High school student Selin Alara Ornek sees a future using machine learning, the NVIDIA Jetson platform for edge AI and robotics to create a robotic guide dog for the visually impaired.
The project, called IC4U, is one of seven robots Olnek has developed so far, including BB4All, a school support robot that helps prevent bullying with real-time notifications and health monitoring.
About the manufacturer
Ornek, a high school senior from Istanbul, Turkey, has always been passionate about the intersection of AI, social impact and robotics. A self-taught roboticist, she built IC4U using a Jetson Development Kit as a sandbox for exploration and experimentation.
She is a member of AI4ALL, a non-profit program with a mission to make AI more diverse and inclusive, and the New York Academy of Sciences. A global figure in the field of robotics, she has been recognized at European Youth Awards and Women in Tech Global Awards events. She won first place at the 2021 Istanbul Bosphorus Robot Cup and third place at the 2023 OpenCV AI Competition.
Her inspiration
Olnek was inspired to found IC4U after seeing guide dogs helping their owners on a trip to France, with his own late dog, Corsan, also being a key inspiration.
“I began to think that if a blind person lost their dog, they would not only lose their best friend, but they would also lose their eyes,” Olnek said.
The project was built to provide visually impaired people with companionship, regardless of age or health restrictions.
Her Jetsons Project
Olnek initially used the ultrasonic sensors in IC4U’s eyes to detect obstacles, but after serving as a panelist at the World Summit on AI in 2021, she decided to develop a new AI application that would enable the robot dog to mimic a real dog.
The ultrasonic sensor could only detect objects in front of the IC4U, so Ornek wanted to expand the detection range to the entire perimeter of the robot.
The solution is to use sound sensors mounted on the robot’s ears, allowing IC4U to orient itself in the direction of the sound and process visual information collected by the integrated ZED 2i wide-angle 3D AI camera, which captures visual data over a wider range and helps detect information such as the size and speed of an object.
“We used the NVIDIA Jetson Nano development kit to power the ZED 2i camera and provide high-quality image processing,” said Ornek. “We were so impressed with the performance of the ZED 2i camera that we didn’t want to limit it to simple object recognition tasks.”
She began thinking about other ways IC4U could help the visually impaired: IC4U’s improved data processing from Jetson-powered high-resolution sensors could detect objects in cities, such as stop signs, traffic light colors, and even the denominations of paper currency.
Additionally, Ornek is using the Jetson Nano to add shopping capabilities to IC4U via web scraping from publicly available resources, with the goal of expanding this functionality in the future by partnering with online retailers.
Back to School
Longer term, Olnek hopes to deploy IC4U for smart city and smart spaces applications and continue exploring AI applications with next-generation platforms such as Jetson Orin.
As a recipient of the Karen McKellin International Leaders of Tomorrow Award, she will begin studying computer science at the University of British Columbia this fall on a full scholarship, and she strives to inspire other young people, especially girls, that technology can be fun.
Students and educators with a valid, accredited university or education email address can sign up to purchase a Jetson Orin Nano or Jetson AGX Orin development kit at a discounted price. Students and educators in the United States can sign up for the discount by visiting Sparkfun. If you’re in another country, check your eligibility (login required).
Learn more about the NVIDIA Jetson platform and Jetson AI courses and certifications from the NVIDIA Deep Learning Institute.
