A car that sees before it moves.
An autonomous two-wheel drive robot powered by an Arduino Uno that scans its surroundings using high-frequency 40 kHz ultrasonic sonar pulses to navigate around obstacles in real time.
Autonomous obstacle avoidance demonstration with ultrasonic pulse ranging and active servo radar scanning.
18 photos documenting the chassis assembly, wiring, calibration, and live exhibition demonstration.
An engineering summary of acoustic ranging physics, servo kinematics, motor drive electronics, and autonomous decision logic.
| Subsystem | Hardware component | Interface | Operating value |
|---|---|---|---|
| Microcontroller | Arduino Uno R3 (ATmega328P) | Pins 4–10, 5V / Vin | 16 MHz |
| Acoustic sensor | HC-SR04 ultrasonic module | TRIG pin 9, ECHO pin 8 | 40 kHz |
| Scanning actuator | TowerPro SG90 micro servo | PWM pin 10 | 0° to 180° |
| Motor controller | L298N dual H-bridge | Pins 7, 6 (Right), Pins 5, 4 (Left) | 2A peak |
| Power supply | 9V DC battery pack | Isolated rail, common ground | 9V / 5V |
The HC-SR04 emits an inaudible 40 kHz acoustic burst when triggered by a 10 µs HIGH pulse on pin 9. Sound travels through air at approximately 340 m/s (29.1 µs/cm).
The sensor measures round-trip flight duration on pin 8. The Arduino calculates one-way distance:
The HC-SR04 sensor has a narrow beam angle of approximately 15°. Mounting it to an SG90 servo allows an active 180° radar sweep across right (0°), center (90°), and left (180°) positions to identify open escape trajectories.
The L298N motor driver converts low-power logic signals into high-current polarity switching for the two geared motors. Motor power is decoupled from the Arduino logic rail with a shared common ground to eliminate inductive brownout resets.
Servo centers at 90°. Both motors drive forward (MRb=HIGH, MLb=HIGH). Continuous cruise speed.
Maintains current motion state to prevent oscillation (flip-flopping) caused by acoustic measurement jitter.
1. Stop all motors (100 ms).
2. Sweep servo across 0° (right), 180° (left), and 90° (center) at 500 ms intervals.
3. Drive backward for 500 ms.
4. Stop for 100 ms.
5. Pivot turn left for 500 ms.
6. Re-evaluate distance in the next loop cycle.
Six questions exploring acoustics, thermodynamics, sensor blind zones, and motor control edge cases.
Nearer. Sound travels faster in hot air, so the echo returns sooner. The code assumes a fixed speed of sound, so it calculates a shorter distance. The error is small (a cm or two), but it is always in the “too close” direction.
The sensor faces forward, so the car backs up blind. It can hit a wall, a person’s foot or another robot behind it. The fix is a rear sensor, or only reversing for a short, safe distance.
Time isn’t angle. As the battery drains, the motors slow down and the car turns less. Floor grip, wheel slip and uneven motors also change the result. Encoders on the wheels or a gyroscope measure the actual turn instead of guessing it.
They interfere, because each sensor hears the others’ pings and echoes. This is the same “phantom obstacle” problem as Quiz Q2. The fix is to fire them one at a time, with a short pause between, so each one only listens for its own echo.
When no echo comes back (nothing in range, or the echo is lost), the timing function typically returns 0, which turns into a distance of 0 cm. Without > 0, the car would treat that as an obstacle right in front of it and stop for nothing.
The HC-SR04 has a blind zone: it can’t measure very short distances (its minimum is around 2 cm). The echo returns while the sensor is still finishing its own ping, so it can’t tell them apart. Objects that close are effectively invisible to it.
Project files in Website/Assets. Click anywhere on an asset row to view, or Download to save the file.
| Asset name | Filename | Format | Actions |
|---|
If your have explored our project or have visited the exhibition at St. Aloysius School on 6th Nov, 2026, please leave a feedback.