IoT Parking Slot Indicator with IR Sensors and Live Display

Drivers circle parking lots hunting for a free bay while half the bays sit empty one floor up — the information exists at the sensor but never reaches the driver. The fix is a per-slot sensing layer with a display at the entrance. This project builds that layer as a student prototype: each parking bay gets an IR obstacle sensor, an ESP32 polls all slots, and the status is shown on red/green LEDs per bay, a 16x2 LCD summary at the "entrance", and a cloud dashboard with a live slot map. The firmware debounces sensor readings so a person walking past a bay does not flip its state, and only stable changes are published. The dashboard keeps a simple occupancy history, which turns the demo into a small data story about peak hours.

1. Each IR sensor watches its bay; the ESP32 scans all sensor pins in a loop every few hundred milliseconds.
2. Raw readings pass through the debounce filter — a slot flips state only after N consecutive consistent samples (default 5).
3. On a confirmed change, the ESP32 updates the slot's LED (green for free, red for occupied) and refreshes the LCD summary.
4. The new state is published over Wi-Fi to the cloud dashboard, which repaints the live slot map within seconds.
5. Every change is timestamped in the dashboard log, building the occupancy history used for the peak-hours view.
6. The LCD always shows the current totals, so the entrance display works even if Wi-Fi drops — only the cloud map needs connectivity.

Tech stack:

• ESP32 development board (Wi-Fi)
• Arduino IDE (C/C++ firmware)
• IR obstacle-avoidance sensor modules (one per slot)
• Red/green LED pairs with resistors
• 16x2 LCD with I2C backpack
• Cloud dashboard (ThingSpeak/Blynk class) over MQTT/HTTP
• Miniature parking-bay model with demo vehicle
• 5 V DC supply

• [Per-slot IR sensing] Each parking bay has an IR obstacle-avoidance sensor; the ESP32 polls all slots in a scan loop with per-sensor sensitivity adjustment.
• [Debounced state logic] A slot changes state only after consecutive consistent readings, so pedestrians or passing shadows do not cause flickering status.
• [LED per bay] Red/green LED pairs at each slot give instant at-a-glance status, wired through current-limiting resistors.
• [Entrance LCD summary] A 16x2 LCD shows total free slots and per-slot status (e.g. "FREE: 2/4, A:OCC B:FREE") for the driver at the gate.
• [Live cloud slot map] Slot states are published over Wi-Fi (MQTT/HTTP) to a dashboard showing a live map of the lot with free/occupied coloring.
• [Occupancy history] The dashboard logs state changes with timestamps, producing a simple peak-hours view of lot usage.
• [Scalable slot count] The firmware's slot table is configurable — the demo ships with four bays and the code supports expansion by adding sensor channels.

• Working prototype: ESP32, 4 IR sensors, LEDs, LCD, miniature parking bays with demo car
• Complete ESP32 firmware (slot scanning, debounce logic, LED/LCD drivers, cloud publishing)
• Cloud dashboard setup guide (slot map configuration, history charts)
• Wiring diagram and connection table
• Sensor sensitivity adjustment guide
• Project report PDF (background, sensing and debounce design, methodology, working, results)
• PPT presentation for final review
• Viva Q&A preparation document (IR sensing, debouncing, ESP32 GPIO expansion, MQTT telemetry)

• IR sensors detect presence within approximately 2–30 cm; the demo bays are sized to this range — a real lot needs longer-range or different sensing (listed as future scope).
• Direct strong sunlight can saturate IR receivers; the demo is designed for indoor/shaded demonstration.
• Debouncing adds a short delay (about a second) before a state change registers — a deliberate trade-off against flicker, documented in the report.
• The dashboard needs Wi-Fi; the LEDs and LCD keep working locally without it.
• Occupancy history is only as granular as the dashboard's logging interval; it shows trends, not second-by-second analytics.