AA Aili Alamdari / Portfolio
← Back to work

Team project · 2025

Modul'accès

SensingLiDAR
ControlsArduino
Modeled gain+27% flow
Concept illustration of the automated sliding door: a small door opens for a pedestrian, both doors open for large objects
Concept illustration — how the two-panel opening logic works in practice.
The problem

Three problems, one entrance

The brief started with a single entryway that was failing three different groups at once, so the fix had to solve all three together, not trade one off against another:

  • 01 Poor accessibility for people with reduced mobility — wheelchairs, stretchers, and equipment don't fit through a standard-width opening.
  • 02 Poor thermal regulation — a door held open (or opened fully for every visitor) lets conditioned or heated air escape.
  • 03 Queue formation at peak hours — a single fixed-width door becomes a bottleneck when traffic spikes.

How it works

The system uses a LiDAR sensor to measure the size of whoever — or whatever — is approaching the entrance, then chooses one of two door configurations automatically:

  • Full opening — triggered when the sensor reads something wider than a standard pedestrian: a wheelchair, a stretcher, or someone carrying equipment.
  • Partial opening — triggered for a regular pedestrian, opening only the smaller of the two door panels. This keeps the airflow — and the heat loss — to a minimum during winter months, while still moving people through quickly.

Mechanically, the two panels ride on a dual guide-rail system — a steel upper rail carrying eight nylon-and-aluminum wheels, with separate aluminum support rails for the small and large door panels (see the guidance-system drawing below). We designed the whole rail and carriage assembly in CAD before building it.

Built and tested end-to-end

This wasn't just a simulation — we built a physical prototype ourselves, wrote the full Arduino control code from scratch, and demonstrated it live in front of a jury. Getting from a rail drawing to a door that reliably opens the correct amount, for the correct object, on the first try meant a lot of iteration on the sensor thresholds and the drive mechanism.

Modeled result: a 27% improvement in people flow through the entrance — all while keeping airflow to a minimum, which is exactly what a Canadian winter calls for.

Gallery