Hackathon Fall 2026

Sentry Car

A Raspberry Pi 5 robot, driven from a laptop, that moves, turns, explores, speaks, and senses.

arpitran/hackathon-fall-2026

The idea

Built milestone by milestone toward an AI agent

Plan

Eight milestones, 24 hours

#MilestoneStatus
M0Prep: order parts, bring up the Pi 5in progress
M1Eyes on the Pi: person detection on the boardtodo
M2Hello GPIO: blink an LED, read a buttontodo
M3Wheels: Car class + drive CLItodo
M4Laptop control: FastAPI + WebSocket + watchdogtodo
M5Sentry vision: live stream with person boxestodo
M6Stretch: voice or obstacle sensingtodo
M7Demo video + docstodo

Hardware

Parts

Brain

Raspberry Pi 5 with active cooler

Motor control

ESP32-S3 co-processor, commanded by the Pi over USB serial

Drive

L298N driver, four BO motors and wheels (skid steering), separate 4×AA battery pack (6V)

Eyes

Logitech USB webcam

Voice

USB speaker (the Pi 5 has no 3.5mm jack)

Body

Doubled cardboard chassis, breadboard for prototyping

Architecture

Laptop to Pi to ESP32-S3 to motors

Architecture diagram: laptop talks WebSocket to the Pi 5, which sends USB serial commands to an ESP32-S3 that drives the L298N and motors; the Pi reads a USB webcam; motors run on a separate AA pack

Wiring

ESP32-S3 to driver to motors, one shared ground

Wiring diagram: ESP32-S3 GPIO to L298N to four motors, two per side, with a shared ground bus and a separate battery pack

Breadboard

ESP32-S3 on the breadboard, wired to the Pi and the driver

Breadboard layout: the ESP32-S3 sits across the centre gap, connected to the Pi by a USB cable, to an LED circuit by male-to-male wires, and to the L298N inputs by male-to-female wires

Electronics

Why an LED needs a resistor

A Pi pin puts out 3.3 V and should supply about 16 mA at most. The resistor decides how much current flows.

Pi GPIO pin3.3 V
→
Resistor220 Ω
→
LED, drops about 2 Vlit
→
PiGND
none471002204701k10k
0 mA16 mA pin limit30 mA
5.9 mA

Red and yellow LEDs drop about 2 V. Blue, white and green drop about 3 V, so use a smaller resistor (100 Ω) if one looks dim. Drag the slider; press Esc to get the arrow keys back.

Vision

Person detection, already working

A YOLO11n detector runs on webcam frames with tracking and a live FPS counter. It is the base for the sentry view in M5.

uv run --extra vision detect-people

Demo

Live demo

Drive the car from the laptop, watch the person boxes.

Fallback: recorded demo video

Next

After the hackathon

← → / space / click