Hello everyone,
I am a 15-year-old student from Turkey. I’ve developed a wearable device for the visually impaired, using a Raspberry Pi 4 and YOLOv8 for real-time object detection.
My goal is to migrate this project to the NVIDIA Jetson platform to improve performance and leverage CUDA/TensorRT. I am currently learning the documentation and looking for advice from the community on the best hardware module for an Edge AI project of this scale.
Any guidance or learning resources you could share would be greatly appreciated.
Hello @kuzeycecenn28,
Welcome to the Isaac ROS forum and thanks for the post!
Congratulations on building such a cool project at 15. Jetson could be a great next step if you want to take advantage of CUDA and TensorRT acceleration.
For hardware choice: If you want to follow the latest Isaac ROS 4.4 setup exactly, the current Jetson platform listed in the system requirements is Jetson Thor with JetPack 7.1, ROS 2 Jazzy, and a 128 GB or larger NVMe SSD.
If your priority is a smaller, lower-power, more wearable-friendly prototype, Jetson Orin is still a strong option, but you should match it with the Isaac ROS release that supports Orin, such as Isaac ROS 3.2 with JetPack 6.1/6.2 and ROS 2 Humble.
Jetson Orin Nano is a great entry device for learning and prototyping. However, for heavier Isaac ROS workloads, a higher-memory Orin or Thor module will give you more headroom.
Since your project already uses YOLOv8, you may want to look at the isaac_ros_yolov8 package and TensorRT workflow too.
Isaac ROS 4.4 Getting Started: Getting Started — Isaac ROS
Isaac ROS 3.2 for Jetson Orin support: Getting Started — isaac_ros_docs documentation Isaac ROS YOLOv8 package: isaac_ros_yolov8 — Isaac ROS
Great work so far. Please feel welcome to try Jetson!
"Thank you for the guidance! My ultimate career goal is to become an Embedded Systems & AI Engineer at NVIDIA.
I am a highly determined and patient developer with a strong focus on every single detail of my projects. I don’t just want to build things; I want to perfect them. I am committed to putting in my absolute best to ensure my ‘Eye Assistant’ device—and any future project—works flawlessly.
Your advice on Jetson Orin is a huge step toward my dream, and I’m ready to dive deep into your documentation to master these technologies. Looking forward to the day I can contribute to the NVIDIA team!"
Great to hear, best of luck!
We’ll close this topic for now, feel free to open a new one if you have more questions down the road.
Thanks for the support and the good wishes! I’ve already registered for the ‘Build It Yourself’ series and I’m really looking forward to learning more. I will keep working on my projects and will reach out again if I need further technical guidance. See you at the sessions!