Dusun IoT offers custom single board computer based on Rockchip RK3588J SoC for AGV/AMR robot.
Automatic Guided Vehicles (AGV) and Autonomous Mobile Robots (AMR) are prevalent across industries, from automated warehousing and logistics in smart factories to hotel service robots and new energy charging AGVs. Utilizing AGV robots for automated transport is crucial for operational efficiency and cost reduction.
As AGV/AMR technology advances, the market demands higher performance and quality, including autonomous navigation, environment sensing, condition monitoring, and multi-robot networking. These demands place greater requirements on the AI performance, communication capabilities, and scalability of the control system.
The RK3588J Single Board Computer Solution features an Octa-core Cortex-A76/A55 CPU and a Mali-G610 GPU, offering high computational and graphical processing capabilities. This solution excels in AGV/AMR applications by providing powerful computing resources, flexible multimedia processing, extensive interfaces, and robust AI capabilities. It ensures efficient handling of complex tasks such as path planning, obstacle avoidance, and visual recognition. The hybrid multi-core AMP system architecture supports Linux, RTOS, and Bare-metal combinations, meeting the high real-time and stability demands of AGV applications.
This board is for reference only and can be customized according to needs.
The RK3588J Single Board Computer includes Rockchip’s self-developed tri-core NPU, capable of collaborative or independent operation to flexibly allocate computing power and prevent redundancy. With a combined compute capability up to 6 TOPS, it supports INT4/INT8/INT16/FP16 mixed computation, allowing seamless model conversion from frameworks like TensorFlow, MXNet, PyTorch, and Caffe. This enables AGV/AMRs Robot to execute complex AI algorithms such as image recognition, object detection, and speech recognition, thereby enhancing automation and intelligence.
Introducing INT4/ INT8/ INT16/ FP16 mixed operations, enabling network model based on frameworks such as TensorFlow/ MXNet/ PyTorch/ Caffe
The RK3588J AGV embedded Single Board Computer supports various interfaces, including USB, PCIe, and SATA, enabling seamless connection to sensors and peripheral devices. This includes vision navigation, inertial navigation, laser sensors, safety edges, emergency stop buttons, and audio-visual alarms. Critical peripherals like cameras, LiDARs, and radars enhance AGVs’ sensing and interaction capabilities. The HDMI interface supports external displays for visual operations and human-machine interaction.
The RK3588J Single Board Computer excels in multimedia processing:
Integrated GPU and VPU supporting 8K video decoding and 4K video encoding enhance AGV applications requiring video monitoring and real-time video analysis.Capable of real-time processing and transmission of high-definition video streams, supporting multiple video inputs and outputs, efficiently compressing and decompressing video to reduce storage and bandwidth needs.
To meet the multi-robot coordination control requirements of AGV robots, the RK3588J Single Board Computer solution applied by Dusun IoT in the AGV robot system control can realize on-board Gigabit Ethernet ports, Wi-Fi, LTE, LoRa modules, etc. It supports the private LoRa protocol, which is suitable for long-distance data acquisition and control in AGV applications, helping to achieve high-speed network transmission and control.
Built to industrial-grade standards, the RK3588J embedded mainboard incorporates refined power management to reduce energy consumption and a watchdog function for a stable operating environment.
Excellent heat dissipation allows for stable operation under wide temperature ranges and complex electromagnetic conditions, meeting AGV/AMR application needs across various scenarios.
Dusun provides comprehensive SDKs and BSPs for AGV/AMR applications, requiring only application development without the need for underlying driver migration. With extensive experience in AGV mainboard hardware design, Dusun offers design suggestions and check services, supporting the complete AGV embedded motherboard design and development from scratch. For local edge computing, Dusun provides various AI tools, including RKNN API, TensorFlow Lite, ONNX Runtime, and Android NNAPI.
