Getting Started

Hardware setup

TI’s TDA4VM SoC houses dual core A72, high performance vision accelerators, video codec accelerators, latest C71x and C66x DSP, high bandwidth realtime IPs for capture and display, GPU, dedicated safety island and security accelerators. The SoC is power optimized to provide best in class performance for perception, sensor fusion, localization and path planning tasks in robotics, industrial and automotive applications.

For more details visit https://www.ti.com/product/TDA4VM

TDA4VM SK

TDA4VM Edge AI Starter Kit (SK) is a low cost, small form factor board designed to bring smart cameras, robots and intelligent machines to life. It packages the same TDA4VM SoC as the EVM version. For more information related to the board, full list of peripherals supported, pin settings for boot modes and more visit TDA4VM-EdgeAI-SK.

To run the demos on TDA4VM SK you will require,

  • TDA4VM Edge AI Starter Kit

  • USB camera (Any V4L2 compliant 1MP/2MP camera, Eg. Logitech C270/C920/C922)

  • Full HD eDP/HDMI display

  • Minimum 16GB high performance SD card

  • 100Base-T Ethernet cable connected to internet

  • UART cable

  • External Power Supply or Power Accessory Requirements

    1. Nominal Output Voltage: 5-20VDC

    2. Maximum Output Current: 5000 mA

    3. Refer to TDA4VM-EdgeAI-SK for more details.

Connect the components to the SK as shown in the image.

_images/board_connections_tda4vm_sk.jpg

Fig. 4 TDA4VM Starter Kit for Edge AI connections

Set the boot pins to SD boot mode as shown in the following image.

_images/TDA4VM-SK-SD-Boot.png

Fig. 5 TDA4VM Starter Kit for Edge AI boot pins

TDA4VM EVM

The Processor SDK Linux for Edge AI can be used to run the edge AI applications on the TDA4VM evaluation module as well. For more information related to the board, the full list of peripherals supported, pin settings, boot modes and more visit TDA4VM-EVM.

To run the demos on TDA4VM EVM you will require,

  • TDA4VM EVM

  • USB camera (Any V4L2 compliant 1MP/2MP camera, Eg. Logitech C270/C920/C922)

  • Full HD eDP display

  • Minimum 16GB high performance SD card

  • 100Base-T Ethernet cable connected to internet

  • UART cable

  • 12V power supply for the EVM.

Connect the components to the EVM as shown in the image,

_images/board_connections_tda4vm_evm.jpg

Fig. 6 TDA4VM evaluation module board connections

USB Camera

UVC (USB video class) compliant USB cameras are supported in both EVM and SK. The driver for the same is enabled in SDK. The SDK has been tested with C270/C920/C922 versions of Logitech USB cameras. Please refer to Getting Error when trying to capture from multiple USB cameras simultaneously to stream from multiple USB cameras simultaneously.

OV5640 YUV sensor

OV5640 CSI camera module is supported by both TDA4VM SK and EVM. It is a 2MP sensor with built-in ISP, which can transmit raw YUYV frames over CSI lanes. This camera module can be ordered from https://www.leopardimaging.com/product/cmos-sensor-modules/mipi-camera-modules/li-am65x-csi2. The camera can be connected to the bottom side of the Edge AI SK board as shown below. The OV5640 camera module requires an additional power supply with following requirements

  • Output Voltage: 5 VDC

  • Output Current: 2.0 A

For added flexibility you can order a Samtec extender cable from here, https://www.samtec.com/products/hqcd

_images/csi_camera_connection.png

Fig. 7 OV5640 CSI based YUV sensor connection with TDA4VM Starter Kit for Edge AI

Note

By default OV5640 is disabled in the SDK. After connecting the camera you can enable it by specifying the dtb overlay file in /run/media/mmcblk0p1/uenv.txt as below,

name_overlays=k3-j721e-edgeai-apps.dtbo k3-j721e-eaik-csi2-ov5640.dtbo

Reboot the board after editing and saving the file.

Please refer Camera sources (v4l2) to know how to list all the cameras connected and select which one to use for the demo.

Software setup

Preparing SD card image

Download the ti-processor-sdk-linux-sk-tda4vm-etcher-image.zip image and flash it to SD card using Balena etcher tool available at:

https://www.balena.io/etcher/

The tool can be installed either on Windows/Linux. Just download the etcher image and follow the instructions to prepare the SD card.

_images/balena_etcher.png

Fig. 8 Balena Etcher tool to flash SD card with Processor SDK Linux for Edge AI

The etcher image is created for 16 GB SD cards, if you are using larger SD card, it is possible to expand the root filesystem to use the full SD card capacity using below steps

#find the SD card device entry using lsblk (Eg: /dev/sdc)
#use the following commands to expand the filesystem
#Make sure you have write permission to SD card or run the commands as root

parted -s /dev/sdX resizepart 2 '100%'
e2fsck -f /dev/sdX2
resize2fs /dev/sdX2

#replace /dev/sdX in above commands with SD card device entry

Power ON and Boot

Ensure that the power supply is disconnected before inserting the SD card. Once the SD card is firmly inserted in its slot and the board is powered ON, the board will take less than 20sec to boot and display a wallpaper as shown in the image below.

_images/boot_wallpaper.jpg

Fig. 9 TDA4VM Starter Kit wallpaper upon boot

You can also view the boot log by connecting the UART cable to your PC and use a serial port communications program.

For Linux OS minicom works well. Please refer to the below documentation on ‘minicom’ for more details.

https://help.ubuntu.com/community/Minicom

When starting minicom, turn on the colors options like below:

# For TDA4VM SK
sudo minicom -D /dev/ttyUSB2 -c on

# For TDA4VM EVM
sudo minicom -D /dev/ttyUSB0 -c on

For Windows OS Tera Term works well. Please refer to the below documentation on ‘TeraTerm’ for more details

https://learn.sparkfun.com/tutorials/terminal-basics/tera-term-windows

Note

Baud rate should be configured to 115200 bps in serial port communication program. You may not see any log in the UART console if you connect to it after the booting is complete or login prompt may get lost in between boot logs, press ENTER to get login prompt

Once Linux boots login as root user with no password. Upon login a script under /opt/edge_ai_apps/init_script.sh is executed which does the below,

  • Makes the wallpaper showing on the display disappear. You will notice “Please wait…” message. This kills weston compositor which holds the display pipe.

  • The display pipe can now be used by ‘kmssink’ GStreamer element while running the demo applications.

  • The script can also be used to setup proxies if connected behind a firewall.

Connect remotely

If you don’t prefer the UART console, you can also SSH to the device if the IP address is known. If the IP address gets automatically assigned then you may have to connect via UART once and find out the allotted IP address by running ifconfig as below:

root@j7-evm:/opt/edge_ai_apps# ifconfig

With the IP address one can ssh directly to the board, view the contents and run the demos.

For best experience we recommend using VSCode which can be downloaded from here.

https://code.visualstudio.com/download

You also require the “Remote development extension pack” installed in VSCode as mentioned here:

https://code.visualstudio.com/docs/remote/ssh

_images/vs_code.png

Fig. 10 Microsoft Visual Studio Code for connecting to TDA4VM Starter Kit for Edge AI via SSH

Installing Dependencies

On a fresh install, you need to first run setup_script.sh as below,

root@j7-evm:/opt/edge_ai_apps#./setup_script.sh

This takes care of installing all the dependencies required to run the demo applications,

  • Clone Tensorflow repo and dependencies to get header files required to build C++ applications

  • Clone ONNX-RT repo and dependencies to get header files required to build C++ applications

  • Clone the edgeai-tiovx-modules repo, build and install the modules library.

  • Clone the edgeai-gst-plugins repo, build and install the custom GStreamer plugins.

  • Clone the edgeai-tidl-tools repo for running standalone inference examples and Jupyter notebooks

  • Compile the C++ apps under apps_cpp folder

Warning

Without installing these dependencies, the Python and C++ demos under edge_ai_apps will not run.

Note

  • We can run the setup_script.sh multiple times. The script checks for what is already downloaded and then executes the next steps like build and install if applicable.

  • If the scripts are executed on Yocto Linux then we need to re-run the setup_script.sh when switching to a Docker session. This is because of the differences between GLIB version in Yocto Linux and Ubuntu 20.04 Docker image.

  • The same applies when switching from Docker back to Yocto Linux. The setup_script.sh must be run once before running any other applications or demos.

Note

In case you are behind VPN, the script will not work since proxy is not handled for tools like git and wget. As a workaround the script can be executed in a docker container. Please refer to Docker Environment for building and running a docker container. Alternatively you can mount the SD card on a linux PC and run the script, make sure you have write permission to the SD card.