The development board runs AI on the device using two processors. It supports voice, vision, and robot control. Find out more!
Arduino is preparing to release VENTUNO Q, a development board for running artificial intelligence on edge devices with real-time control. The board uses two processors. An AI processor from the Qualcomm Dragonwing IQ8 Series runs AI workloads, including generative models. An STM32H5 microcontroller handles tasks such as motor control and actuation. The AI processor includes a neural processing unit with up to 40 dense TOPS.
VENTUNO Q includes 16 GB RAM and supports up to 64 GB storage. The board can run AI on the device instead of the cloud. Possible uses include voice assistants that run local language models, gesture-based mirrors, and information kiosks that use speech recognition and text-to-speech.
It can also be used for robotics. Examples include pick-and-place arms using vision, robots that follow users, and machines that move using visual SLAM. Other uses include monitoring and inspection systems. Devices can analyze security video, process traffic data, or inspect products using vision-language models to find defects or missing parts.
The main processor runs Ubuntu and Debian. The microcontroller runs Zephyr OS with the Arduino Core. Applications can be built in the Arduino App Lab environment using Arduino sketches or Python. Models are available for speech recognition, gesture recognition, pose estimation, and object tracking. These run offline through Qualcomm AI Hub.
The board also works with Edge Impulse Studio for building AI models. The board can run as a single-board computer or connect to a PC. Hardware interfaces include CAN-FD, PWM, and GPIO. Robotics development can use ROS 2. The board also includes connectors for MIPI-CSI cameras, audio, displays, and 2.5-Gb Ethernet. It works with Arduino UNO shields, Arduino Modulino nodes, Qwiic sensors, and Raspberry Pi HATs.
“VENTUNO Q reflects our shared commitment to make edge AI more powerful and more accessible,” said Nakul Duggal, EVP and Group GM, Automotive, Industrial and Embedded IoT, Qualcomm Technologies, Inc. “By uniting Arduino’s developer ecosystem with the power of Dragonwing processors, we are making advanced edge AI available to millions of developers worldwide.”
