Smart parking that counts free spaces

The problem

Searching for a parking space at malls and hospitals wastes visitors' time and adds to congestion inside the car park, since drivers only find out whether there's a free space after entering and circling two floors. At heart the problem is simple: the information exists — each space is either free or taken — but there's no way to collect it and show it at the entrance.

How it works

1. SensingA distance sensor mounted above each space detects whether there's something below it.
2. ClassificationAn object present for long enough means a car, not a passing person.
3. CountingThe system collects the status of all spaces and counts how many are free and taken.
4. DisplayA board at the entrance shows the number of free spaces and how they're spread across floors.
5. GuidanceColored indicators above each space: green for free, red for taken.
6. FollowingAn online dashboard shows occupancy and peak times.

Project features

• It tackles a daily problem everyone faces and can be explained in one sentence.
• The colored indicators guide the driver from a distance without reading any text.
• Occupancy data helps the car park's management decide on pricing or expansion.
• An easy demo model with four miniature spaces and small cars.
• It can expand to dozens of spaces with the same architecture.

Expected challenges

These are the main obstacles to expect during the build, laid out clearly so you know them before you start.

• The distance sensor is affected by direct sunlight and shiny car surfaces.
• Telling a parked car from a person standing or a shopping cart.
• Running wires to dozens of spaces is costly, and a wireless solution also raises the cost.
• Powering each sensor in an open car park.
• Dust building up on the sensor changes its readings over time.

The consultation gives you a solution for each one.

Expected committee questions

Prepare your answers in advance: these questions come up often in defenses of similar projects.

1. Why was this type of sensor chosen to detect cars?
2. How does the system handle a car parked across two spaces, or a motorcycle?
3. How does data from all the sensors reach the entrance board?
4. How can the system be expanded to multi-story car parks?

Practice the defense, question by question