A popular convenience store recruited AAEON to help them create an AI inference-based automated self-checkout that could reduce the time customers spent queuing and deliver a more convenient and streamlined way to shop.
How the BOXER-8642AI Revolutionized Smart Retail
(1) Intel RealSense D405 3D Depth Camera Support
■ Intel 3D RealSense D405 x 6
■ 10Gbps USB 3.2 Gen 2 x 8
■ Frame Drop Rate > 0.1%
(2) World-Class Image Classification
■ NVIDIA Jetson AGX Orin (275 TOPs)
■ Precise 3D Object Recognition
■ Real-Time, Error-Free Data
(3) Retail-Ready Hardware
■ Fanless & Compact Design
■ -25°C to 55°C
■ 12V-24V for Power Supply Fluctuation Protection

With sprawling areas dedicated to shopping, restaurants, leisure lounges, and even accommodation, the passenger experience in airports has greatly improved. With these developments however, come barriers to customers with mobility issues. To remedy this, one company contacted AAEON in search of a product suitable for their autonomous passenger transport vehicle, finding the BOXER-8645AI to be the perfect fit!
How AAEON’s BOXER-8645AI Drove the Project’s Success:
(1) Equipped with eight FAKRA connectors for GMSL2 camera integration, the BOXER-8645AI provided real-time high-resolution image data transmission.
(2) The system’s DB-9 ports offered isolated CANBus and RS-232/422/485 signals interfaces for motor control and sensor integration for collision avoidance.
(3) The BOXER-8645AI provided up to 275 TOPs of AI performance through its NVIDIA® Jetson AGX Orin™ module, providing the vehicle with the tools needed to navigate its surroundings.
(4) MIL-STD-810G standard tolerance to both vibration and shock ensured the application remained operational within such a dynamic setting, ensuring longevity of use.

Thanks to the PICO-RAP4, AAEON’s client was able to develop a fully automated service robot that could be deployed in various settings across industries ranging from retail to agriculture.To run the AI inference models required to act autonomously, the robot relied on the PICO-RAP4's embedded Intel® Core™ i7-1370PE CPU, which offered 6 performance-cores, 8 efficient-cores and 20 threads to provide extremely high processing power and CPU frequencies capable of handling demanding AI tasks. It also maintained power efficiency through its efficient cores for background operations. While not the primary source of the robot’s AI functionality, the PICO-RAP4's CPU also incorporated several advanced technologies, including Intel® Deep Learning Boost, Intel® Gaussian & Neural Accelerator, and Intel® AVX2, which were instrumental in accelerating AI workloads and enhancing machine learning performance.
To manage the majority of the application’s AI workload, the PICO-RAP4's M.2 2280 M-Key slot, which supported a PCIe 4.0 x4 interface, was equipped with a high-spec AI acceleration module. The PICO-RAP4’s high-bandwidth PCIe interface allowed the AI acceleration module to process the client’s diverse image recognition, natural language processing, and decision-making algorithms without any latency.

For RoboCup 2022 and 2023, held in Portugal and France respectively, VDL Robot Sports made use of AAEON’s BOXER-8251AI, featuring the NVIDIA® Jetson Xavier™ NX. For 2024’s competition, held in the Netherlands, VDL Robot Sports decided to upgrade their platform to capitalize on the innovations AAEON’s AI@Edge Solutions range had made during the intervening period, selecting the BOXER-8651AI as the system to help elevate their team's performance to the next level.
When it came to meeting the expectations of VDL Robot Sports’ autonomous robotic system, AAEON's BOXER-8651AI hit the back of the net. Featuring the NVIDIA® Jetson Orin NX, the BOXER-8651AI provided was on target, with the integrated NVIDIA Ampere GPU featuring 1024 CUDA® and 32 Tensor Cores offering exceptional inferencing capabilities. Moreover, the system supported the NVIDIA JetPack™ software development kit, allowing the engineers at VDL Robot Sports to optimize the performance of their AI models, including the use of TensorRT for high-performance deep learning inference and CUDA for parallel computing.

A client was seeking a drone solution capable of leveraging AI inferencing and high-resolution imaging to provide real-time data on crop health, diseases, and environmental conditions across extensive agricultural landscapes. After careful consideration, the client selected AAEON's de next-V2K8.
Given the complexity of the project, the de next-V2K8’s extremely small 86mm x 55mm form factor was a very attractive feature, allowing the client to minimize the overall load on the drone, enabling better flight dynamics and extending flight time. In addition, the minimal space taken up by the board meant the client was able to allocate more of their payload capacity to sensors, cameras, and other necessary equipment.
Powered by the AMD Ryzen™ Embedded V2718 with Radeon™ Graphics, the de next-V2K8 leveraged 8 cores and 16 threads of processing power. This level of performance, along with the board’s 16GB of onboard LPDDR4x system memory, enabled fast high-resolution image processing. Its integrated Radeon™ Graphics also meant the board supported enhanced data visualization, adding to its capacity to handle large visual datasets from peripheral cameras.

Resilient in-vehicle security systems that anticipate threats, rather than merely withstand them have evolved into a specialized sector within the security industry. These systems are crucial in guaranteeing the safety and security of assets and personnel, especially in high-risk environments.
The BOXER-8658AI's unique selection of interfaces was a major factor in its selection for the project. With the PC able to support eight PoE cameras via RJ-45 ports, the client’s security system could maintain comprehensive surveillance of the vehicle’s surroundings. This multiple PoE interface allowed for simultaneous power and data transmission, simplifying deployment and eliminating the need for a separate power source when installing cameras for the application. Additionally, the inclusion of multiple USB 3.2 Gen 2 ports and serial ports (DB-9 and DB-15) facilitated communication with various vehicle systems, sensors, and industrial equipment, enhancing operational versatility and integration capabilities.
Given the project’s narrow specifications, the BOXER-8658AI was the obvious choice for the client’s in-vehicle security system. As the factor without which the project would be a non-starter, the product’s environmental resilience was a key factor. Designed to meet MIL-STD-810G standards for shock and vibration resistance, the BOXER-8658AI was guaranteed to work in harsh environments on the road. Furthermore, the solution’s broad -15°C to 60°C temperature range guaranteed consistent performance across varying climates.

With their proprietary AI model, AAEON’s client aimed to analyze customer behaviors in department stores to understand the preferences of visitors. Using this data, they would then be able to craft marketing strategies that resonated with similar visitors, converting passersby into customers. This company chose AAEON’s UP Squared Pro 710H Edge as the embedded system responsible for implementing their AI algorithm, collecting behavioral data, running its inference model, and optimizing in-store digital advertising to appeal to visitors in real-time.
Armed with one integrated Hailo-8™ edge AI processor, along with a second Hailo-8™ edge AI acceleration module installed via its M.2 2280 M-Key slot, the client’s model was optimized and translated to the Hailo-8 format using the Hailo-8™ Dataflow Compiler, a software toolset provided as part of Hailo’s AI software suite. With two Hailo-8™ AI acceleration modules, the device had a total of 52 TOPS of inferencing power at its disposal, allowing the client’s algorithm to analyze customer behaviors and identify demographic trends and preferences from the visual data collected.

A company was tasked with providing 500 smart parking meters to a city in Taiwan, and chose AAEON’s BOXER-8221AI Compact Fanless Embedded BOX PC to power the application. The smart parking meter’s mechanics were comprised of two street-level cameras, mounted to bollard fixtures adjacent to each roadside parking space. AAEON’s BOXER-8221AI, featuring the NVIDIA® Jetson Nano™, enabled the smart parking meters to determine the duration that a car had parked in a space using real-time license plate recognition, and automatically issue tickets to drivers via text, with the help of the company’s software, which calculated the fee with their own algorithm. With NVIDIA’s Maxwell™ GPU architecture allowing processing speeds up to 472 GFLOPs, the smart parking meters could process several high-resolution images at the same time, ensuring the accuracy of the parking data, while analyzing parking patterns through its AI learning capabilities.

ComBox Technology and LARGA are two software developers and systems integrators who have partnered together to provide a solution for their car share customers. Their AI inference is capable of accurately determining if a driver is smoking by using visual data to not only look for smoke, but also analyze driver behavior and the position of the driver’s hands. The system can also recognize if the driver is holding an object such as a cellphone or cigarette, providing even more accurate and flexible analysis. To provide their customers with an effective package solution, ComBox and LARGA turned to AAEON and the VPC-3350S mobile NVR with AI Core X.

The customer chose the BOXER-6646-ADP from AAEON’s Box PC range as the system responsible for powering their eGate application. With dimensions of just 220mm x 154mm x 61.6mm and a fanless design, the customer was able to integrate the BOXER-6646-ADP into the limited deployment space without any issues. Despite this, the system’s support for both 12th and 13th Generation Intel® Core™ Processors meant that it had more than enough computing capacity to handle high-intensity workloads while maintaining power-efficiency.
When it came to peripheral device support, the BOXER-6646-ADP also delivered. The system was able to accommodate both the primary and secondary display screen requirements of the application through its two HDMI ports. Additionally, two of the system’s three USB 3.2 Gen 2 ports were utilized to install the cameras necessary for capturing passenger data. The BOXER-6646-ADP’s third USB 3.2 Gen 2 port was occupied by an RFID reader. The RFID reader’s purpose was to scan passenger passports to acquire the corresponding information for the customer’s software to compare with the data obtained by the system’s USB cameras.

The proposed autonomous ships were intended for a range of purposes, including seabed mapping, carrying out routine maritime patrols, and to aid in search and rescue operations. With a strong need for a robust and compact computing solution equipped with high processing power (11th Gen Intel), expandable storage, and the ability to seamlessly integrate LIDAR, SONAR, and GPS functions, AAEON's BOXER-6643-TGU proved to be the perfect fit.
The challenges of navigating the vast and unpredictable open sea, including harsh environmental conditions and limited installation space, were met and exceeded with the versatility and resilience of this AAEON’s expertly designed rugged system. AAEON's BOXER-6643-TGU proved to be the linchpin in the success of this innovative maritime project. With its unwavering performance in the face of challenging conditions, seamless sensor integration, powerful data processing, and secure data storage, it stands as a testament to AAEON's commitment to delivering cutting-edge technology solutions that empower customers to excel in their most demanding endeavors.

A renowned industry-leading manufacturer of versatile camera platforms was seeking a solution to power their high-specification, medium-to-long-range rugged surveillance system. The end product was primarily targeted towards the security sector, where it would find application in various settings, such as highway traffic monitoring, border surveillance, and perimeter security.
Having determined that it fulfilled all the requisite suitability criteria, the company made the decision to incorporate AAEON's PICO-TGU4 embedded Pico-ITX Board. This choice was driven by the major advantages presented by this model, including its robust embedded 11th Generation Intel® Core™ i5 processor and soldered LPDDR4x system memory. Additionally, the board offered a degree of versatility for supplementary wired connectivity through its I/O and wireless communication protocols via its expansion slots.

A large power station in the United States sought to deploy a robotic solution to undertake its equipment testing and maintenance needs. Due to the station’s large floorspace and variety of equipment, a human solution would require employees to travel large distances and perform analysis of complex equipment in tight, potentially hazardous spaces. Therefore, the company felt that a compact, four-legged robotic unit would be the perfect fit.
Equipped with an AMD Ryzen™ V2000 processor, LPDDR4x system memory, and up to 128GB of NVMe storage onboard, the PICO-V2K4 had all core facets required for deployment integrated, comprehensively meeting the needs of the application without additional components. Further, the board contained a wealth of connectors to provide the robot with peripheral device function, such as two USB 3.2 Gen 2 running at 10Gbps for infrared cameras, four COM ports for LIDAR sensor support, and four simultaneous displays courtesy of HDMI, DP via USB Type-C, and either LVDS or eDP interfaces. The integrity of the defect information returned to the station’s data center was also maintained via onboard TPM 2.0, preventing information security breaches.

In the MAXER-2100, the vendor found the ideal platform for their software, running their inferencing model to rapidly identify defects in PCB components, but the crucial thing that the MAXER-2100 brought to the quality inspection ecosystem was its use of using deep learning to refine the client’s inferencing model as it received and processed a greater volume of data, resulting in a faster and more accurate quality inspection process overall.
The vendor (Hawkeye solution) acquired images of components, solder joints, traces, and other features of the PCBs via an AOI device. Once captured, these images were transmitted to the MAXER-2100, powered by 13th Generation Intel® Core™ LGA 1700 socket-type processors. With NVIDIA® GPU cards installed via PCIe slots, the MAXER-2100 was able to run the vendor’s AI inferencing model to determine the pass/fail status of each component image.

A high-quality wooden furniture manufacturing company opted for a single board computer (SBC) over larger system-level alternatives, envisioning a more discreet integration into their existing factory infrastructure. After careful consideration and evaluation, the furniture manufacturer concluded that AAEON's PICO-ADN4 was the perfect fit for the stringent requirements of their project. This decision highlighted a preference for a streamlined, cost-effective solution that could seamlessly integrate into their existing processes while delivering the automation capabilities essential for optimizing their assembly line.

Artificial Intelligence has been central to Industry 4.0 from the very beginning, and the manufacturing sector is increasingly looking for safer, more efficient, and cost-effective means of maintaining production. Car manufacturing, more than most, has always been open to the innovation and improvement offered by AI solutions. An example of this is the automation of sheet metal stamping in the manufacturing process, whereby sheet metal made up of the raw metal materials is stamped into the manufacturers desired shape.
Automated guided vehicles (AGVs) look set to be the next iteration of this innovation, with a leading car manufacturer recently incorporating the BOXER-6641 Fanless Embedded Box PC in their manufacturing process. The relationship between AGVs and car manufacturing appears to be a natural fit, with AGVs able to transport a greater volume of material to assist in the production process when compared to a human being or a forklift.

A global pioneer in vacuum technology planned to upgrade its industry-leading leak detector, which is renowned for being one of the fastest and most effective tools for component leak testing across industries such as automotive, semiconductor, and HVAC. Helium is commonly used as a tracer gas for such testing due to its inert nature and lack of reactivity with other elements. The product in question is a flexible, portable standalone system that allows for both integral testing, used to determine component leak-tightness, and sniffer testing, which pinpoints leak locations.
Alongside the testing unit, the company required a remote controller that could provide a user-friendly interface, operate under various conditions, and interact with the data from the testing unit. After careful consideration, they identified AAEON’s RTC-710RK 7" Rugged Tablet as the perfect solution.

A compact fanless body, powerful inferencing capabilities, and an FCC Class B classification make the BOXER-8641AI-Plus the premier smart healthcare solution. An advanced I/O featuring 10G LAN facilitates high-speed, high-bandwidth data transmission for peripheral devices, while onboard TPM 2.0 maintains robust security for data in transit. Designed for seamless integration with existing equipment setups, AAEON’s newest embedded AI system brings out-of-band (OOB) management and over-the-air (OTA) update functionality to the BOXER series, making it easier than ever to maintain, enhance, and update your software remotely.

A leader in the European "in vitro diagnostics (IVD)" market wanted to upgrade their multiplex testing machines, used to perform a wide array of blood analysis functions such as immunoassay, clinical chemistry, and molecular testing. However, there were several key attributes the company felt the single-board-computer used to power such sophisticated machinery needed to have.
The GENE-ADP6, equipped with 12th Generation Intel® Core™ processors (formerly Alder Lake), was installed as part of a larger mechanical unit, in which blood samples were placed and analyzed on-board using chemiluminescence as a measurement principle for disease markers. CCD cameras connected to the GENE-ADP6 via its four USB 2.0 ports were used to convey the information captured from light emissions and provide an output of the results. Using this method, multiple tests could be completed using a smaller blood sample, dramatically increasing the efficiency of the blood analysis process.