LOOKING FOR THE RIGHT

INDUSTRIAL SOLUTION?

Discover embedded, communication and automation solutions tailored to your application.

ROScube-I Starter Kit

ROS Starter Kit with Intel Core Motherboard and MXM Graphics
  • Manufacturer Adlink Technology Inc.
  • Category: ROS Robots AV
  • The ADLINK ROScube-I Starter Kit is a high-performance embedded computing platform designed for robotics and AI applications. It integrates Intel® 8th/9th Generation Core™ i7, i5, or i3 processors with support for MXM Type A/B graphics modules, enabling GPU-accelerated AI computing. The system is fully compatible with ROS (Robot Operating System) and ROS 2, providing developers with a robust foundation for building intelligent robotic systems. It is part of ADLINK’s ROSCube series, which is optimized for real-time control, machine vision, and autonomous navigation tasks.
- ROS/ROS 2 Compatibility: Fully supports both ROS and ROS 2 frameworks, ensuring seamless integration with open-source robotic software libraries for motion control, vision, and navigation.
- Intel® Core™ Embedded Processor Options: Available with 8th/9th Gen Intel® Core™ i7-9700TE, i5-8500T, or i3-8100T processors, offering scalable performance for diverse robotics workloads.
- MXM Graphics Module Support: Compatible with ADLINK EGX-MXM series GPU modules (P1000, P2000, P3000, P5000), supporting Type A/B MXM modules up to 120W for AI acceleration.
- Rich I/O Interface: Provides extensive connectivity including multiple Gigabit Ethernet ports, USB 3.1/2.0, serial ports, digital I/O, and M.2 expansion slots for wireless and storage modules.
- AI and Vision Acceleration: Supports Intel® OpenVINO™ toolkit for optimized deep learning inference and computer vision applications.
- Compact Embedded Design: Small form factor (197.72 x 167.32 mm) suitable for integration into robotic platforms and industrial environments.
- Flexible Expansion: Includes PCIe x8 edge connector, M.2 slots (E, B, M keys), and optional TPM 2.0 and eSIM modules.
- Robust Environmental Tolerance: Operates in temperatures from 0°C to 60°C (without MXM) or 0°C to 55°C (with MXM), with humidity tolerance up to 90% non-condensing.
- Software Ecosystem: Bundled with ADLINK’s Neuron SDK, Ubuntu 18.04 LTS environment, and middleware support for ROS/ROS 2 and OpenVINO™.

Country of origin: Taiwan R.O.C.

1. Product Overview

Product Name: ROScube-I Starter Kit
Model Series: ROScube-I Series
Article Number: ROScube-I Starter Kit
Manufacturer: ADLINK Technology, Inc.
Product Type: ROS/ROS 2-based embedded computing platform with Intel® Core™ processor and MXM graphics module support

The ADLINK ROScube-I Starter Kit is a high-performance embedded computing platform designed for robotics and AI applications. It integrates Intel® 8th/9th Generation Core™ i7, i5, or i3 processors with support for MXM Type A/B graphics modules, enabling GPU-accelerated AI computing. The system is fully compatible with ROS (Robot Operating System) and ROS 2, providing developers with a robust foundation for building intelligent robotic systems. It is part of ADLINK’s ROSCube series, which is optimized for real-time control, machine vision, and autonomous navigation tasks.




2. Key Features

- ROS/ROS 2 Compatibility: Fully supports both ROS and ROS 2 frameworks, ensuring seamless integration with open-source robotic software libraries for motion control, vision, and navigation.
- Intel® Core™ Embedded Processor Options: Available with 8th/9th Gen Intel® Core™ i7-9700TE, i5-8500T, or i3-8100T processors, offering scalable performance for diverse robotics workloads.
- MXM Graphics Module Support: Compatible with ADLINK EGX-MXM series GPU modules (P1000, P2000, P3000, P5000), supporting Type A/B MXM modules up to 120W for AI acceleration.
- Rich I/O Interface: Provides extensive connectivity including multiple Gigabit Ethernet ports, USB 3.1/2.0, serial ports, digital I/O, and M.2 expansion slots for wireless and storage modules.
- AI and Vision Acceleration: Supports Intel® OpenVINO™ toolkit for optimized deep learning inference and computer vision applications.
- Compact Embedded Design: Small form factor (197.72 x 167.32 mm) suitable for integration into robotic platforms and industrial environments.
- Flexible Expansion: Includes PCIe x8 edge connector, M.2 slots (E, B, M keys), and optional TPM 2.0 and eSIM modules.
- Robust Environmental Tolerance: Operates in temperatures from 0°C to 60°C (without MXM) or 0°C to 55°C (with MXM), with humidity tolerance up to 90% non-condensing.
- Software Ecosystem: Bundled with ADLINK’s Neuron SDK, Ubuntu 18.04 LTS environment, and middleware support for ROS/ROS 2 and OpenVINO™.




3. Technical Specifications

| Parameter | Specification |
|----------------|------------------|
| Processor Options | Intel® Core™ i7-9700TE (1.8GHz, 12M cache, 35W TDP) / Intel® Core™ i5-8500T (2.1GHz, 9M cache, 35W TDP) / Intel® Core™ i3-8100T (3.1GHz, 6M cache, 35W TDP) |
| Chipset | Intel® Q370 Chipset |
| Memory | Dual-channel DDR4 SO-DIMM, non-ECC, 2400 MHz; up to 32 GB (RSM-57/55), up to 16 GB (RSM-53) |
| Display Output | 1x internal HDMI (vertical connector) |
| Ethernet | 4x Gigabit Ethernet (3x Intel® i210-AT, 1x Intel® i219-LM) |
| Serial Ports | 1x RS-232/422/485 pin header, 1x RS-232 pin header |
| USB Ports | 4x USB 3.1 Gen 1 Type-A, 2x USB 3.1 Gen 1 pin headers, 2x USB 2.0 pin headers |
| Digital I/O | 4x DI, 4x DO via 10-pin 2.0mm wafer (0.5A by BIOS selection) |
| M.2 Expansion | 1x M.2 E Key (1630/2230) for Wi-Fi/Bluetooth; 1x M.2 B Key (2242/2280) for SATA storage; 1x M.2 M Key (2242/2280) for SATA/PCIe x4 storage |
| PCIe Expansion | 1x PCIe x8 Gen2 edge connector (from 2x PCIe x4 root ports, up to 50W) |
| Power Connector | PCIe power connector up to 12V @ 3.5A |
| Optional Modules | eSIM, TPM 2.0 |
| Power Input | DC 12V ±5% via Molex DC-in jack |
| AC Adapter (Optional) | 240W AC/DC adapter (12V @ 20A) |
| Dimensions | 197.72 mm (W) x 167.32 mm (L) / 7.784” x 6.587” |
| Mounting | ADLINK proprietary mounting holes and CPU cooler bracket |
| Operating Temperature | 0°C to 60°C (without MXM), 0°C to 55°C (with MXM) |
| Storage Temperature | -40°C to 85°C |
| Operating Humidity | 10% to 90%, non-condensing |
| EMC Compliance | EN 55032 / EN 55024 |
| Software Environment | Ubuntu 18.04 LTS |
| SDK | ADLINK Neuron SDK |
| Middleware | ROS/ROS 2, Intel® OpenVINO™ |




4. Functionality

The ROScube-I Starter Kit functions as a compact, high-performance embedded computing platform optimized for robotics and AI workloads. It serves as the central processing unit for robotic systems, integrating sensor data, executing control algorithms, and managing communication between subsystems.

How It Works:
- The embedded Intel® Core™ processor executes real-time control and AI inference tasks.
- The optional MXM GPU module accelerates deep learning and computer vision workloads.
- ROS/ROS 2 middleware provides a modular framework for robot software development, enabling integration of perception, planning, and actuation modules.
- The Neuron SDK simplifies deployment of AI models and ROS nodes, while OpenVINO™ optimizes inference performance on Intel® hardware.

Use Cases and Applications:
- Autonomous mobile robots (AMRs) and service robots
- Industrial automation and collaborative robotics
- Machine vision and inspection systems
- AI-enabled edge computing and sensor fusion
- Research and prototyping for robotics and AI applications

Compatibility:
- Fully compatible with ROS and ROS 2 ecosystems
- Supports Ubuntu 18.04 LTS operating system
- Compatible with Intel® OpenVINO™ toolkit for AI acceleration
- Expandable via M.2 and PCIe interfaces for additional storage, wireless connectivity, or GPU modules




5. Components & Accessories

Included Components:
- ROScube-I embedded motherboard (RSM-57, RSM-55, or RSM-53 depending on configuration)
- Integrated Intel® Core™ processor
- Pre-installed DDR4 SO-DIMM memory (8 GB or 16 GB depending on model)
- Internal HDMI connector
- Power input interface (Molex DC-in jack)

Optional Accessories:
- Wireless Module: Intel® Wireless-AC 9260 M.2 2230 Dual-Band 2x2 Wi-Fi + Bluetooth 5 kit (P/N: 91-95266-0010)
- DDR4 SO-DIMM Memory:
- 8 GB DDR4 2400 MHz (P/N: 29-6B800-L430)
- 16 GB DDR4 2400 MHz (P/N: 29-6BC00-L430)
- MXM Graphics Modules: EGX-MXM-P1000, P2000, P3000, P5000 (Type A/B, up to 120W)
- AC/DC Power Adapter: 240W adapter with 8-pin ATX connector (P/N: 31-62164-0000-A0)

Connectors and Interfaces:
- HDMI, USB 3.1/2.0, Ethernet, RS-232/422/485, digital I/O, M.2 slots, PCIe x8 edge connector




6. Installation & Setup

Requirements:
- DC 12V ±5% power input or optional 240W AC/DC adapter
- Ubuntu 18.04 LTS operating system environment
- ROS/ROS 2 and Neuron SDK installation for development

Mounting Options:
- ADLINK proprietary mounting holes for secure integration into robotic chassis or enclosures
- Proprietary CPU cooler bracket for thermal management

Configuration Steps:
1. Install memory modules and optional MXM GPU.
2. Connect power supply via Molex DC-in or AC/DC adapter.
3. Attach peripherals (Ethernet, USB, HDMI, serial devices).
4. Install Ubuntu 18.04 LTS and configure ROS/ROS 2 environment.
5. Deploy Neuron SDK and OpenVINO™ for AI model optimization.
6. Calibrate sensors and actuators through ROS nodes.




7. Performance & Capabilities

Processing Performance:
- Up to 8-core Intel® Core™ i7-9700TE CPU with 12 MB cache and 35W TDP ensures high computational throughput for real-time robotics tasks.
- Dual-channel DDR4 memory architecture supports up to 32 GB for data-intensive applications.

Graphics and AI Acceleration:
- MXM GPU modules (up to 120W) deliver high-performance parallel processing for AI inference, image recognition, and SLAM (Simultaneous Localization and Mapping).
- Intel® OpenVINO™ toolkit enhances deep learning inference efficiency on both CPU and GPU.

Connectivity and Expansion:
- Four Gigabit Ethernet ports enable multi-sensor and multi-device networking.
- Multiple USB 3.1 ports support high-speed peripheral connections such as cameras and LiDAR sensors.
- M.2 and PCIe expansion slots allow flexible system customization.

Efficiency and Power:
- Optimized for 12V DC input with low power consumption (35W CPU TDP).
- Optional 240W power adapter supports additional GPU and peripheral loads.

Limitations:
- Operating temperature limited to 55°C when MXM GPU is installed.
- No integrated display output beyond internal HDMI connector.
- Requires external storage and wireless modules for full functionality.




8. Standards & Compliance

- EMC Compliance: EN 55032 (Electromagnetic Compatibility for Multimedia Equipment), EN 55024 (Immunity Requirements for Information Technology Equipment)
- Safety and Reliability: Designed for industrial-grade operation with wide temperature and humidity tolerance.
- Software Standards: Compatible with open-source ROS/ROS 2 frameworks and Ubuntu LTS distributions.
- Hardware Standards: Compliant with Intel® Q370 chipset specifications and MXM Type A/B GPU interface standards.
- Security Options: Optional TPM 2.0 module for hardware-based encryption and secure boot.




9. Additional Information

Warranty:
Standard ADLINK manufacturer warranty (duration subject to regional policy).

Technical Support:
ADLINK provides comprehensive technical documentation, SDK support, and developer resources for ROS/ROS 2 integration.

Maintenance:
- Regular firmware and BIOS updates available through ADLINK support portal.
- Software updates for Neuron SDK and OpenVINO™ ensure ongoing compatibility with new AI frameworks.
- Recommended periodic inspection of cooling system and connectors in high-duty environments.

Product Lifecycle and Updates:
- Specifications subject to change without notice (latest revision: May 12, 2020).
- Continuous software support for Ubuntu 18.04 LTS and ROS/ROS 2 distributions.

Manufacturer Contact:
ADLINK Technology, Inc.
Website: [www.adlinktech.com](https://www.adlinktech.com)




Summary

The ADLINK ROScube-I Starter Kit is a robust, scalable embedded computing platform purpose-built for robotics and AI applications. Combining Intel® Core™ processing power with optional MXM GPU acceleration, it delivers the computational performance required for advanced perception, planning, and control tasks. Its compatibility with ROS/ROS 2, extensive I/O options, and modular design make it ideal for developers and integrators building next-generation autonomous systems. With industrial-grade reliability, flexible expansion, and comprehensive software support, the ROScube-I Starter Kit provides a complete foundation for intelligent robotics development and deployment.

No accessories available.