SK-TDA4VM Setup Guide
Hardware requirements
- A Texas Instruments TDA4VM single-board computer
- A USB-C power cable to power the TDA4VM board
- A microSD card
- A desktop or laptop computer for flashing the microSD card
- A way to connect the microSD card to the computer (a microSD slot or microSD reader)
- An Ethernet cable
- An HDMI cable
Required downloads
Download the following files to your computer:
Download the PROCESSOR-SDK-LINUX-SK-TDA4VM — Linux SDK for edge AI applications on TDA4VM Jacinto™ processors image.
Next, download and install the Balena Etcher for your desktop/laptop OS. You will use the Balena Etcher to flash the microSD card.
Flash the image
Important
You must extract the image from the zip file before flashing the microSD card.

Insert the microSD card into a reader connected to your computer.
Launch Balena Etcher.
Click Flash from File to open the file selector.
Navigate to and select the image you downloaded.
Click Select Target to choose the storage device corresponding to your microSD card from the selector window.
Click on the desired device, then click Select to continue.
Click Flash!. If you receive a warning concerning the size of the microSD card, ensure that you have inserted the proper microSD and also selected the proper device, then click, Yes, I’m sure to flash the board. The flashing and verification process may take 10-20 minutes, depending on your system.
Tip: How to think about building a machine
While the Etcher is flashing your microSD card, we recommend reading How to think about building a machine.
On completion of the flashing and validation process, remove the microSD card from your computer and insert it into the TDA4VM.

Install Viam dependencies on the TDA4VM
Connect the board to Ethernet.
Connect the board to a monitor with the HDMI cable.
Connect the board to power using the USB-C power cable.
Use the credentials and IP address displayed in the upper right-hand corner of the monitor to SSH into the board.
From the SSH session on the TDA4VM board:
Clone the TDA4VM repo:
git clone https://github.com/viam-labs/tda4vm-setup.gitNavigate to the setup directory:
cd tda4vm-setup/Make the server setup script executable:
chmod +x tda4vm-viam-setup.shLaunch the setup script to install
viam-serverdependencies:./tda4vm-viam-setup.shOnce this process completes, the board will reboot.
Install viam-server
viam-server is the open-source software that drives your hardware and connects your device to the cloud.
Install viam-server on the computer or single-board computer (SBC) that is directly connected to your hardware (for example sensors, cameras, or motors):
Make sure your computer or SBC is powered on and connected to the internet.
Create a Viam account on app.viam.com. You can configure and manage devices and data collection in the web UI.
Create a new machine using the Add machine button in the top right corner of the LOCATIONS tab in the app. A machine represents your device.
On your machine’s page, click View setup instructions.
Select the appropriate architecture for your machine: Linux / Aarch64, Linux / x86_64, or Linux / Armv7l.
On most Linux operating systems, you can run
uname -mto confirm your computer’s architecture.If you selected Linux / Aarch 64 or Linux / x86 also select your installation method:
viam-agent(recommended): installs viam-agent, which will automatically install (and update) viam-server and provide additional functionality such as provisioning and operating system update configuration.manual: installs onlyviam-serveron your machine.
Follow the instructions on the page to install
viam-serverand connect it to the cloud with your machine’s unique credentials.After you install
viam-server, a secure connection is automatically established between your machine and Viam. When you update your machine’s configuration,viam-serverautomatically gets the updates.You are ready to configure supported hardware on your machine.
Try an example
Once you’ve installed viam-server and your machine has come online, if your machine has a webcam, you can try an example project:
Click the + icon next to your machine part in the left-hand menu and select Insert Fragment.
Add the
DeskSafariGamefragment by theRobot Landorganization and click Insert Fragment. This adds a number of resources to your machine:- a camera component which connects to the webcam
- machine learning resources to run a model and apply it to the camera stream
- control logic that implements a game
Save your config and review the available resources on the CONFIGURE tab.
Log into this Viam application with your Viam credentials and select your machine. The application provides a UI for playing the game
Select a camera and press the Start Game button. The goal of the game is to find and show specific objects to the camera.
If you’d like to learn how to create this game, see the Desk Safari tutorial.
Should the game not work, return to your machine in the Viam web UI and check the LOGS tab for errors.
Next steps
Need assistance?
You can also ask questions in the Community Discord and we will be happy to help.
Was this page helpful?
Glad to hear it! If you have any other feedback please let us know:
We're sorry about that. To help us improve, please tell us what we can do better:
Thank you!