NVIDIA Jetson Neuron Board
The Jetson Neuron Board (NVIDIA Neuron Board) from Mistral is a feature-rich, performance-oriented base board designed to meet the compelling needs of modern AI, Deep Learning and inference on the edge. Designed to mate with NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX modules, the Neuron Base acts as an NVIDIA Nano carrier board, NVIDIA Jetson TX2 NX carrier board or NVIDIA Xavier NX carrier board that enables all the features of the SoM for quick product development and benefit meet stringent time to market requirements.
The NVIDIA Neuron Board is available in two Variants – NB Basic and NB Turbo. The NB Basic brings out the required basic interfaces and connectivity features of the NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX such as Ethernet, CSI, USB, HDMI, M.2 NVMe Slot, mmWave RADAR etc. The NB Turbo is richer in terms of additional functional features and interfaces such as GMSL, FPD Link III, Wi-Fi 802.11ac, BT5.0 and 9-axis IMU Sensors.
Neuron Base Board
NVIDIA Nano Carrier Board
With its highly optimized form-factor, performance-rich I/Os and software support, Mistral’s Neuron Base is a NVIDIA Nano carrier board, NVIDIA TX2 NX carrier board and NVIDIA Xavier NX carrier board that offers a flexible and scalable platform to get products to market faster at reduced development cost. Based on a dual board architecture, the NVIDIA Nano carrier board can be mated with NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX SoMs over a compatible SODIMM socket.
The Neuron Base is designed to enable all the functional features of NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX, NVIDIA Jetson Xavier NX SoMs and aids faster product development and prototyping. It provides I/O accessibility and power to the NVIDIA Jetson Nano, TX2 NX and Xavier NX SoMs.
NVIDIA Nano carrier board is available in two variants namely, NB Basic and NB Turbo.
The NVIDIA Neuron Board addresses all crucial requirements that are key to design, develop and prototype embedded edge computing devices, which demand high-performance factors and runs complex AI workloads. The NVIDIA Nano carrier board is ideal for realizing AI-based machine vision applications such as Edge Camera with object detection and recognition, Human Activity Recognition, Robotics, Drones, Autonomous Navigation, Radar-Camera Sensor Fusion and Smart Retail among others.
The Neuron Base can be integrated with Mistral’s mmWave Radar Modules (60GHz Industrial Radar or 77GHz Automotive Radar) to design Camera and Radar Fusion Solutions ideal for Industrial and Automotive applications.
The Neuron Board is offered with NVIDIA JetPack SDK, which enables quick development of AI applications utilizing libraries supporting major AI frameworks, computer vision, graphics, multimedia, and more.
Radar and Camera
The Neuron Base can be mated with NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX SoMs over a compatible SODIMM socket.
The Neuron Base is designed to enable all the functional features of NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX, NVIDIA Jetson Xavier NX SoMs and aids faster product development and prototyping.
The NVIDIA Neuron Board is available in two variants.
* For Active Cooling, Fan control option is provided
The NVIDIA Nano carrier board / NVIDIA TX2 NX carrier board / NVIDIA Xavier carrier board for NVIDIA Jetson Nano, NVIDIA TX2 NX and NVIDIA Jetson Xavier NX is available in two variants namely NB Basic and NB Turbo.
The NB Basic provides all rudimentary interfaces and connectivity features of the NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX SOMs such as Ethernet, CSI, USB, HDMI, M.2 M, mmWave RADAR etc. needed for product development. The NB Turbo is a more sophisticated platform with more functional features and powerful multimedia interfaces such as GMSL, FPD Link III, Wi-Fi 802.11, Bluetooth and 9-axis IMU Sensors.
Based on customer requirements, either variant of NVIDIA Neuron Board can be ordered with Jetson SOM modules as mentioned below.
Variants and Pricing Info
Recommended Modules for Sensor Fusion
To realize Camera-Radar sensor fusion, the following modules can be integrated with NVIDIA Neuron Board.
mmWave RADAR Modules
Mistral’s mmWave RADAR Modules, based on Texas Instruments single-chip FMCW sensors, enable intelligent and autonomous decision making at the edge! The modules are available in 77GHz and 60GHz frequency bands and are ideal for low power, self-calibrating, ultra-accurate RADAR systems in industrial, Automotive and Smart City Applications.
Model: Mistral’s 60GHz (IWR6843) / 77GHz (AWR1843) RADAR Modules
Camera Module (2 lane CSI)
The camera module with CSI-2 MIPI 2-lane interface from Leopard Imaging is specially designed to mate with NVIDIA Jetson platforms.
Recommended Power Adapter for Neuron Base
Mistral recommend the following 24V/65W and 12V/60W AC/DC power supply adapters that support universal input (90~264 Vac).
Part No: 102-3417-ND [12V/5A Adapter]
Part Number: 102-3418-ND [24V/2.71A Adapter]
AI enabled Sensor Fusion Kit
The NVIDIA Jetson SoM based AI-enabled Sensor Fusion Kit (AI-SFK) from Mistral is an integrated, performance oriented, Camera Vision and mmWave RADAR fusion platform that meet the compelling needs of modern AI, Deep Learning, and Video Analytics applications.
The AI-SFK (Camera Vision + mmWave RADAR) is uniquely designed to provide range precision along with superior imaging capabilities for applications such as ADAS, autonomous vehicle, smart retail, industrial 4.0, robotics, smart building, and smart city among others. The camera & mmWave RADAR complement each other for object detection, classification, range, velocity and elevation parameters. The platform provides vision analytics and RADAR data fusion in real-time.
The Sensor Fusion Kit consists of:
- Mistral’s Neuron Base Board (NB-Turbo) mated with NVIDIA Jetson Xavier NX SoM
- Performs object detection, classification and annotates video stream with RADAR range data
- Mistral’s 77GHz mmWave RADAR Module
- Consists of FMCW transceiver with integrated PLL, Baseband ADC, R4F controller with accelerator, SPI flash, DSP, etc.
- Camera Module
- CSI-based 8 MP / 4K colour sensor supporting up to 21 fps
The platform delivers RADAR point cloud and video data:
- The RADAR point cloud data consists of object’s ID, range, angle, velocity and signal strength
- Camera stream inferenced using NVIDIA’s Jetson-inference module & Tensor-RT and annotated for different types of objects
- Mistral’s Sensor fusion demo algorithm co-relates and combines RADAR target with pre-trained objects (humans)
The SFK also supports NVIDIA JetPack SDK, which enables the development of AI applications with accelerated libraries supporting all major AI frameworks as well as computer vision, graphics and multimedia among others.
In addition to providing engineering services for product/system development, Mistral provides customization services around SFK for:
- Custom camera integration
- Other sensor integration like thermal, LiDAR etc
- Application software
- ML/DL algorithms development and integration
- Remote firmware upgrade
- Audio and video
- Cloud and mobile apps
- Power optimization
- Mechanical enclosure customization
- FCC/CE certification support
- Chirp Profile Tuning support
Radar and Camera
NVIDIA Jetson Xavier NX
- 384-Core NVIDIA Volta GPU with 48 Tensor Cores
- 6-Core NVIDIA Carmel ARM v8.2 64-bit CPU
- 2x NVDIA Engines
- 7-Way VLIW Vision Processor
- 8 GB 128-bit LPDDR4x
- 16 GB eMMC 5.1
- 10/100/1000 Base-T Ethernet
- CSI2-2 Lane Camera Interface
- Sony’s iMX219 from Leopard Imaging
- Mistral’s 77GHz AoPCB RADAR module based on TI’s AWR1843 SoC
- FMCW Transceiver with Integrated PLL, Transmitter, Receiver, Baseband and ADC
- 76 to 81GHz Coverage with 4GHz available Bandwidth
- Four Receive Channels and Three Transmit Channels
- C674x DSP for FMCW Signal Processing
- ARM Cortex R4FBased Radio Control System for Object Tracking and Classification
- M.2 M key based 250GB SSD
- Gigabit Ethernet
- Wi-Fi 802.11 a/b/g/n/ac
- BT 5.0
- 1 x CSI2 – 2 Lanes
- 1x FPD Link III
- 1x GMSL
- 2x USB 2.0
- 1x USB 3.1
- HDMI 2.0
- HDMI 2.0
- DMIC / I2S via Expansion Header
- 9-axis IMU Sensor
- Expansion Header
- I2C, I2S, GPIOs, Reset
- Cooling: Active cooling with Fan
- Power: 12V/5A to 24V/2.5A
- Operating Temperature Range: -25 C to 80 C
Sensor Fusion Demo
Object Detection With Ranging demo using Mistral’s AI-enabled Sensor Fusion Kit. The Sensor Fusion Kit integrated with people identification algorithm detects multiple people within a range of approx. 10 meters. The Kit can be integrated with a wide variety of object recognition algorithms to make the Neuron Kit suitable for ADAS, Traffic Management, Smart Building, Smart City and Industrial Automation applications.
Designed to mate with NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX modules, the NVIDIA Neuron Board from Mistral is a feature-rich, performance-oriented NVIDIA Nano carrier board designed to meet the compelling needs of modern AI, Deep Learning and inference on the edge. The Neuron Board supports Ubuntu 18.04 Linux OS. The board supports VisionWorks, a software development package provided by NVIDIA for computer vision and image processing. Mistral also provides TensorRT SDK for high-performance deep learning inference.
- Supports Ubuntu 18.04 Linux OS natively
In addition to the above-mentioned software, the AI-SFK board also supports:
- ROS1 Melodic
- RADAR / Camera Fusion
Designed to mate with NVIDIA Jetson Nano, NVIDIA Jetson TX2 NX and NVIDIA Jetson Xavier NX modules, the NVIDIA Nano carrier board and NVIDIA Xavier NX carrier boards from Mistral are a feature-rich, performance-oriented base board designed to meet the compelling needs of modern AI, Deep Learning and inference on the edge.