KIDA-Robot-v01

KIDA: Kinetic Interactive Drive Automaton

https://github.com/CursedPrograms/WHIP-Robot-v00
https://github.com/CursedPrograms/NORA-Robot-v00
https://github.com/CursedPrograms/KIDA-Robot-v00
https://github.com/CursedPrograms/ComCentre

Overview

KIDA is an advanced autonomous robot platform built on Raspberry Pi 5, featuring dual-camera vision (night vision + AI camera), a Hailo-8 AI accelerator (13 TOPS), and local LLM capabilities. The robot supports voice interaction, autonomous navigation, and real-time AI inference.

Hardware Specifications

Computing

Chassis & Motion

Power System

Sensors & Cameras

Audio

Software Stack

Operating System

Recommended: Raspberry Pi OS (Bookworm)
Kernel: 6.12.62+rpt-rpi-2712

AI & ML

| Component | Technology | |———–|————| | LLM | Ollama (DeepSeek-R1:1.5b, Gemma3:4b) | | STT | OpenAI Whisper | | TTS | Piper / ElevenLabs API | | Vision | Hailo-8 AI Processor, IMX500 |

Python Requirements

playsound
openai-whisper
sounddevice
numpy
whisper
SpeechRecognition
pygame
requests
elevenlabs==0.2.26
torch
torchaudio

GPIO / Wiring Reference

Motor A (Left)

Motor B (Right)

| L298N Pin | Function | Pi GPIO | |———–|———-|———| | IN3 | Direction | GPIO 22 | | IN4 | Direction | GPIO 23 |

Power Schematic

[12V Battery Pack — 3S 21700 @ 3.7V each]
 ├── + ─────────► L298N VS        (motor power input)
 ├── + ─────────► LM2596S IN+     (step-down input for Pi)
 ├── – ─────────► L298N GND
 └── – ─────────► LM2596S IN–

[LM2596S Output]
 ├── OUT+ ──────► Pi 5V  (GPIO pin 2 — or Pi UPS via USB-C [Recommended])
 └── OUT– ──────► Pi GND (GPIO pin 6 or 9)

Installation

1. Clone & Set Up Python Environment

sudo apt update
sudo apt install python3-venv python3-pip

python3 -m venv ~/kida-venv
source ~/kida-venv/bin/activate

pip install --upgrade pip
pip install -r requirements.txt

2. Install OpenAI Whisper

pip install git+https://github.com/openai/whisper.git

3. Install Audio Dependencies

sudo apt install pulseaudio jackd2 alsa-utils portaudio19-dev python3-pyaudio

4. Install Piper TTS

pip install piper-tts

5. Install ElevenLabs (optional)

pip install git+https://github.com/elevenlabs/elevenlabs-python@v3

6. Install & Set Up Hailo

sudo apt update
sudo apt install hailo-all

Verify the installation:

hailortcli fw-control identify

Expected output:

Device: Hailo-8
PCIe Address: 0001:03:00.0
Firmware Version: x.x.x

Troubleshooting: If the driver is not detected after install, a DKMS module may have failed to build. Pin back to version 4.19 which has stable kernel module support:
sudo apt-mark hold linux-image-rpi-2712 linux-headers-rpi-2712
sudo apt install hailort=4.19.0 hailo-all=4.19.0 -y
sudo apt-mark hold hailort hailo-all
sudo reboot

Manual Hailo Build (if package unavailable)

sudo apt install -y git build-essential cmake python3-dev python3-pip

git clone https://github.com/hailo-ai/hailort.git
cd hailort && mkdir build && cd build
cmake .. && make -j$(nproc) && sudo make install

echo "export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH" >> ~/.bashrc
source ~/.bashrc
pip install hailort

Running KIDA

Activate the virtual environment

cd /home/kida-01/Desktop/Kida-Robot
source venv/bin/activate

Run the main script

python scripts/main.py

Run LLM

ollama run deepseek-r1:1.5b
# or
ollama pull gemma3:4b-it-qat

Test cameras

rpicam-hello --list-cameras

# AI Camera inference (camera slot 1)
rpicam-hello --camera 1 -t 0 \
  --post-process-file /usr/share/rpi-camera-assets/imx500_mobilenet_ssd.json \
  --viewfinder-width 1920 \
  --viewfinder-height 1080 \
  --framerate 30

Autostart on Boot

Option A: systemd service (recommended)

sudo nano /etc/systemd/system/kida.service

[Unit]
Description=KIDA Main Controller
After=graphical.target

[Service]
User=kida-01
WorkingDirectory=/home/kida-01/Desktop/Kida-Robot
ExecStart=/home/kida-01/kida-venv/bin/python scripts/main.py
Restart=always
RestartSec=5

[Install]
WantedBy=graphical.target

sudo systemctl daemon-reload
sudo systemctl enable kida.service
sudo systemctl start kida.service

Option B: Camera preview service

sudo nano /etc/systemd/system/kida-camera.service

[Unit]
Description=KIDA Camera Live Preview
After=graphical.target

[Service]
User=kida-01
Environment=DISPLAY=:0
ExecStart=/usr/bin/python3 /home/kida-01/Desktop/Kida-Robot/scripts/camera_preview.py
Restart=always
RestartSec=5

[Install]
WantedBy=graphical.target

sudo systemctl daemon-reload
sudo systemctl enable kida-camera.service
sudo systemctl start kida-camera.service

Check status / logs:

sudo systemctl status kida-camera.service
journalctl -u kida-camera.service -f

Option C: crontab

crontab -e
# Add:
@reboot cd /home/kida-01/Desktop/Kida-Robot && python3 main.py &

Option D: Desktop autostart

mkdir -p ~/.config/autostart
nano ~/.config/autostart/kida.desktop

[Desktop Entry]
Name=KIDA Controller
Exec=python3 /home/kida-01/Desktop/Kida-Robot/main.py
Type=Application
X-GNOME-Autostart-enabled=true

RTC Setup

dtparam=rtc_bbat_vchg=3000000
sudo mount -o remount,rw /boot/firmware
sudo nano /boot/firmware/config.txt
sudo hwclock -w
sudo hwclock -v -r

FFmpeg Setup (Windows)

Download the latest static build from: https://www.gyan.dev/ffmpeg/builds/
Extract and rename the folder to ffmpeg, placed at C:\ffmpeg

Ensure this structure exists:

C:\ffmpeg\bin\
├── ffmpeg.exe
├── ffplay.exe
└── ffprobe.exe

Add C:\ffmpeg\bin to your system PATH via System Properties → Environment Variables
Verify: open a new terminal and run ffmpeg -version

Diagnostics

# Check Hailo kernel module
lsmod | grep hailo

# Identify Hailo device
hailortcli fw-control identify

# Check PCI devices
lspci | grep Hailo

# Kill conflicting camera processes
ps aux | grep -E 'libcamera|picamera'
sudo kill -9 <PID>

# Check which process is using the camera
sudo fuser -v /dev/video0

# Kernel version
uname -r

This site is open source. Improve this page.