Enabling AI on Low Power Endpoint Devices -QuickLogic & SiFive Freedom Aware
summary
QuickLogic teams with SiFive Freedom Aware templates and the acquired Sense ML tooling to push AI into battery-powered endpoint devices for predictive maintenance. The SenseML data capture lab collects and labels large, metadata-rich time series; the analytical toolkit searches ~100 feature extractors and ~20 classifiers and returns top candidates with memory, MIPS and power tradeoffs, cutting POC time from months to days. Hardware uses a dual-CPU real-time+housekeeping design and can include an FPGA for >100 kHz sampling; ports already run on RISC-V and common ARM boards.
AI-written summary · about
chapters (14)
- 0:00Intro
- 0:26Resource constrained devices
- 1:46SenseML
- 1:56Platform Segment
- 2:21Mainstream is doing one thing
- 2:34AI Taxonomy
- 3:08Machine Learning
- 4:10Resources
- 6:21Open Standards
- 9:08SenseML analytic toolkit
- 11:39Supported platforms
- 12:56Predictive Maintenance
- 13:24RealTime Component
- 14:45FPGA
related talks
21mUnlocking the potential of low-power embedded devices with Edge AI software
29mAccelerating the RISC-V Revolution: Unleashing Custom Silicon with Revolutionary Design Platforms
26mVision AI at the Edge for Industrial Inspection
48mThe Power of Edge AI: Revolutionizing Future of Real-Time, Autonomous Processing
55mEdgeAI - AI in the Palm of Your Hand
19mMission-Critical AI in Space and Sky: SWaP-Constrained Intelligence Wi...
26mNext generation machine learning for mobile and embedded platforms
22mAI On The Farthest Edge