Robotic Suite

Advantech Robotic Suite is an integrated software environment designed to accelerate the development, deployment, and management of robotic applications.

It reduces integration complexity by providing pre-validated software stacks, containerized environments, and ready-to-use workflows for modern Robot development.

It is more than just an SDK – it is a complete development and execution ecosystem.

Support & Community

• Advantech AIM-Linux Community: https://forum.aim-linux.advantech.com/c/robotics-autonomous-systems-ros-ros2/40
• Robotics Suite is thumbs up or down? Fork, use and take polls while getting your chance to win a developer grand prize. https://forum.aim-linux.advantech.com/t/fork-test-and-critique-good-or-bad-tell-us-what-you-think-about-the-robotics-suite/1479

What’s Included

Robotic Suite brings together the software stacks, containerized environments, example workflows, and learning resources needed for modern AMR and manipulator application development:

• Core Techniques – Focused, code-level tutorials covering reusable robotics building blocks, including HSV color-profile creation, color-block detection and camera-frame position estimation.
  Each technique can be studied independently and integrated into larger AMR or manipulator applications.

• ROS 2 – Core middleware for robotic applications, including: ros2_control, rqt, rviz2, robot_localization, imu_tools, Navigation2, rtabmap_ros, LIO-SAM, cartographer, slam_toolbox
  These provide motion control, visualization, sensor fusion, SLAM, and autonomous navigation out of the box.

• Isaac ROS – NVIDIA GPU–accelerated perception and robotics pipelines, including: isaac_ros_nvblox, isaac_ros_visual_slam, isaac_ros_yolov8, isaac_ros_unet, isaac_ros_centerpose, isaac_ros_occupancy_grid_localizer, isaac_ros_cumotion_moveit
  These packages cover 3D reconstruction, visual SLAM, object detection, segmentation, pose estimation, local occupancy mapping and arm control for high-performance perception workloads.

  

• QIR ROS SDK – Qualcomm NPU–accelerated perception and robotics pipelines, including: ocr_service, sample_object_detection, sample_object_segmentation, sample_depth_estimation, simulation_sample_pick_and_place, simulation_follow_me
  These packages and examples cover OCR, object detection, segmentation, depth estimation, simulation, and follow-me workflows using Qualcomm NPU acceleration.

  

• Autoware – Open-source stack for autonomous driving and advanced applications

  

• Containerized runtime – Pre-integrated and version-controlled environments to simplify setup, and ensure reproducibility.

This allows you to focus on building your application logic instead of spending time on basic integration and environment setup.

Download

• Installer

Compatible Products

Robotic Suite is optimized and validated on selected Advantech platforms, including:
NVIDIA:

• AFE-A702 / ASR-A702
  NVIDIA® Jetson Thor™ AMR Control System & Control Board

• AFE-R750 / ASR-A701
  NVIDIA® Jetson Orin™ NX / AGX Orin™ AMR Control System & Control Board

Qualcomm:

• AFE-A503 / ASR-A503
  Qualcomm® IQ9075M Robot Control System & Control Board

• ASR-D501
  Qualcomm® QCS6490 Compact companion/mission computer

Development Utilities

Robotic Suite provides learning and development resources organized into three main categories:

• Example Workflows – Ready-to-run perception, AMR, and examples for understanding complete robotics application workflows.
• Core Techniques – Code-level tutorials covering reusable robotics techniques that can be integrated into your own applications.
• Development Environment – Pre-configured ROS 2, Isaac ROS, QIR ROS SDK, and Autoware environments for application development.

Core Techniques

• Color Profile Creation
  Collect HSV samples from a live RGB image and create reusable color profiles for color-based detection.

• Color Block Detection & Position Estimation
  Detect colored blocks and estimate their camera-frame position and approximate orientation using RGB, aligned depth, and camera information.

---

Example Workflows

ROS Base

Manipulation

• Manipulator
  Explore a recorded vision-based manipulator workflow that connects color-block perception and camera-frame position estimation with robot-arm motion visualization.

Mobility

• 2D LiDAR Mapping
  Create a 2D occupancy grid map from a 2D LiDAR using slam_toolbox, and visualize it in RViz.

• 2D LiDAR Navigation
  Learn the basic concepts of Nav2-based navigation and obstacle avoidance with a 2D LiDAR.

• 3D LiDAR Mapping
  Generate a 3D map from a 3D LiDAR using lidarslam_ros2, and inspect the result in RViz.

---

ISAAC ROS

• Multi-Sensor Function
  Combine multiple sensors into a unified perception pipeline using Isaac ROS.

• 3D Scene Persistence
  Build and maintain a 3D representation of the environment for long-term perception.

• 3D Camera Obstacle Avoidance
  Use 3D cameras and Isaac ROS to detect obstacles and plan safe motion around them.

---

QIR ROS

• 2D Object Depth Estimation
  Use a recorded 2D camera stream to run sample_depth_estimation and sample_apriltag simultaneously, demonstrating how multiple QIR ROS SDK packages can share the same sensor input.

Development Environment

Robotic Suite provides pre-configured development environments to help users quickly get started with robotics application development.

• ROS Jazzy
  ROS 2 Jazzy is pre-installed and configured on the host system, providing a stable and ready-to-use foundation for ROS-based development.

• ROS Humble
  ROS 2 Humble is pre-installed and configured on the host system, providing a stable and ready-to-use foundation for ROS-based development.

• ISAAC ROS
  A fully integrated Docker-based environment is provided, including pre-installed ISAAC ROS dependencies.
  This allows users to quickly build and run GPU-accelerated perception and AI applications without complex setup.

• QIR ROS SDK
  A fully integrated Docker-based environment is provided, including pre-installed QIR ROS SDK dependencies.
  This allows users to quickly build and run applications without complex setup.

• Autoware
  A containerized Autoware environment is provided, enabling users to explore autonomous driving workflows such as perception, planning, and control with minimal setup effort.

Typical Workflow with Robotic Suite

1. Explore a ready-to-run example – Use the provided ROS bag–based perception, AMR, or manipulator examples to understand the application workflow without requiring physical sensors or robots.
2. Build the robot workflow – Learn the main components for your application:
   • Perception – Multi-function camera processing, depth estimation, detection, segmentation, and scene understanding.
   • AMR – Mapping, localization, navigation, and obstacle avoidance.
   • Manipulator – Perception, position estimation, motion planning, and robot-arm action.
3. Extend with platform SDKs and Core Techniques – Use Isaac ROS, QIR ROS SDK, or reusable Core Techniques to add accelerated perception and application-specific functions.
4. Adapt the example to your hardware – Replace the provided ROS bag with real sensor data and configure the required topics, TF relationships, robot description, drivers, and parameters.
5. Deploy and validate on Advantech platforms – Run the customized application in the provided containerized environment and verify its behavior on the target platform.