Skip to main content

Hey everyone! Brett here, back from a much-needed family trip to Maui and excited to jump back into the Elderly Guardian Challenge.

My Metis AI Accelerator System hardware arrived shortly before my trip, and I’ve successfully got the Metis board plugged in and recognized by my development machine. Huge thanks to the community and Axelera AI for making the setup process pretty straightforward, even after a few days away from the keyboard, it was simple to get started.

Early Technical Progress:
Right now, I’m focusing on the core of the project: real-time fall detection using computer vision. I’m currently wiring up YOLOv8 pose estimation on the Metis AI Accelerator System, leveraging its high-throughput performance to process camera feeds locally. 

Upcoming Integration Plans:
Next steps will be integrating with Alexa for voice check-ins and automating responses (like turning on lights and notifying caregivers) based on the system’s alerts. I’ll also be refining my camera placement to maximize coverage in key living spaces and make the fall detection as robust as possible.

Why this matters:
As I settle back into routine after my trip, I’m reminded of how important peace of mind is for families, especially when we can’t always be there in person. I’m excited to see what the next few weeks bring as I bring together the Metis hardware, YOLOv8, Alexa, and smart home tech to build something truly meaningful for seniors and their families.

If anyone has tips for best practices with YOLOv8 on Metis, or ideas for creative Alexa routines, I’d love to connect!

Looking forward to sharing more progress soon.

Great to hear about the update, I’m also targeting YOLOV8 pose and am in the early stages of trying to run it and navigate the SDK so I have no useful tips yet, anything you find i’ll be interested to know.

Some general issues I found was:
How to solve "ModuleNotFoundError: No module named 'axelera.app'" | Community

This and other errors (like Metis not detected) is mostly solved with a restart. As somebody mentioned in that thread, the difference between running the files can be important.


Hope you had a great holiday ​@moorebrett0!

Can’t wait to see Elderly Guardian progress! Years ago, when I first trained in electronics, I worked for a company that made warden call and panic systems, many of which were in sheltered housing or residental care homes. Something like this would have been incredible, so I can really see not only the safety benefits, but the security people could feel in their own homes from having this running.


Hi ​@moorebrett0 

This is a very exciting and useful project. What are your plans for fall detection? Pose detection + tracking and feeding data into some action recognition model? Or feeding time series keypoints data into some logic that figures out if someone in falling?


Reply