All News | Boards | Chips | Devices | Software | LinuxDevices.com Archive | About | Contact | Subscribe
Follow LinuxGizmos:
Twitter Facebook Pinterest RSS feed
*   get email updates   *

Jetson Nano based robotics kit connects to SparkFun sensors

Aug 16, 2019 — by Eric Brown — 826 views


SparkFun has launched a version of Nvidia’s Jetson Nano based JetBot AI Kit robotics kit equipped with its Qwiic pHat, a wide-angle camera, and more. Also new: a SparkFun DLI Kit for Jetson Nano to use in conjunction with an Nvidia AI course.

When Nvidia launched its Linux-powered Jetson Nano module and $99 Jetson Nano Development Kit in March, it posted specs and instructions on GitHub for using the kit to build out a mobile JetBot robot. In late July, Waveshare launched a JetBot AI Kit based on the design available with ($216) or without ($100) the Nano Dev Kit. Now SparkFun has released a more advanced — and expensive — version of the open source robotics kit selling for $275 with the Nano Dev Kit or $175 without. SparkFun also launched a SparkFun DLI Kit for Jetson Nano focused on deep learning (see farther below).



SparkFun JetBot AI Kit, fully assembled
(click images to enlarge)

The SparkFun JetBot AI Kit follows other SparkFun robotics kits such as the Raspberry Pi Zero W based SparkFun Autonomous Kit for Sphero RVR. The new kit “includes everything you need to get started with JetBot minus a Phillips head screwdriver, a basic soldering setup, and an Ubuntu desktop GUI,” says SparkFun.

The JetBot taps the Jetson Nano’s 128-core Maxwell GPU along with ROS (Robot Operating System) nodes code to run facial recognition, object tracking, auto line following, and collision advance algorithms. The Nano’s Maxwell GPU provides CUDA-X AI libraries and [email protected] support. The 70 x 45mm Nano module also integrates 4x 1.43GHz Cortex-A57 cores and 4GB LPDDR4.

The 100 x 80mm Jetson Nano Dev Kit is similar to a Raspberry Pi 3B+. It offers a GbE port with PoE support, a microSD slot, 4x USB 3.0 ports, and a micro-USB device port with power support. The SBC is further equipped with 4K-ready HDMI 2.0 and eDP ports, a MIPI-CSI input, a 40-pin GPIO header, and an M.2 slot that holds the Intel wireless card available on the Jetson AI Kit. (For more details on the Nano Dev Kit, see our Jetson Nano report.)

Like the Waveshare kit, the SparkFun JetBot AI Kit ships with a 64GB microSD card with a pre-flashed SparkFun JetBot image that includes the JetBot ROS software. It also adds AWS RoboMaker with AWS IoT Greengrass.

The kit is the only JetBot kit currently on the market “that moves beyond the standard JetBot examples,” claims SparkFun. “Many other popular AI frameworks like TensorFlow, PyTorch, Caffe and MXNet are supported, and Jetson Nano is capable of running multiple neural networks in parallel to process data and drive action,” says the company.



SparkFun JetBot AI Kit parts (left) and SparkFun Qwiic pHat for Raspberry Pi
(click images to enlarge)

Another key addition is the SparkFun Qwiic pHat for Raspberry Pi. Backed up by the included Jetson GPIO Python library, which offers “some new python compatibility with the SparkFun Qwiic ecosystem,” the partial HAT offers 4x connectors that can plug in dozens of sensors and other gizmos within the Jetson Nano environment, says SparkFun. The pHAT “makes it easy to integrate more than 30 daisy-chainable sensors without soldering.”

Other kit features include a Leopard Imaging 145-degree FOV wide-angle camera with ribbon cable, and EDIMAX WiFi adapter, and a SparkFun Serial Controlled Motor Driver. There’s also a SparkFun Micro OLED Breakout from the Qwiic family.

 
SparkFun DLI Kit for Jetson Nano

The SparkFun DLI Kit for Jetson Nano is designed for developers, researchers, students, and hobbyists who want to explore AI concepts. The kit ships with a Jetson Nano Dev Kit and a 32GB microSD card with content that supports Nvidia’s free, self-paced, online Deep Learning Institute (DLI) course, Getting Started on AI with Jetson Nano. The course teaches how to collect image data and use it to train, optimize, and deploy AI models for custom tasks like recognizing hand gestures or performing image regression for locating a key point in an image.



SparkFun DLI Kit
(click image to enlarge)

The kit is further equipped with the Raspberry Pi Camera and a 5V, 4A power supply. There’s also a micro-USB cable and a 2-pin jumper.

 
Further information

The SparkFun JetBot AI Kit is available for $274.95 with the Jetson Nano Dev Kit or $174.95 without. The SparkFun DLI Kit for Jetson Nano sells for $164.95 with the Jetson Nano Dev Kit. More information may be found in the announcement and blog post, as well as product pages for the SparkFun JetBot AI Kit and SparkFun DLI Kit for Jetson Nano.
 

(advertise here)


Print Friendly, PDF & Email
PLEASE COMMENT BELOW

Please comment here...