Alexa, Attack Intruders

If our doom at the hands of our robot overlords is coming, I for one welcome the chance to get a preview of how they might go about it. That’s the idea behind Project Icarus, an Alexa-enabled face-tracking Nerf turret. Designed by [Nick Engmann],  this impressive (or terrifying) project is built around a Nerf Vulcan, a foam dart firing machine gun mounted on a panning turret that is hidden behind a drop-down cabinet door. This is connected to a Pi Zero equipped with a Pi camera. The Zero is running OpenCV and Google Firebase, which connects it with Amazon’s Alexa service.

It works like this: you say “Alexa, open Project Icarus”. Through the Alexa skill that [Nick] created, this connects to the Pi and starts the system. If you then say “Alexa, activate alpha”, it triggers a relay to open the cabinet and the Nerf gun starts panning around, while the camera mounted on the top of it searches for faces. The command “Alexa, activate beta” triggers the Nerf to open fire.

[Nick] has put together a nice build, and has written up the process so you can create your own. Some of it is a bit rough, though: it takes a few seconds to respond, so a quick attacker would already be inside. Perhaps an irising door and slide-out platform would be more robust, and an addition to recognizing faces so you don’t end up shooting the wrong person would be a prudent addition.

3 thoughts on “Alexa, Attack Intruders

  1. Alexa, Double Whammy.

    A couple pinhole cameras would let the pi do the IFF comparisons prior to being activated, though this puts your house under 24/7 surveillance.

  2. Can i just say that Alexa seems incredibly slow in all videos, regardless of hacks or native functionality?

    I mean dont get me wrong, props for the nerf sentry, but if Alexa takes 20-30 seconds to start shooting you might aswell just throw stuff yourself. :P

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.