Intelligent Systems | Ngā Pūnaha Māhoe

Design systems that collect trustworthy data

A robot that measures nothing is just a toy. A robot that measures something is a scientist.

Wero — the big question

How do we design, program, and deploy intelligent systems that collect trustworthy data?

What you will investigate

Curriculum strands

Physical Science Yr 9–10: Energy, circuits, matter measurement
Technology Yr 9: Electronics & Mechatronics, Computer Science, Digital Technologies
Technology Yr 10: Logic gates, feedback, modularity, AI ethics (Futures Literacies)

Technology strands

Electronics and Mechatronics
Computer Science
Digital Technologies

1–2 termsSupport level: Teacher support when you need it

Studies in this world

See all studies →

advanced6–10 sessions

NuiBot Science Assistant Build

NuiBot is a science assistant — not a toy. What will you ask it to help you investigate?

Microcontroller assistants extend what classrooms can measure — if sensors are calibrated and placed for the question, not the gimmick.

You will assemble (or commission) a NuiBot, calibrate its sensors, and use it as your evidence-gathering partner in a class investigation.

Venture/WeroObserve/KiteInfer/Whakaaro

advanced5–8 sessions

NuiBot Sensor Deployment Challenge

Where you put NuiBot's sensors changes the story it tells. Where will you place it?

Longitudinal sensor deployments generate community-valuable datasets when sampling plans are documented and cross-checked.

You will design and run a real deployment of NuiBot's sensors to answer a question that needs longitudinal evidence.

Observe/KiteInfer/WhakaaroEvaluate/Tohu

developing5–8 sessions

Arduino Data Logger Investigation

An Arduino can listen for hours longer than you can. What would you ask it to listen for?

Open-source loggers let communities build affordable monitoring — code quality and file handling determine whether data survives the field.

You will build a small Arduino-based logger to capture a single variable on a tight schedule and use the data to answer a real question.

Venture/WeroObserve/KiteInfer/Whakaaro

developing3–5 sessions

AI Ethics and Intelligent Systems Investigation

AI can classify, predict, and recommend — but who trained it, on what data, and who wins or loses when it is wrong?

Intelligent systems shape science, health, and whenua monitoring — communities need people who can critique AI fairly, not just use it.

You will investigate a real AI tool used in science, compare its outputs to human expertise, and argue for fairer or more careful use.

Venture/WeroObserve/KiteInfer/WhakaaroEvaluate/Tohu

What this inquiry community is discovering

Patterns from guided studies in this world — useful ideas to test, not answers to copy.

early pattern

Sensor placement mattered more than sensor type

Groups that repositioned their sensor based on their evidence found more meaningful variation in their readings than groups that changed sensors.

Evidence pattern

Across investigations, teams who mapped readings in multiple positions often found a location where the signal changed clearly, even when using the same sensor type.

What it might mean: The investigation should tell you where to put a sensor. Changing hardware doesn’t help if the placement can’t detect the variation you care about.

Try this next

Before changing your sensor, map readings at three different positions to find where variation is highest.

See all Intelligent Systems community findings →

What you might make

Maker pathways connect your evidence to a real prototype or build.

Sensor arm3D print
New arm attachment for the exact position your evidence suggests
Your sensor placement data shows where readings are most informative.
Probe holder3D print
Custom holder for the probe type your investigation uses
A custom holder ensures your probe stays at the calibrated position.
Docking stationLaser cut
Storage and charging station for reliable robot deployment
If battery life or positioning affected your readings, a docking station helps.