It took a bit, but I was able to finally gather the bits and pieces to get started on my version of the NVIDIA Jetson Nano JetBot. The JetBot is an Mobile Robot Platform that uses the NVIDIA Jetson Nano for its compute power for AI-powered deep learning autonomous navigation. The Jetson Nano is the latest Single Board computer, or rather System-on-Module in the Jetson line by NVIDIA. The Nano is a more affordable System at $99 US where the Jetson TX2 runs $299 - $749 and the Jetson AGX Xavier at $1,099, although the Nano does have a scaled down set of features.
NVIDIA Jetson Nano Developers Blog
NVIDIA Jetson Nano Getting Started Guide
|GPU||NVIDIA Maxwell™ architecture with 128 NVIDIA CUDA® cores|
|CPU||Quad-core ARM® Cortex®-A57 MPCore processor|
|Memory||4 GB 64-bit LPDDR4|
|Storage||16 GB eMMC 5.1|
|Video Encode||4K @ 30 (H.264/H.265)|
4K @ 60
|Connectivity||Wi-Fi requires external chip|
|Camera||12 lanes (3x4 or 4x2) MIPI CSI-2, DPHY 1.1 (1.5 Gbps)|
|Size||69.6 mm x 45 mm|
|Mechanical||260-pin edge connector|
|Brain||NVIDIA Jetson Nano|
|Motors||2 X Adafruit DC Gearbox Motor -"TT Motor" - 200RPM - 3 to 6VDC (3777)|
|Motor Driver||DC Motor + Stepper FeatherWing Add-on Motor Driver|
|Battery||INIU 10000 mAh Portable Power Bank LED Display Ultra Compact 2 USB Ports Mobile Charger External Battery Backup Powerbank|
|uSD Card||Samsung 64GB 100MB/s (U3) MicroSD EVO Select Memory Card|
|USB cable||Type A to Micro, right angle|
|Caster Ball||1" Inch Delrin Plastic Ball Bearings G1-1 Bearings|
|WiFi/BT||Intel Dual Band Wireless-Ac 8265 W/Bt|
|Camera||Raspberry Pi Camera Module V2-8 Megapixel,1080p|
|Camera Lens||Waveshare IMX219 Camera Module for The Official Raspberry Pi Camera Board V2 8MP Sensor 160 Degree FoV|
|Antenna||2 X Molex Antenna Wi-Fi 3.3dBi Gain 2483.5MHz/5850MHz Film|
|Wheels||2 X Thin White Wheel for TT DC Gearbox Motors - 65mm Diameter|
|OLED||Adafruit PiOLED - 128x32 Monochrome OLED Add-on for Raspberry Pi|
|Camera Mount||3D Printed|
|Caster Base||3d Printed|
|Caster Shroud||3D Printed|
JetBot Software Image
NOTE: Image is 64GB and may need a 128GB Micro SD card
1. Download the JetBot SD Card image from the software-setup link posted previously.
The image used in this example is:
2. Etcher was used to Load the Image on the SD card in this example.
NOTE: For uSD cards 64GB or above, the card must be formatted for ExFAT
a. On Linux, use mkfs-exfat
b. If not already installed, in the case of Ubuntu, use apt-get to install the required package
$sudo apt-get install exfat-fuse exfat-utils
c. Format the SD card
$ sudo mkfs.exfat -n sdxc64 /dev/sdh mkexfatfs 1.2.3 Creating... done. Flushing... done. File system created successfully.
d. Launch Etcher and select the image to install
e. A SD card 64GB or greater will not show automatically in Etcher, so it is necessary to click on Select Drive to select the SD card
f. Once the card is selected, click Flash to copy the image to the SD card.
NOTE: A Warning may appear regarding the size of the image. Just click Continue
g. If performed on Ubuntu using Etcher, it is necessary to launch GParted to allocate the allocate space on the SD card. Otherwise the Nano will not boot.
Click Fix to complete the process.
h. Once the GParted process is complete. close the application and safely eject the SD card and insert it back into the Nano
The JetBot Hardware listed in the Bill of Materials is sourced from multiple online sources, however some items may not be available. Considering the JetBot image is hardware specific, it is best to sick with what is listed in the BOM or a fair equivalent. In some cases, there are multiple options for a particular item, such as the wheels where the wider and white rimmed wheels were chosen in this instance rather than the skinnier black wheels. The type of Micro SD card seems to be important to align with the JetBot image size.
NVIDIA Jetson Nano
The JetBot Chassis, Camera Mount, Caster Mount and Caster Shroud were printed on a XYZ da Vinci Mini 3d Printer.
A 3D printed gripper was added for future use.
Chassis with Caster attached
Wheels and Motor Driver
The BOM calls out the Adafruit DC Motor + Stepper FeatherWing Add-on Motor Driver which has mounting holes that align with it on the 3D printed chassis. The motors are standard Hobby Geared DC Motors from Adafruit rated at 200 rpm. The Wheels are Thin White Wheel for TT DC Gearbox Motors that attach to the DC Motors,
The OLED Display is a tiny Adafruit PiOLED that is used to display the system IP Address and related data at boot and during operation.
Camera and Camera Lens
The Camera is a Raspberry Pi Camera V2 that will have its camera module replaced with a Waveshare IMX219 Camera Module for 160 Degree Field of View
WiFi and Antenna
The WiFi module is a M.2 Intel Dual Band Wireless-Ac 8265 W/Bt that attaches to the carrier board for both WiFi and Bluetooth 4.2 support. There will be 2 Molex Wi-Fi 3.3dBi Gain 2483.5MHz/5850MHz Film Antenna attached to extend the WiFi range.
With the hardware in hand and the 3D printer parts completed, its time to assemble the JetBot which will be covered in the next post.