Phone Bot

Difficulty

Intermediate

BYOB Kit

$30

Build Time

2-3 Weekends

Arena

Navigation · AI Demo

Overview

That old smartphone in your drawer has a camera, GPS, accelerometer, gyroscope, magnetometer, WiFi, Bluetooth, a touchscreen, a speaker, and a microphone — all packed into a board smaller than a playing card. Mount it on a wheeled chassis and you've got an instant robot brain with more sensors than most purpose-built kits provide.

This build uses the phone as a sensor and compute platform. A companion app communicates with an Arduino motor controller via Bluetooth. The phone handles the hard stuff (vision, navigation, AI); the Arduino handles the simple stuff (spinning motors).

The Build

Assemble the chassis, mount the phone, pair via Bluetooth, and start driving. The phone's camera feed becomes your robot's eyes; its sensors become its spatial awareness.

The 2WD chassis comes with two gear motors, wheels, and a front caster wheel. Mount the motors, attach the wheels, and secure the caster. Mount the battery holder and Arduino on the chassis deck. The phone cradle attaches to the front, angled slightly downward for ground-level vision.

Connect the motors to the L298N driver, wire the driver's control inputs to the Arduino, and connect the Bluetooth module. The Arduino receives drive commands via Bluetooth and translates them into motor PWM signals. Tank-style steering: each motor runs independently for differential drive.

// Bluetooth command → motor action
void loop() {
  if (Serial.available()) {
    char cmd = Serial.read();
    if (cmd == 'F') forward(200);
    if (cmd == 'L') turnLeft(150);
// ... full implementation in project files
  

Sideload the included Android APK onto your old phone. The app provides: virtual joystick for manual driving, live camera view, GPS position tracking, and sensor data display. Pair the phone with the HC-05 Bluetooth module (default PIN: 1234).

The app includes three vision modes: color tracking (follow a colored object), line following (track a dark line on a light floor), and face detection (turn toward detected faces). The phone's processor handles the computer vision; motor commands are sent to the Arduino at 10Hz.

For outdoor use, switch to GPS navigation mode. Set waypoints by tapping on the map view. The phone computes heading error using its magnetometer and adjusts motor commands to steer toward each waypoint. Accuracy is about 3-5 meters with phone GPS — good enough for park-scale navigation challenges.