0731-85311818
Equipped with the all-new Jetson AGX Orin series Fully support mainstream deep learning frameworks and algorithms、 Cloud edge collaborative reasoning, 8-channel high-definition cameras, multi-channel high-speed data transmission, and rich communication interfaces
Equipped with NVIDIA's all-new Jetson AGX Orin series, the 2048 core NVIDIA Ampere architecture can be configured with up to 64 Tensor Cores GPU, The maximum frequency is 1.3GHz. Up to 12 core ARM can be used Cortex-A78AE v8.2 64 bit CPU, maximum frequency of 2.2GHz, 3MB L2+6MB L3. Embedded with 2 NVDLA engines, providing 275 TOPS high-performance computing power, with performance better than Jetson AGX Xavier has increased by 5 times, enabling autonomous driving functions such as sensor fusion, environmental perception, image fusion, path planning, and decision control
Support TensorFlow, PyTorch, and Caffe/ All popular AI frameworks such as Caffe2, Keras, and MXNet provide platform optimized deep learning inference frameworks and tools, making it easier for developers to deploy various deep learning models. By implementing powerful algorithm functions such as image recognition, object detection and localization, pose estimation, semantic segmentation, video enhancement, and intelligent analysis, they can meet high computing needs such as processing large amounts of sensor data, running deep learning algorithms, real-time environment perception, path planning, and decision control
On the basis of massive front-end data, super powerful computing power, and advanced algorithm combination, wireless networks can be leveraged to achieve collaborative integration and innovation of all factor resources, building an autonomous closed-loop cloud edge collaborative reasoning framework for front-end collection and execution, real-time inference at the edge, and back-end training and upgrading, realizing intelligence in the cloud and intelligence on the end
Supporting the fusion perception of multiple high-definition cameras and LiDAR, comprehensively, in real-time, sensitively, and accurately controlling changes in the surrounding environment, providing key references for intelligent driving path planning and decision-making control, laying the foundation for the implementation of SV3D, NoA, and other functions, and ensuring the safe operation of intelligent devices such as robots and unmanned vehicles in various environments
Support access to 8-way GSML2 high-definition cameras to capture high-resolution images of the vehicle's surrounding environment, which are used to identify road signs, lane lines, traffic signals, pedestrians, cyclists, other vehicles and other targets, and provide real-time visual information for the auto drive system
Built in dual layer USB 3.0, HDMI, RS232, RS485, CAN, GPIO, Gigabit Ethernet, microphone and other device interfaces, the device interfaces are highly integrated to meet the needs of various external devices and sensor connections, comprehensively improving the scalability of the platform
| Basic Parameters | |||
|---|---|---|---|
| Product Name | EX-03 Intelligent driving domain controller | Product Dimensions(WxDxH) | 197x174x52mm |
| Operating Temperature | -20-50℃ | Operating Humidity | 95%@40℃ (non-condensing) |
| Processor parameters | |||
| NVIDIA RION Series | AGX Orin 64GB AGX Orin 32GB |
||
| CPU | 12-core NVIDIA Arm" Cortex A78AE v8.2 8-core NVIDIA Arm" Cortex A78AE v8.2 |
||
| GPU |
2048-core NVIDIA Ampere GPU with 64 Tensor Cores 1792-core NVIDIA Ampere GPU with 64 Tensor Cores |
||
| Memory | 64GB 256-bit 32GB 256-bit |
||
| Maximum AI reasoning ability | 275TOPs 200TOPs |
||
| I/O | |||
| LAN | 2XLAN(1000Mbps) | GPIO | 1xGPIO |
| USB3.0 | 4xUSB3.0 | AUDIO | 1xAUDIO |
| HDMI | 1xHDMI | TYPE-C | 1xTYPE-C |
| MIC | 1xMIC | GSML2.0 | 8xGSML2.0 |
| DC IN | 1xDC IN(12V) | ||
| Expansion Slots | |||
| WIFI | 1xWIFI | ||
| Operating System | |||
| Linux | Ubuntu 22.04 LTS with JetPack Above5.1 | ||