1984 Big Brother is Watching you (and helping with savings)
summary
A home-level analytics stack captures device traffic (including PLC over powerline to keep data local), funnels events into a message queue and stream processor, and applies sliding‑window feature extraction plus Q‑learning and online reinforcement to adapt policies. The deployed system runs on a few hundred sensors, learns routines (e.g., gradual wake‑ups, presence mimicry), detects mold, wiring faults, fire/CO risk and intrusions with multi‑sensor verification, and updates automation rules when behavior or constraints (broken leg, schedule change) shift.
AI-written summary · about
chapters (19)
- 0:00Smart home automation intro
- 3:59Speaker’s IT background
- 7:55Smart homes vs devices
- 10:46Intelligent building
- 13:41Need for smart homes
- 17:34Market reality of smart homes
- 21:30Privacy and security concerns
- 24:39DIY smart home setup
- 27:46Truly smart home
- 31:04Cutting cloud dependency
- 33:55AI learning automation
- 35:59Avoiding single points of failure
- 37:05Wake up automation
- 38:52Security and energy savings
- 40:15Multi-verification for alarms
- 41:42Health and safety apps
- 42:46Mold prevention sensors
- 43:28Detecting depression via audio
- 44:14Conclusion & Q&A
related talks
29mLife on the Edge
05mPredictive maintenance with sensor "BOB"
11mSmart Automation Business Successful Experience Sharing – Gregg Schaefer
21mIoT, Wavelets and Machine Learning in the Smart Home
30mHousehold INFOSEC in a Post-Sony Era #bbuzz
38mEldercam: Event-driven AI for Surveillance
51mThings That Go Bump in the Night: Detecting Problems in the Internet of Things
18mFrom Smart Homes to Smart Grids