Session by session
- Session 11 · Ask & understandWeek 6
Features without recordings: peaks, gaps and duration
Evidence: a question, sketch or new vocabulary
- Session 12 · Build & designWeek 6
Collect a labelled knock-pattern dataset
Evidence: a design draft, dataset or build
- Session 13 · ImplementWeek 7
Train a classifier and a timing-rule baseline
Evidence: a working version, explained once
- Session 14 · Test & improveWeek 7
Evaluate on a separate recording session
Evidence: a test log with at least one failure
- Session 15 · Explain & reflectWeek 8
Explain which patterns the model confuses
Evidence: an individual explanation
- Local Python/scikit-learn
- knock-feature CSV logged in P01
Split by recording session; compared with a timing-rule baseline.
A model, an evaluation study or an AI-checking workflow.
Completion needs the artefact, an honest test log, an individual explanation and no open safety or privacy issue.
Projects in the same block
- RoboticsG12-P01
Sound-Sensor Characterisation Bench
A characterised sound-sensor channel with a measured noise floor and a justified trigger threshold.
Sessions 01-05All plans - RoboticsG12-P02
Fail-Safe Relay Switch
A relay-switched low-voltage lamp that is off by default and always obeys a manual switch.
Sessions 06-10All plans - AIG12-P03
Knock-Pattern Classifier
A classifier that tells three knock patterns from background noise using timing features, not recordings.
Sessions 11-15All plans - IntegratedG12-P04
Knock-to-Light Link
A laptop model that sends named lamp commands to a relay controller that rejects unknown, stale or unsure input.
Sessions 16-20All plans
For school leadersChoose a starting point.
Build from evidence.
Pick the classes and a plan. We map the timetable, kit and safety checks with you, then pilot one class first.
- Prospectus and class-wise plan
- Kit and readiness check
- Pilot one class first
- Evidence at every milestone
Let's plan your pilot.
Share a few details and we will send the right plan for your classes.