🚦

THE AI CLUB · ENGINEER LAB

Traffic Tech Lab

⭐ 0 XP🏅 Engineer in training

LEVEL 1 · Wake up the traffic light

Connect a timer, write the traffic sequence and bring the lights to life!

YOUR PROGRESS

👣 Byte’s Building MapNot sure what to click? Follow these little steps and build the signal yourself!

STEP 1 OF 6

Meet your mission

✅ YOUR MINI CHECKLIST

🏙️ Future City Junction

Live simulation — YOUR program controls the signals

● PAUSED

● LIVE TRAFFIC

00:00 · phase: 0s

⚠️ Safety problem! Debug your rules before vehicles proceed.

🎛️ Create a traffic situation

SpeedSystem is idle

Byte: I'm here with clues, not answers. Connect your timer to get started!

Step 1 · Choose and connect

Install sensors, then wire them to the controller.

0 connected

👆 Wiring: Tap a sensor's blue plug, then tap the matching controller IN socket. For outputs, tap a controller OUT plug, then its actuator socket. You can make a wrong connection and learn by testing!

SENSOR OUTPUTS

🧩 THE CONTROLLER

INPUT SOCKETS (IN)

OUTPUT SOCKETS (OUT)

ACTUATORS

Pick a sensor from the tray to begin.

Step 2 · Program your traffic brain

Write the rules. Your code actually controls the simulation!

TRAFFIC_CONTROLLER.pseudoIF • THEN • ELSE • END IF

📖 My syntax cheat sheet and colour preview

IF condition THEN action; optional ELSE IF and ELSE; close with END IF. Use AND / OR for combined conditions. Conditions read the sensor and light values at the start of each second. If multiple rules set the same output, the last assignment wins. Timer counts seconds since a traffic-light colour changed. Use timer, pedestrian_button, vehicle_sensor, siren_sensor, car_light, side_light, pedestrian_signal. Values: GREEN, YELLOW, RED, WALK, WAIT, PRESSED, RELEASED, DETECTED, CLEAR, TRUE, FALSE. Change variables by writing car_light = YELLOW.

Step 3 · Test, observe, debug!

Your program runs through real test scenarios.

—Passed

—Needs work

0Debug attempts

Your lab report appears here after you run tests. Wrong answers are useful evidence — fix the code and test again.

🔒

AI Upgrade LabComplete the four automation missions to unlock this experiment.

Remember: writing IF–ELSE rules is automation. In this lab you'll train a small model using examples.

🧠

AI Upgrade · Can a signal learn?You're about to train a real tiny machine-learning classifier locally in your browser!

Rule-based: YOU write every condition. AI: You give labelled examples and the model learns which examples are similar. The AI only suggests green duration; your safety rules stay in charge.

1 · Training examples

Edit any example, add new ones, then train again. Labels represent traffic demand, not guaranteed answers.

Hour	Main cars	Side cars	Label	×

2 · Try a new traffic situation

These are new inputs, not labelled training rows.

Time (hour)Main carsSide cars

Train your model to see what it predicts.

🛑 Safety first: AI cannot override pedestrian and emergency rules. After training, your pseudocode can use ai_green_target instead of a fixed number (e.g., IF timer >= ai_green_target THEN). You must still write the rule.

Automation vs AI

30sExample fixed green duration

—AI-suggested green duration

0Training examples

Reflect: Did you tell the system every answer, or did it learn from labelled examples? Try retraining with poor data. What changed?

📓 Engineer Notebook — Open when you need help

📡 What is a sensor?A sensor collects information, such as whether a pedestrian has pushed a button. It provides an input, not an action.

⚙️ What is automation?An automated controller follows the rules you write. IF button PRESSED THEN change its outputs.

💬 What is pseudocode?A readable plan for a program. You can use IF, ELSE, AND and OR to say what the controller should do.

🔀 AND or OR?AND requires both conditions. OR requires either one. Try them in your Level 3 truth table.

🔁 What is debugging?Run your program, look at failures, locate the mistaken rule, improve it and test again.

🧠 What is AI?Unlike fixed rules alone, a trained AI model finds patterns in example data and uses them to make predictions. Predictions can be wrong.