🚗 Raspberry Pi 3 RC Car

A three-wheeled robot car controlled from any phone browser over Wi-Fi, built on a Raspberry Pi 3 and programmed entirely in Python. Features real-time obstacle avoidance with an HC-SR04 ultrasonic sensor, automatic LED brightness via an LDR and MCP3008 ADC, hardware PWM motor control, and a touch-friendly joystick web interface.

---

📸 Gallery

  

Left: top-down view showing the RPi 3, motor driver, breadboard, HC-SR04 and LDR circuit. Right: the mobile web joystick used to drive and steer the car.

---

✨ Features

• Mobile web controller — touch joystick + accelerate button + reverse toggle, served directly from the Pi over Wi-Fi (no app install needed)
• Obstacle avoidance — HC-SR04 ultrasonic sensor progressively caps forward speed as the car approaches an obstacle; full stop at < 8 cm
• Automatic LED — LDR reads ambient light via MCP3008 ADC; LED dims in daylight and brightens in darkness
• Hardware PWM — smooth, jitter-free motor speed control using the Raspberry Pi's true hardware PWM channels
• Differential steering — independent left/right wheel speed model for smooth turning
• Fail-safe — if the browser connection is lost for > 0.5 s, the car stops automatically
• Clean shutdown — Ctrl+C stops motors and releases GPIO resources gracefully

---

🛠️ Hardware

Bill of Materials

Component	Model / Notes	Qty
Single-board computer	Raspberry Pi 3 Model B / B+	1
Motor driver	L298N dual H-bridge	1
DC gear motors	6 V TT motors (yellow/black)	2
Wheels	Compatible with TT motor shaft	2
Ultrasonic sensor	HC-SR04	1
ADC	MCP3008 (10-bit, SPI)	1
Light sensor	LDR (photoresistor) + 10 kΩ resistor (voltage divider)	1
LED	Any colour + current-limiting resistor (~220 Ω)	1
Power supply	4× AA batteries (6 V) for motors; USB power bank for Pi	1 each
Chassis	Laser-cut wood 2WD platform	1
Breadboard	Full-size	1
Jumper wires	Male–male and male–female	assorted

GPIO Wiring

All pin numbers use BCM (Broadcom) numbering — this is what gpiozero and RPi.GPIO use by default.

Motor driver (L298N)

Signal	BCM GPIO	Physical Pin	Notes
Left motor IN1	4	7	Direction control
Left motor IN2	5	29	Direction control
Right motor IN1	6	31	Direction control
Right motor IN2	12	32	Direction control
Left motor ENA (PWM)	18	12	Hardware PWM channel 0
Right motor ENB (PWM)	19	35	Hardware PWM channel 1

Ultrasonic sensor (HC-SR04)

Signal	BCM GPIO	Physical Pin
TRIG	20	38
ECHO	21	40

⚠️ The HC-SR04 ECHO pin outputs 5 V. Use a voltage divider (1 kΩ + 2 kΩ) to bring it down to ~3.3 V before connecting to the Pi.

ADC & light sensor (MCP3008 + LDR)

The MCP3008 connects to the Pi's hardware SPI bus (SPI0):

MCP3008 Pin	Raspberry Pi (BCM)	Physical Pin
VDD, VREF	3.3 V	1
AGND, DGND	GND	6
CLK	11 (SCLK)	23
DOUT (MISO)	9	21
DIN (MOSI)	10	19
CS/SHDN	8 (CE0)	24
CH0	—	LDR voltage divider output

LED

Signal	BCM GPIO	Physical Pin
LED anode (+ resistor)	25	22
LED cathode	GND	any GND

---

📁 Project Structure

raspberry-pi-car/
├── images/
│   ├── car_top_view.jpg
│   └── car_with_web_controller.jpg
├── src/
│   ├── main.py            # Entry point — 50 Hz control loop
│   ├── car.py             # Car class (motors + LED + LDR via ADC)
│   ├── constants.py       # All hardware config and tuning parameters
│   ├── ultrasonic.py      # HC-SR04 sensor driver
│   └── web_controller.py  # Flask server + mobile joystick UI
├── .gitignore
├── LICENSE
├── README.md
└── requirements.txt

---

⚙️ Installation

1. Prerequisites

• Raspberry Pi 3 running Raspberry Pi OS (Bullseye or later, 32/64-bit)
• Python 3.11+ (python3 --version)
• The Pi and your phone must be on the same Wi-Fi network

2. Enable hardware interfaces

Hardware PWM

Add the following line to /boot/config.txt (or /boot/firmware/config.txt on newer OS versions):

dtoverlay=pwm-2chan

Then reboot: sudo reboot

SPI (for MCP3008)

sudo raspi-config
# → Interface Options → SPI → Enable → Finish

3. Clone the repository

git clone https://github.com/<YOUR_USERNAME>/raspberry-pi-car.git
cd raspberry-pi-car

4. Install Python dependencies

pip install -r requirements.txt

If you get a permission error, use pip install --break-system-packages -r requirements.txt on newer Raspberry Pi OS versions.

---

🎮 Usage

Start the car

python src/main.py

The terminal will print the server URL and live telemetry:

Web server started.
Open on your phone: http://<RASPBERRY_PI_IP>:8080

rev=False  dir=+0.00  acc=0.00  dist= 85.3 cm  light=1.92 V  led=0.00  L=0.00  R=0.00

Control from your phone

1. Find the Pi's IP address: hostname -I
2. Open http://<PI_IP>:8080 in your phone browser (Chrome / Safari)
3. Use the joystick to steer, hold Accelerate to go forward, tap Reverse to toggle reverse mode
4. Press Ctrl+C in the terminal to stop everything cleanly

---

🧠 How It Works

Architecture

The project follows a clean single-responsibility module design:

main.py  ──reads──▶  web_controller.py   (Flask REST API + mobile UI)
         ──reads──▶  ultrasonic.py        (HC-SR04 distance measurement)
         ──uses ──▶  car.py               (motors + LED + LDR)
                         └──reads──▶  constants.py  (all config)

Obstacle avoidance

The PROXIMITY_TABLE_CM in constants.py defines a piecewise-linear speed cap:

Distance  →  Max forward speed
≥ 100 cm  →  100 %
   60 cm  →   70 %
   25 cm  →   30 %
   15 cm  →   15 %
 ≤  8 cm  →    0 % (full stop)

The car's actual forward speed is joystick_acceleration × speed_cap, so it slows down gradually instead of braking suddenly. In reverse, the cap is disabled and speed is limited to MAX_REVERSE_SPEED = 0.4.

Automatic LED

The LDR forms a voltage divider with a 10 kΩ resistor. As ambient light increases, LDR resistance decreases and the voltage at the MCP3008 input rises. The compute_led_brightness() method maps:

voltage ≥ 1.74 V  →  LED off   (bright environment)
voltage ≤ 0.50 V  →  LED full  (dark environment)
in between        →  linear interpolation

Differential steering

compute_wheel_speeds() implements a simple differential-drive model: the outer wheel keeps the full base speed while the inner wheel is reduced by |steering| × STEERING_GAIN. This produces smooth, proportional turns without abrupt speed changes.

---

🎓 Academic Context

This project was developed as the final project for the Electronic Systems course in the Mathematical Engineering and Artificial Intelligence degree (iMAT) at Universidad Pontificia Comillas ICAI, Madrid.

---

🚀 Future Improvements

• Autonomous navigation
• Computer vision using OpenCV
• Object detection with AI models
• Path planning algorithms
• ROS 2 integration

📄 License

This project is licensed under the MIT License — feel free to use, modify, and distribute it with attribution.