● ADLINK's EGX-MXM-A1000, EGX-MXM-A2000, and EGX-MXM-A4500 are the first modules to adopt NVIDIA's embedded GPU based on the NVIDIA Ampere architecture. ● ADLINK's MXM graphics modules provide high-performance GPU acceleration in a compact and power-efficient MXM form factor, bringing edge computing and embedded AI to many vertical markets such as healthcare, manufacturing, and transportation. ● Robust design that withstands harsh temperature environments, shock, vibration, and corrosion under severe conditions.

●ADLINK's compact and rugged AI-enabled AVA-RAGX is built on the latest industrial version of the NVIDIA Jetson AGX Xavier module with 32 TOPS, providing reliability and performance for railway applications in demanding and harsh operating environments. ●AVA-RAGX expands ADLINK's portfolio of EN 50155 compliant AVA platforms, offering railway solution providers the flexibility to choose the optimal configuration for digital transformation. ●ADLINK continues its "Elite" partnership with NVIDIA to leverage the power of infrastructure edge AI to build advanced railway solutions.

●ADLINK's proprietary 80-core Arm-based COM-HPC server-type module sets a new standard ratio for edge performance and power. With up to 80 ArmV8.2 64-bit cores, a maximum of 2.8GHz, and moderate power requirements at 175W TDP, users can reliably and predictably handle compute-intensive workloads. ●The COM-HPC Ampere Altra complies with Arm SystemReady SR, allowing for immediate support of many standard operating systems, hypervisors, and software. ●ADLINK has built a super-quiet water-cooled tower system called the "AVA Developer Platform," which comes with a default of 32 cores based on the COM-HPC Ampere Altra module and its carrier. The COM-HPC Ampere Altra module is also included in the "AVA-AP1," a high-durability vehicle prototyping platform co-developed by ADLINK, Arm, and Ampere, which features a default of 80 cores.

● The new portal enhances the remote management capabilities of ADLINK's deep learning acceleration platform, the "DLAP-211 series." ● Real-time system information and timely alerts enable immediate response from remote locations. ● Monitoring, alerts, and updates are enhanced by Allxon Device Management Solutions (Allxon DMS).

The ADLINK EVA SDK is an AI (artificial intelligence) edge visual analysis software platform that supports AI models optimized by TensorRT, OpenVINO, and ONNX Runtime, enabling AI developers, AI software companies, and enterprise system integrators (SI) to develop AI vision proofs of concept (PoC) in just two weeks. Since the software platform is pre-installed on ADLINK's AI vision hardware, it facilitates the development of edge AI vision applications with an intuitive GUI and ready-to-use plugins, allowing users to focus on training and integrating AI models. The EVA SDK is a one-API framework that supports open-source plugins for each stage of AI vision projects, including image capture and preprocessing, AI inference, post-processing, and analysis.

● The 3U VPX module is based on the 11th generation Intel Core i7 processor technology (formerly codenamed Tiger Lake-H), enhancing data and graphics performance as well as artificial intelligence (AI) acceleration capabilities. ● The design compliant with the Sensor Open System Architecture (SOSA) allows for easy reconfiguration and upgrades, providing a cost-effective, rapidly developed and deployed, non-proprietary, and open architecture for embedded computing capabilities. ● Based on the principles of the Module Open System Approach (MOSA), ADLINK actively collaborates with partners to develop custom-made COTS products based on open standards, enabling system integrators in the aerospace and defense sectors to bring products to market quickly.

●ADLINK's Express-TL COM Express Type 6 module is equipped with up to 8-core 11th generation Intel Core, Xeon W, and Celeron 6000 processors, as well as 128 GB of memory. ●The Express-TL is the first COM Express module to support PCI Express Gen 4x16, efficiently providing twice the bandwidth of traditional COM Express modules. ●The Express-TL is designed to operate in harsh temperatures ranging from -40°C to 85°C, making it suitable for continuous use in industrial environments for 24 hours a day over 10 years.

The ultra-compact edge DAQ solution "MCM-216/218" for voltage and current measurement functions as a standalone edge device that does not require a host PC. The MCM-216/218 is based on the ARM Cortex-A9 processor and features built-in 16 or 8 channels of 16-bit analog voltage or current inputs, making it suitable for 24/7 sensor measurement and condition monitoring IIoT applications. The MCM-216/218 supports distributed and remote data acquisition, edge analytics, monitoring control, and data acquisition (SCADA).

● The cPCI-A3525 complies with the latest PICMG CPCI-S.0 CompactPCI Serial Rev. 2.0 standard, providing enhanced performance. ● The robust and SWaP-optimized 3U module is ideal for next-generation mission-critical applications in railway transportation, aerospace and defense, and industrial automation. ● The addition of the cPCI-A3525 to ADLINK's extensive CompactPCI product line demonstrates the company's continued commitment to technology.

The CM5-P1000 is the first COTS PC/104 module to feature the NVIDIA Quadro P1000 GPU. The robust and optimal SWaP solution is ideal for aerospace, defense, and industrial applications. The addition of the CM5-P1000 to ADLINK's PC/104 products clearly demonstrates the company's unwavering commitment to technology.

● The AI smart camera integrates a compact all-in-one device designed for industrial AI inference with an NVIDIA Jetson Xavier NX module. ● The highly integrated format and simple cabling enable high-performance, compact, and low-bandwidth requirements for easily using AI in edge applications. ● With a pre-installed development environment, an intuitive GUI, and a wealth of plugin options are provided, accelerating proof of concept and reducing time to market.

● ADLINK's EGX-MXM-T1000, EGX-MXM-RTX3000, and EGX-MXM-RTX5000 are the first modules to adopt embedded GPUs based on NVIDIA's Turing architecture. ● ADLINK's embedded MXM graphics modules enhance edge computing and embedded AI, accelerating the computing needed for applications constrained by SWaP. ● Embedded graphics modules compliant with the MXM (Mobile PCI Express Module) standard.

●ADLINK offers a full line of EtherCAT solutions, including master and slave modules, at a cost-effective price. ●EtherCAT systems have been proven in test applications to reduce total cost of ownership (TCO) by up to 10% and increase productivity by up to 30%. ●The speed of development and deployment in machine automation and industrial automation, including glass inspection, assembly equipment, lens testing, automatic optical inspection, and dispensing, will be enhanced.

The SMARC AI on-module LEC-IMX8MP is equipped with NXP's i.MX 8M Plus, which integrates an NPU with a maximum of 2.3 TOPS. It combines NXP NPU, VPU, and GPU computing in a compact size, making it suitable for a wide range of AI-based applications such as industrial IoT, smart homes, and smart cities. ADLINK offers a ready-to-go IoT prototyping platform with the I-Pi SMARC IMX8M Plus, integrating production-grade components and software portability, along with flexibility and upgradability for proof of concepts (PoC) like Raspberry Pi.

ADLINK and Canonical have agreed that ADLINK will be able to provide customers with devices that have Ubuntu Linux, supported by Canonical, installed and licensed. By being certified, it is ensured that patches and the latest versions of Ubuntu Linux are reliably installed on ADLINK's devices, guaranteeing that the devices run the latest operating system and security revisions. Customers will receive a complete and thoroughly validated solution, allowing them to reduce time to market and ensure the security of their IoT solutions.

●By using the MECS-6110 edge server, it becomes easier to develop network function virtualization (NFV) that executes network functions via software on white box uCPE (universal customer premises equipment) by extracting network resources from dedicated hardware. ●The MECS-6110 realizes a universal edge platform that can efficiently provide multiple network services and security features, such as software-defined wide area network (SD-WAN) and next-generation firewall (NGFW), to enterprises of all sizes. ●ADLINK's MECS-6110 edge server offers an open platform certified for Intel Select Solutions for uCPE, enabling communication service providers (CommSP) to quickly enter the market through rigorous pre-deployment testing that verifies system performance.

The ADLINK DLAP x86 series is the most compact GPU-equipped deep learning acceleration platform among commercially available products, enhancing the performance of AI applications in industrial, manufacturing, and healthcare environments. The DLAP x86 series is designed to strengthen compute-intensive AI inference and learning tasks, optimizing SWaP and AI performance at the edge while providing excellent power efficiency and cost performance. Additionally, with a heterogeneous architecture that integrates high-performance CPUs and GPUs in a compact size, along with a design that excels in thermal efficiency, it is ideal for AI applications that require advanced computational performance.

- ADLINK's latest ASM series offers sharp, high-precision image quality and excellent reproduction of high-definition images in color and grayscale, promoting rapid image-guided therapy. - The adoption of a robust and hygienic design that allows for quicker and easier cleaning reduces the time required for disinfection in operating rooms and emergency care environments. - ADLINK's surgical monitors are certified as Class I medical devices based on the European MDD to ensure the safety of patients and operators and facilitate system-level integration.

The latest PXES-2314T from ADLINK is a 4-slot Thunderbolt 3 PXI Express chassis that can be deployed for a wide range of inspection and measurement applications. The compact system structure enhances portability, allowing mobile testing to be performed without sacrificing system bandwidth. Although it has become a low-cost and convenient form factor, the high performance and flexibility of PXI Express remain intact.

The ADLINK MLC-AL/MLC-KL series features a hygienic design that is convenient for cleaning and sterilization, reducing the risk of infection for patients and frontline healthcare workers. The MLC-AL and MLC-KL support a wide range of systems, including mobile X-ray, ventilators, blood and tissue analysis, and vital sign monitoring, helping to streamline patient care. They are equipped with 13.3-inch and 15.6-inch full HD displays that support advanced image reproduction with PCAP multi-touch functionality in a compact housing.

ADLINK's NEON-1000-MDX AI Smart Camera series has been recognized for demonstrating innovation in IoT (Internet of Things) products and technologies by winning the 3rd IoT Innovator Award from Compass Intelligence. The NEON-1000-MDX AI Smart Camera integrates ADLINK Edge vision analysis software for industrial machine vision, making it easier to implement AI machine vision applications in new or existing environments. This award is the sixth award received by ADLINK's machine vision edge AI products since 2020. ADLINK Technology has won the 2020 IoT Innovator Award from Compass Intelligence for its NEON-1000-MDX AI Smart Camera series in the field of machine vision technology. The annual 3rd IoT Innovator Award from Compass Intelligence is given to companies that demonstrate outstanding innovations that impact the market and contribute to the advancement of IoT.

ADLINK Technology has launched the first octa-core COM Express Type 6 compact module equipped with the new AMD Ryzen V2000 APU, which integrates Radeon graphics (cExpress-AR). With double the core count, the cExpress-AR is ideal for demanding graphic-based applications such as: • Ultrasound image processing • 4K high-speed video encoding and streaming • Embedded gaming and infotainment • AI inference at the edge "The AMD Ryzen V2000 APU features high-performance graphics computing cores, allowing it to be used in AI inference applications without the need for an additional GPU. This is one way the cExpress-AR contributes to overall system cost reduction," commented Alex Wang, ADLINK's COM Express product manager.

ADLINK Technology has announced the latest NEON-1000-MDX smart camera series along with the Smart Pallet solution. This series integrates the Intel Movidius Myriad X VPU, which supports an interface model optimized for OpenVINOTM, and ADLINK's Edge Vision software that allows for the rapid and easy addition of AI-based machine vision to existing environments. With this all-in-one solution, machine vision engineers and developers can not only quickly build, test, and deploy machine learning (ML) models but also have tools to connect, stream, and automate machine vision tasks for IoT solution teams and system integrators.

ADLINK Technology has released the latest AI-enabled smart camera series, NEON-2000-JT2, which facilitates the deployment of machine vision in AI-based factory automation. This new series of smart cameras, equipped with NVIDIA's Jetson TX2 AI module, features an all-in-one design and a compact footprint, with a vision software set pre-installed, reducing the complexity of developing and integrating AI machine vision. With the NEON-2000-JT2, production efficiency can be improved in labor-intensive manufacturing industries where quality assurance and inspection are essential, such as food and beverage, consumer packaged goods (CPG), and agriculture. One customer has implemented ADLINK's AI machine vision solution for product inspection, achieving over 95% accuracy while inspecting more than 50 times the number of products compared to before. ADLINK believes that customers can further expand their achievements with the NEON-2000-JT2 series.

ADLINK Technology Inc., a global leader in edge computing aiming to contribute to society and industry by connecting people, places, and things through AI, has led the development of Eclipse Cyclone DDS to become a Tier 1 ROS 2 middleware in collaboration with the Eclipse Foundation IoT community, the OpenADx working group, and the ROS 2 Technical Steering Community. With the latest release of ROS 2, Foxy Fitzroy (foxy), the development of robotics and autonomous vehicles has become easier, faster, and more reliable and secure. Eclipse Cyclone DDS is the next-generation implementation of the OMG (Object Management Group) DDS (Data Distribution Service) standard, reflecting ADLINK's experience in innovative development work from the outset.

ADLINK Technology has announced the ROScube-X series of real-time ROS 2 robot controllers for advanced robotics applications. The ROScube-X series from ADLINK features the NVIDIA Jetson AGX Xavier module, which includes a variety of interfaces such as an integrated Volta GPU, dual deep learning accelerators, and a GMSL2 camera connector that can integrate cutting-edge robotics systems. It is also compatible with the fully equipped resources provided by the NVIDIA JetPack SDK, making it ideal for robotics applications that require advanced AI computing performance with minimal power consumption.

The ADLINK ROScube-I series is a ROS 2-compatible robot controller that features Intel Xeon E, 9th generation Intel Core i7/i3, and 8th generation Intel Core i5 processors. It is equipped with advanced I/O connectivity that supports a wide range of sensors and actuators, catering to the diverse needs of robotic applications. The ROScube-I also supports an expansion box that includes the "Intel Distribution of OpenVINO" toolkit, which is useful for enhancing functionality and performance, along with an Intel VPU card for computing AI algorithms and interfaces. Robot systems based on the ROScube-I are supported by ADLINK's Neuron SDK, a platform specifically designed for specialized robotic applications such as autonomous mobile robots (AMR).

ADLINK Technology today announced the release of the next-generation Matrix embedded computer, designed to enhance the responsiveness and durability of edge AI and computing applications. The next-generation Matrix series, engineered to boost performance, offers innovation and robustness to enable data-driven decision-making at the edge, achieving improvements in efficiency, productivity, and security across the industry. The next-generation Matrix series elevates system responsiveness to the next level with its heterogeneous computing architecture. The new Matrix series combines CPU computing based on the 9th generation Intel Xeon and Core processors with GPU computing from the NVIDIA Quadro Embedded GPU, accelerating compute-intensive high-resolution medical imaging and enabling faster and more accurate diagnoses.

PICMG (PCI Industrial Computer Manufacturers Group) is in the final stages of developing the latest COM-HPC computer-on-module specification to meet the needs of the highly advanced industrial edge computing market. The COM-HPC specification is intended not to replace COM Express but to expand upon its success, with a release planned by the second quarter of 2020. Edge computing is a rapidly growing field that demands unprecedented levels of performance. The system memory requirements for current and future high-end embedded multicore processors are increasing very sharply. As memory requirements approach 128 GB or 256 GB, current common form factors like COM Express can no longer provide a cost-effective solution. This is why COM HPC is needed. Five sizes have been defined that can be used as server modules capable of accommodating up to eight DIMM sockets onboard.

ADLINK Technology, one of the main contributors to the SMARC SGET committee, has announced the release of the SMARC module revision 2.1. The SGET (Standardization Group for Embedded Technologies) has just introduced the latest future-proof specification 2.1. The new features are only additional items that do not affect the multiplexing of existing signals and the pins of the edge connector assigned in the 2.0 specification, making the specification fully backward compatible with the current specification 2.0. The main additions are as follows: - Anticipating the needs of the AI and robotics markets, support for up to four MIPI CSI ports - Expansion of Ethernet ports through multiplexing of SERDES signals on the third and fourth PCIe x1 interfaces - Various minor additions for initially unused pins such as plus 2 GPIO pins and PCIe clock request signals