++ velocity_scaler_node - Configurable gain for motor tuning
Stability Validation	C++ waypoint_looper_node - Automated waypoint loops with metrics
TF Bridge	C++ odometry_bridge_node - Optional odom→base_link broadcast

---

System Architecture

┌─────────────────────────────────────────────────────────────────────────┐
│                           LAPTOP (Navigation Station)                   │
│  ┌─────────────────┐  ┌─────────────────┐  ┌─────────────────────────┐  │
│  │ timestamp_fixer │→ │ pointcloud_to_  │→ │     SLAM Toolbox        │  │
│  │   (C++ Node)    │  │   laserscan     │  │   (2D Map + Localize)   │  │
│  └────────┬────────┘  └─────────────────┘  └─────────────────────────┘  │
│           │                                                              │
│           ↓                                                              │
│  ┌─────────────────┐  ┌─────────────────┐  ┌─────────────────────────┐  │
│  │ octomap_server  │  │     Nav2        │→ │   velocity_scaler       │  │
│  │  (3D Mapping)   │  │ (Path Planning) │  │     (C++ Node)          │  │
│  └─────────────────┘  └─────────────────┘  └───────────┬─────────────┘  │
└───────────────────────────────────────────────────────────────────────┬─┘
                                                                        │
                              WiFi / ROS 2 DDS                          │
                                                                        ↓
┌─────────────────────────────────────────────────────────────────────────┐
│                        HUSARION ROSBOT XL                               │
│  ┌─────────────────┐  ┌─────────────────┐  ┌─────────────────────────┐  │
│  │  Ouster OS1     │  │ robot_localization│ │    Motor Drivers       │  │
│  │  (via Jetson)   │  │     (EKF)        │  │  ← /cmd_vel            │  │
│  └─────────────────┘  └─────────────────┘  └─────────────────────────┘  │
└─────────────────────────────────────────────────────────────────────────┘

---

Hardware Requirements

Component	Model	Purpose
Robot Platform	Husarion ROSbot XL	Mobile base with motor control
3D LiDAR	Ouster OS1-64	360° point cloud perception
Coprocessor	NVIDIA Jetson	Ouster driver host
Workstation	Ubuntu 22.04 Laptop	Navigation, SLAM, visualization

See hardware_bom.md for detailed bill of materials.

---

Software Prerequisites

• OS: Ubuntu 22.04 (Jammy)
• ROS 2: Humble Hawksbill
• C++ Standard: C++17
• Python: 3.10+

---

Installation

1. Install ROS 2 Dependencies

sudo apt update
sudo apt install -y \
  ros-humble-navigation2 \
  ros-humble-nav2-bringup \
  ros-humble-slam-toolbox \
  ros-humble-pointcloud-to-laserscan \
  ros-humble-robot-localization \
  ros-humble-tf2-ros \
  ros-humble-octomap-server

2. Install Clearpath Packages

Follow the Clearpath Robotics Documentation for base configuration tools.

3. Clone and Build

# Clone repository
cd ~/ros2_ws/src
git clone https://github.com/RuGG12/Autonomous-Mobile-Robot-Perception-Navigation.git

# Install dependencies
cd ~/ros2_ws
rosdep install --from-paths src --ignore-src -r -y

# Build (C++ and Python hybrid package)
colcon build --packages-select inspector --symlink-install
source install/setup.bash

---

Usage

Launch Full Navigation Stack

# Terminal 1: LiDAR Pipeline (C++ timestamp sync + 3D→2D conversion)
ros2 launch inspector lidar_converter.launch.py

# Terminal 2: Navigation + SLAM (C++ velocity scaler + Nav2 + SLAM Toolbox)
ros2 launch inspector clearpath_standard_demo.launch.py

Launch 3D Mapping (Optional)

# Runs Octomap server parallel to 2D SLAM
ros2 launch inspector 3d_mapping.launch.py

Run Stability Validation

# Automated waypoint loop testing with metrics logging
ros2 launch inspector waypoint_validation.launch.py loop_count:=10

# Check results
cat stability_log.txt

---

Package Structure

inspector/
├── src/                              # C++ Source Files
│   ├── timestamp_fixer_node.cpp      # PointCloud2 timestamp sync
│   ├── velocity_scaler_node.cpp      # Velocity gain control
│   ├── odometry_bridge_node.cpp      # Odom→TF bridge
│   └── waypoint_looper_node.cpp      # Stability validation
├── launch/                           # Launch Files
│   ├── lidar_converter.launch.py     # LiDAR pipeline
│   ├── clearpath_standard_demo.launch.py  # Full nav stack
│   ├── 3d_mapping.launch.py          # Octomap integration
│   └── waypoint_validation.launch.py # Stability testing
├── config/
│   ├── nav2_params.yaml              # Navigation parameters
│   └── waypoints.yaml                # Test waypoint coordinates
├── CMakeLists.txt                    # C++ build configuration
└── package.xml                       # Package manifest

---

C++ Node Details

timestamp_fixer_node

Purpose: Fixes Ouster OS1 hardware clock mismatch (boot time ~700s vs system wall time)

ros2 run inspector timestamp_fixer_node \
  --ros-args -p input_topic:=/ouster/points -p output_topic:=/ouster/points_fixed

Features:

• Zero-copy UniquePtr callback for high throughput
• SensorDataQoS for reliable sensor data
• Multi-threaded executor

---

velocity_scaler_node

Purpose: Applies configurable gain to velocity commands for motor response tuning

ros2 run inspector velocity_scaler_node \
  --ros-args -p scale_factor:=1.5

Parameters:

• scale_factor (double, default: 1.5) - Velocity multiplier
• input_topic (string, default: /cmd_vel_inspector)
• output_topic (string, default: /cmd_vel)

---

odometry_bridge_node

Purpose: Broadcasts odom → base_link TF from odometry messages

ros2 run inspector odometry_bridge_node \
  --ros-args -p publish_tf:=false

⚠️ Note: Set publish_tf:=false (default) when robot_localization EKF is running to prevent TF tree conflicts.

---

waypoint_looper_node

Purpose: Validates navigation stability via repeated waypoint loops

ros2 run inspector waypoint_looper_node \
  --ros-args --params-file config/waypoints.yaml

Output: Logs success/failure metrics to stability_log.txt:

Loop 1 | WP 0 | (0.00, 0.00) | SUCCESS | 2.34s
Loop 1 | WP 1 | (2.00, 0.00) | SUCCESS | 3.12s
...
SUMMARY:
Total Loops: 10 | Success Rate: 97.5%

---

Configuration

Navigation Tuning

Edit config/nav2_params.yaml:

• Speed limits, acceleration
• Costmap resolution and inflation
• Goal tolerances

Waypoint Testing

Edit config/waypoints.yaml:

waypoint_looper:
  ros__parameters:
    waypoints:
      x: [0.0, 2.0, 2.0, 0.0]
      y: [0.0, 0.0, 2.0, 2.0]
      yaw: [0.0, 0.0, 1.57, 3.14]
    loop_count: 10

LiDAR Mounting

Update static transform in launch/lidar_converter.launch.py:

arguments=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'base_link', 'os_sensor']

---

Network Configuration

All devices must be on the same network with matching ROS_DOMAIN_ID:

# On all devices
export ROS_DOMAIN_ID=0

---

Troubleshooting

Issue	Solution
TF lookup errors	Ensure lidar_converter.launch.py is running first
Robot shaking in RViz	Set publish_tf:=false on odometry_bridge (EKF conflict)
No point cloud	Check Ouster driver on Jetson: ros2 topic echo /ouster/points
Navigation fails	Verify /scan topic: ros2 topic hz /scan

---

License

This project is licensed under the MIT License - see LICENSE for details.

---

Author

Rugved Raote
rugvedraote@gmail.com