CircuitSense: Automatic Sensing of Physical Electric Circuits and Generation of Virtual Circuits ...
summary
System automatically senses breadboard pin locations by attaching 80 strain gauges to clips and reading tiny resistance changes via Wheatstone bridges, ADCs and multiplexers, and actively probes components with a 50 Hz square wave (0.5–2 V) to collect electrical responses. It classifies 22 component types with a random forest trained on 11,000 waveforms (claims 100% accuracy in their tests), estimates R/C/L from voltage-divider and RC/inductor laws, runs per-component probing in about five seconds, and struggles only with some parallel configurations requiring small software edits.
AI-written summary · about
chapters (15)
- 0:00Start
- 4:43Metal Clips
- 5:17Sensing Clips
- 5:49Circuit Design
- 6:36Prototype Design
- 7:10Component Location
- 8:20Component Type
- 8:39Active Probing
- 9:06A Component Inserting...
- 10:08Data Collection - IC
- 10:41Component Classification
- 11:16Evaluation
- 11:54Circuit Diagram
- 12:42Component Specification
- 14:40CircuitSense
related talks
20mThe Toastboard: Ubiquitous Instrumentation and Automated Checking of Breadboarded Circuits
15mCircuitStack: Supporting Rapid Prototyping and Evolution of Electronic Circuits
19mCircuitStyle: A System for Peripherally Reinforcing Best Practices in Hardware Computing
06mA Survey and Taxonomy of Electronics Toolkits for Interactive and Ubiquitous Device Prototyping
14mRISC-V and Antmicro’s visual system designer: Everything everywhere all at once
12meTactileKit: A Toolkit for Design Exploration and Rapid Prototyping of Electro-Tactile Interfaces
07mWeaving Schematics and Code: Interactive Visual Editing for Hardware Description Languages
10mForsense: Accelerating online research through sensemaking integration and machine research support.