PiCar Brain

Voice-controlled robot car for Leon. Wake word "Jarvis", speaks Swedish.

Quick Reference

Item	Value
Pi hostname	picar.local
SSH	ssh pi@picar.local (password: leon)
Project path	/home/pi/picar-brain
App service	picar-app (phone control)
Voice service	voice (Hey Jarvis)

Architecture

┌─────────────────────────────┐     ┌─────────────────────────────┐
│     app_control.py          │     │    voice_assistant.py       │
│     (picar-app service)     │     │    (voice service)          │
│                             │     │                             │
│  Picarx (motors/servos)     │     │  Button (GPIO 25)           │
│  Camera (Vilib)             │     │  LED feedback               │
│  SunFounder app (8765)      │     │  Speaker (TTS)              │
│  Socket server (5555) ◄─────┼─────┤  Mic + Whisper              │
│                             │     │  Wake word (Porcupine)      │
│  Executes: forward, turn,   │     │  GPT conversation           │
│  stop, camera pan/tilt      │     │  Socket client → sends cmds │
└─────────────────────────────┘     └─────────────────────────────┘

Both services run simultaneously. Voice sends movement commands to app via localhost socket (port 5555).

Deployment

# On Mac: Edit, commit, push
cd ~/picar-setup/picar-brain
git add -A && git commit -m "message" && git push

# On Pi: Pull and restart
sshpass -p 'leon' ssh pi@picar.local "cd ~/picar-brain && git pull && sudo systemctl restart picar-app voice"

Or step by step:

ssh pi@picar.local
# Password: leon
cd ~/picar-brain
git pull
sudo systemctl restart picar-app voice

Services

Service	Port	What it does
picar-app	8765, 9000	Motors, camera, phone app
voice	5555→	Wake word, TTS, GPT

# Status
systemctl status picar-app voice

# Logs
journalctl -u voice -f
journalctl -u picar-app -f

# Restart
sudo systemctl restart picar-app voice

Fresh Install

1. Flash SD Card

Use Raspberry Pi Imager with Raspberry Pi OS 64-bit Lite.

Configure:

• Hostname: picar
• Username: pi / Password: leon
• WiFi: Your network
• Enable SSH

2. Clone and Setup

ssh pi@picar.local
cd ~
git clone https://github.com/maxwraae/picar-brain.git
cd picar-brain
./setup.sh

3. Add API Keys

cp keys.example.py keys.py
nano keys.py

Add OpenAI API key and Picovoice access key.

4. Test and Reboot

./test.sh
sudo reboot

Files

File	Purpose
voice_assistant.py	Voice assistant (socket client)
app_control.py	Phone app + socket server
exploration.py	Autonomous exploration (disabled)
memory.py	Conversation memory
keys.py	API keys (not in git)

Troubleshooting

Check logs:

tail -f ~/picar-brain/voice.log
journalctl -u voice -n 50

Test socket:

echo '{"action":"stop"}' | nc localhost 5555

Audio not working:

# Check Pipewire is disabled
systemctl --user status pipewire

# Test speaker directly
aplay -D plughw:1,0 /usr/share/sounds/alsa/Front_Center.wav

Git pull conflict:

git fetch origin && git reset --hard origin/main