Open Air Quality Dashboard with Low-Cost Sensor Correction
A dashboard combining official monitoring stations with cheap community sensors, correcting the cheap readings against the reference stations, and being honest about uncertainty. Low-cost particulate sensors read badly in humidity, and correcting for that is the actual work.
Intermediate
A small team, or one strong student willing to learn something new.1 semester, 2 students
5 to ship
3 optional extrasSuggested stack
What you should ship
- Ingestion from at least one official source and one community sensor network
- Calibration model correcting low cost readings against a co-located reference station
- Map and time series interface showing readings with an uncertainty band
- Health guidance mapping current levels to the relevant national or WHO advice, clearly attributed
- Validation reporting corrected sensor error against reference readings on held out days
If you have time left
- Short horizon forecasting using weather inputs
- Alerting when levels cross a threshold in a chosen area
- Detecting and excluding sensors that have failed or drifted
The problem
Official air quality stations are accurate, expensive and sparse. Community sensors are cheap and numerous and read poorly, particularly in high humidity where particulate sensors substantially overestimate. Presenting either alone gives a misleading picture.
What you build
Ingestion from both source types, a correction model fitted where a cheap sensor sits near a reference station, and a public interface that shows corrected values with their uncertainty.
The scientific contribution
The calibration. Fit a model using temperature and humidity alongside the raw reading, validate it against reference data from days not used in fitting, and report the error before and after correction. That number is your result.
Why uncertainty must be visible
A single number on a map implies precision the data does not have. Showing a band, and explaining what it means, is both more honest and a better piece of design work.
The hard part
Sensor drift and failure. Community sensors go bad quietly and keep reporting. Detecting a sensor that has diverged from its neighbours is necessary for the map to stay trustworthy over time.
Scope warning
Particulate matter only. Gas sensors have different and harder calibration problems and adding them will cost you the depth on the part you can do well.
Ideas and guidance, not finished projects
These are project ideas and scoping guidance, published free for students to use as a starting point. I do not build, write, or sell final-year projects, and I do not complete coursework for anyone. Take an idea, make it yours, and build it.
Not sure this one fits you?
I help students pick an idea that matches their skills and their deadline, then scope it down until it is finishable. Guidance only, never done for you.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Every session is free; a few slots open each week.
Follow along
New writing, resources and project ideas land here first.