Let's make awesome pet robots

World Map of Our Customers

Mon, 16 Mar 2026 00:00:00 +0000

Here is a world map of Maker’s Pet customer we have shipped to as of March 16, 2026.

How to Connect L298N Motor Driver to ESP32

Fri, 28 Nov 2025 00:00:00 +0000

L298N is a popular, low-cost motor driver board. Here is how you can connect L298N to ESP32 and drive a ROS2 (Kaia.ai-compatible) differential robot.

Schematic

Wiring

Refer to these wiring example illustrations.

Firmware

Use the default firmware with no changes.

Use the default config.yaml file with no changes.

Power supply

Make sure to power your L298N from a stable power supply.

For example, if your robot runs on batteries, powering your L298N directly from the battery can cause problems when the battery gets low on charge. Specifically, your robot’s motors may not be able to achieve their maximum RPM when the battery voltage is low.

To work around this problem, you may want to put a DC-DC converter between your battery and L298N to stabilize the motor power.

Bringup instructions

Follow video instructions to build and bring-up your robot.

Happy robot building!

How to View Robot Telemetry Using Foxglove

Thu, 11 Sep 2025 00:00:00 +0000

Let’s view ROS2 robot telemetry, only this time let’s use Foxglove.dev instead of ROS2 RViz. Foxglove runs outside of ROS2. For example, I run Foxglove on my Windows PC in a browser, while my ROS2 runs inside a Docker container using Windows WSL.

Launch ROS2 Docker Container

Let’s launch Kaia.ai ROS2-based Docker container. I added -p 8765:8765/tcp to the launch command in order to open port 8765 for Foxglove data streaming.

docker run --name makerspet -it -v c:\maps:/root/maps --rm -p 8888:8888/udp -p 4430:4430/tcp -p 8765:8765/tcp -e DISPLAY=host.docker.internal:0.0 -e LIBGL_ALWAYS_INDIRECT=0 kaiaai/kaiaai:iron

Install Foxglove

In your ROS2 Bash shell run the command below to install the Foxglove bridge ROS2 node.

apt update && apt install -y ros-iron-foxglove-bridge

Next, go to Foxglove.dev and install Foxglove viewer on your PC. I run Foxglove directly from my browser.

Launch Your Robot

Next, launch your ROS2 robot as usual. Please read the full instructions about setting up and operating Maker’s Arduino/ROS2 robots and Kaia.ai software here.

Launch Foxglove

In your ROS2 Bash shell launch the Foxglove ROS2 node:

ros2 launch foxglove_bridge foxglove_bridge_launch.xml port:=8765

Open your Foxglove viewer

click Open Connection to a live robot or server
select Foxglove WebSocket ws://localhost:8765
your Foxglove viewer should connect and start displaying some data
expand View on the left and unhide the /scan topic
- select a color that is more visible
click the Topics tab in the left sidebar. You should see /scan and /tf topics running.

Using Foxglove with Rosbridge

As an alternative, you can use Rosbridge ROS2 node instead of Foxglove bridge.

Both of these are ROS2 nodes and both use WebSockets. According to Foxglove, Foxglove bridge should run a bit faster compared to Rosbridge.

The key advantage of using Rosbridge is that you can use your own client with Rosbridge - instead of the Foxglove viewer.

Rosbridge uses port 9090 by default. Therefore, the Docker launch command should look like this:

docker run --name makerspet -it -v c:\maps:/root/maps --rm -p 8888:8888/udp -p 4430:4430/tcp -p 9090:9090/tcp -e DISPLAY=host.docker.internal:0.0 -e LIBGL_ALWAYS_INDIRECT=0 kaiaai/kaiaai:iron

Install Rosbridge using your ROS2 Bash shell:

apt update && apt install -y ros-iron-rosbridge-server
source /opt/ros/iron/setup.bash

Launch Rosbridge using your ROS2 Bash shell:

ros2 launch rosbridge_server rosbridge_websocket_launch.xml

Launch your Foxglove viewer

click Open Connection to a live robot or server
select Foxglove WebSocket ws://localhost:9090
your Foxglove viewer should connect and start displaying data

Happy robot building!

How to Connect YDLIDAR X3, X3PRO, X2, X2L, X4 and SCL to Maker's Pet ESP32 Boards

Mon, 28 Apr 2025 00:00:00 +0000

YDLIDAR X3, X3PRO, X2 and X2L are a low-cost 2D LiDAR 360-degree distance sensor used in robotics for navigation. Here are instructions to connect YDLIDAR X3/X3PRO and X2/X2L to the LiDAR port on Maker’s Pet BDC-30P and BDC-38P motor driver and ESP32 carrier boards.

Hardware Wiring

Maker’s Pet BDC-30P and BDC-38C4 motor driver and ESP32 carrier boards have an 8-pin LiDAR header connector. The LiDAR header can be connected to a large variety of 2D LiDARs. The LiDAR header pins include:

LiDAR TX
LiDAR RX
+5V power for LiDAR
GND ground for LiDAR
M- LiDAR’s motor negative terminal
M+ LiDAR’s motor positive terminal (connected to +5V power)
PWM for LiDAR motor speed control
EN for LiDAR enable/disable (used by YDLIDAR X4 only)

YDLIDAR X3/X3PRO and X2/X2L have a Molex PicoBlade 1.25mm 4-pin connector. The photo below shows the YDLIDAR X3/X3PRO connector pinout.

The photo below shows the YDLIDAR X2/X2L connector pinout.

For easy connection, Maker’s Pet sells a breakout cable for YDLIDAR X2, X2L, X3, X3 PRO and SCL with 2.54mm female DuPont connectors.

Make the following connections between YDLIDAR X3/X3PRO X2/X2L Lidar and BDC-30P/BDC-38C4 board:

LiDAR TX to board LiDAR header TX
LiDAR GND to board LiDAR header GND
LiDAR +5V to board LiDAR header +5V
LiDAR M_CTR/M_SCTR to board LiDAR header PWM

Firmware Configuration

Edit your Kaia.ai firmware config.yaml configuration file. In the lidar section comment out everything except your LiDAR model:

YDLIDAR X3 for YDLIDAR X3
YDLIDAR X3 PRO for YDLIDAR X3 PRO
YDLIDAR X2 for YDLIDAR X2
YDLIDAR X2L for YDLIDAR X2L

Here is reference documentation for config.yaml.

Here is an example configuring YDLIDAR X3 PRO for use.

lidar:
# model: LDROBOT LD14P
# model: NEATO XV11
# model: SLAMTEC RPLIDAR A1
# model: XIAOMI LDS02RR
# model: YDLIDAR SCL
# model: YDLIDAR X2
# model: YDLIDAR X2L
# model: YDLIDAR X3
  model: YDLIDAR X3 PRO
# model: 3IROBOTIX DELTA-2G
# model: 3IROBOTIX DELTA-2A 115200
# model: 3IROBOTIX DELTA-2A
# model: 3IROBOTIX DELTA-2B
# model: YDLIDAR X4 PRO
# model: YDLIDAR X4

Your board’s config.yaml already contains ESP32 GPIO pin mapping to the board’s LiDAR connector, including LiDAR TX, RX, PWM and EN:

For illustration purposes, here is the ESP32 GPIO pin assignment LiDAR from the BDC-30P config.yaml.

lidar:
  gpio:
    tx: 35
    rx: 27
    pwm: 13
    en: 32

Upload your edited configuration file to ESP32 as Arduino sketch data, see video.

Next, boot your firmware and check the Arduino Serial Monitor. The firmware should print your LiDAR model, e.g. LIDAR model YDLIDAR X3 PRO.

ets Jul 29 2019 12:21:46

rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0030,len:1344
load:0x40078000,len:13964
load:0x40080400,len:3600
entry 0x400805f0

Kaia.ai firmware version 0.8.0-iron
ESP IDF version v4.4.7-dirty
SPIFFS mounted successfully
/network.yaml found; loaded OK
/config.yaml found; loaded OK
To enter web config push-and-release EN, then push-and-hold BOOT within 1 sec
Board model MINI-BDC30P-BODY with BDC-30P, version v1.1.1, manufacturer makerspet.com
LIDAR model YDLIDAR X3 PRO
LIDAR RX buffer size 1024, baud rate 115200
Battery ADC attenuation 7.00, voltage 2.68V
Motor driver type IN1_IN2; motor encoder type AB_QUAD
Motor Max RPM 180.00; encoder PPR 1035.00 TPR 4140.00

Connecting to WiFi NETGEAR48 ... 

ROS2 Software Configuration

Follow this step if you are using Kaia.ai ROS2 PC software for robot mapping, navigation, SLAM and visualization. Otherwise you can skip this step.

Override the default LiDAR model when you launch your robot using the lidar_model argument, see more details:

Set the lidar_model argument as follows:

YDLIDAR-X3 for YDLIDAR X3
YDLIDAR-X3-PRO for YDLIDAR X3 PRO
YDLIDAR-X2-X2L for YDLIDAR X2 and X2L

ros2 launch kaiaai_bringup physical.launch.py robot_model:=makerspet_loki lidar_model:=YDLIDAR-X3-PRO

The full list of supported lidar_model arguments is here.

When your robot’s ESP32 connects to your local ROS2 PC, your local PC should print your LiDAR model LDS model YDLIDAR-X3-PRO.

Your YDLIDAR sensor should be powered on and spinning.

In the meantime, you should see your ESP32 firmware printing out LiDAR RPM around 4.75 or 5. This means your Lidar is spinning at 4.75 rotations per second.

connected, IP 192.168.1.153
Connecting to Micro-ROS agent 192.168.1.113 ... 

Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -60dBm
Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -59dBm
Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -61dBm
Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -62dBm

How to Connect Delta-2A, 2B and 2G LiDAR to Maker's Pet ESP32 Boards

Sun, 27 Apr 2025 00:00:00 +0000

3irobotix Delta-2A, 2B and 2G are a low-cost 2D LiDAR 360-degree distance sensor used in smart vacuum cleaners for navigation. Here are instructions to connect Delta-2G to the LiDAR port on Maker’s Pet BDC-30P and BDC-38P motor driver and ESP32 carrier boards.

Hardware Wiring

LiDAR TX
LiDAR RX
+5V power for LiDAR
GND ground for LiDAR
M- LiDAR’s motor negative terminal
M+ LiDAR’s motor positive terminal (connected to +5V power)
PWM for LiDAR motor speed control
EN for LiDAR enable/disable (used by YDLIDAR X4 only)

Some LiDARs, including 3irobotix Delta-2G, do not have a built-in LiDAR motor control. Maker’s Pet BDC-30 and BDC-38C4 have a special on-board circuit that drives the LiDAR motor directly. Kaia.ai Arduino ESP32 firmware controls that special circuit to perform real-time LiDAR motor speed control.

3irobotix Delta-2A, 2B and 2G use a JST PH 2.0mm 5-pin connector. The photo below shows the Delta-2G connector pinout and connections to Maker’s Pet BDC-30P board with a 30-pin ESP32 dev kit.

connect Delta LiDAR TX to BDC-30P LiDAR TX
connect Delta LiDAR GND to BDC-30P LiDAR GND
connect Delta LiDAR MOT- to BDC-30P LiDAR M-
connect Delta LiDAR +5V to BDC-30P LiDAR +5V
connect Delta LiDAR +5V/MOT+ to BDC-30P LiDAR +5V

Idential wiring applies to the Maker’s Pet BDC-38P board with a 38-pin ESP32 dev kit as well.

Optionally, put a resistor in series with TX because Delta-2G TX outputs 5V signal, not 3.3V. A 100 Ohm to 10k resistor should work.

For easy connection, Maker’s Pet sells a breakout cable for 3irobotix Delta-2A and 2G with 2.54mm female DuPont connectors.

Here are more high-resolution photos of the wiring.

Firmware Configuration

Edit your Kaia.ai firmware config.yaml configuration file. In the lidar section comment out everything except your LiDAR model:

3IROBOTIX DELTA-2A for Delta-2A
3IROBOTIX DELTA-2A for Delta-2A that runs at 115200 baud (I’m not 100% sure of this LiDAR model name)
3IROBOTIX DELTA-2B for Delta-2B
3IROBOTIX DELTA-2G for Delta-2G

Here is reference documentation for config.yaml.

lidar:
# model: LDROBOT LD14P
# model: NEATO XV11
# model: SLAMTEC RPLIDAR A1
# model: XIAOMI LDS02RR
# model: YDLIDAR SCL
# model: YDLIDAR X2
# model: YDLIDAR X2L
# model: YDLIDAR X3
# model: YDLIDAR X3 PRO
  model: 3IROBOTIX DELTA-2G
# model: 3IROBOTIX DELTA-2A 115200
# model: 3IROBOTIX DELTA-2A
# model: 3IROBOTIX DELTA-2B
# model: YDLIDAR X4 PRO
# model: YDLIDAR X4

Your board’s config.yaml already contains ESP32 GPIO pin mapping to the board’s LiDAR connector, including LiDAR TX, RX, PWM and EN:

For illustration purposes, here is the ESP32 GPIO pin assignment LiDAR from the BDC-30P config.yaml.

lidar:
  gpio:
    tx: 35
    rx: 27
    pwm: 13
    en: 32

Upload your edited configuration file to ESP32 as Arduino sketch data, see video.

Next, boot your firmware and check the Arduino Serial Monitor. The firmware should print LIDAR model 3IROBOTIX DELTA-2G.

ets Jul 29 2019 12:21:46

rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0030,len:1344
load:0x40078000,len:13964
load:0x40080400,len:3600
entry 0x400805f0

Kaia.ai firmware version 0.8.0-iron
ESP IDF version v4.4.7-dirty
SPIFFS mounted successfully
/network.yaml found; loaded OK
/config.yaml found; loaded OK
To enter web config push-and-release EN, then push-and-hold BOOT within 1 sec
Board model MINI-BDC30P-BODY with BDC-30P, version v1.1.1, manufacturer makerspet.com
LIDAR model 3IROBOTIX DELTA-2G
LIDAR RX buffer size 1024, baud rate 115200
Battery ADC attenuation 7.00, voltage 2.68V
Motor driver type IN1_IN2; motor encoder type AB_QUAD
Motor Max RPM 180.00; encoder PPR 1035.00 TPR 4140.00

Connecting to WiFi NETGEAR48 ... 

ROS2 Software Configuration

Follow this step if you are using Kaia.ai ROS2 PC software for robot mapping, navigation, SLAM and visualization. Otherwise you can skip this step.

Override the default LiDAR model when you launch your robot using the lidar_model argument, see more details:

Set the lidar_model argument as follows:

3IROBOTIX-DELTA-2A for 3irobotix Delta-2A and the 115200 baud Delta-2A
3IROBOTIX-DELTA-2B for 3irobotix Delta-2B
3IROBOTIX-DELTA-2G for 3irobotix Delta-2G

ros2 launch kaiaai_bringup physical.launch.py robot_model:=makerspet_loki lidar_model:=YDLIDAR-X3-PRO

The full list of supported lidar_model arguments is here.

When your robot’s ESP32 connects to your local ROS2 PC, your local PC should print your LiDAR model LDS model 3IROBOTIX-DELTA-2G.

Your Delta LiDAR should be powered on and spinning.

In the meantime, you should see your ESP32 firmware printing out LiDAR RPM around 4.75 or 5. This means your Lidar is spinning at 4.75 rotations per second.

connected, IP 192.168.1.153
Connecting to Micro-ROS agent 192.168.1.113 ... 

Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -60dBm
Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -59dBm
Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -61dBm
Telem avg 46 max 52ms, LiDAR RPM 4.75, wheels RPM 0.00 0.00, battery 8.00V, RSSI -62dBm

How to Design, 3D Print Bearings and Save Thousands of Dollars

Tue, 18 Mar 2025 00:00:00 +0000

I’m building a robot that somewhat resembles R2D2 - a round body with a spinning semi-sphere head - NOT a clone for copyright reasons and deep respect to the filmmakers! The robot’s spinning head needs a large, thin bearing.

I have tried searching for a bearing like that to purchase - those are not easy to find, costing around $1000! Therefore, I have looked into making a DIY bearing like that. Thankfully, this bearing would not have to carry much weight or move around much. Therefore, this type of bearing can be 3D printed.

Long story short, I have designed and 3D printed a small 60mm OD bearing as a starter prototype - and it worked great! So, I leveled up to 190mm OD, which also worked well. Finally, I have printed a 340mm (13.3 inch) OD bearing on my DIY Voron 2.4.

Here is a YouTube reel illustrating the concept.

And here is a complete from-scratch tutorial for absolute beginners showing step-by-step how to design a radial bearing in Autodesk Fusion 360 (free for personal use as of the time of the writing), 3D printed on my glorious Voron 2.4 350mm (and my old reliable Prusa MK3S) and assembled by yours truly using 1/4” steel bearing balls I purchased on Amazon.

After posting this video on Reddit and Facebook I got (harsh) negative replies saying that 3D printed plastic bearings are of no use. I do beg to disagree. After extensive testing, I can recommend 3D printed bearings for use in light-duty applications including pan-and-tilt hardware - cameras, lights and R2D2-style robot heads. More specifically, the larger your light-duty bearing is, the more money you can save by 3D printing it istead of purchasing one (made from steel). Moreover, thanks to 3D printing you can make your large bearing of just about any custom size you want - as long as the bearing fits your 3D printer.

I printed my bearings usint PETG plastic. If you want to level-up from very-light-duty to something more durable, consider using Nylon, Nylon reinforced with glass beads or Delrin (POM-H) self-lubricating plastic filament.

Last, but not the least. There are radial bearings and there are axial load bearings. My DIY bearing tutorial shows the former, while your specific use case may need the latter.

Happy 3D printing!

Build an Arduino Self-Driving Robot - Video Instructions Series

Sun, 09 Mar 2025 00:00:00 +0000

Here is my video instructions series with step-by-step explanations “How to Build an Arduino/ROS2 Self-driving Robot”.

There are five video lessons:

Self-driving robot assembly video
Software and firmware setup video
Robot bring-up video
Create a map of your place video (manual driving)
Robot self-drives video - create a map, navigate your place automatically
Software, firmware install and bring-up on Ubuntu PC
Finished self-driving robot

Happy learning and building!

How to Connect TB6612FNG Motor Driver to Your ESP32 DIY Robot

Thu, 06 Mar 2025 00:00:00 +0000

TB6612FNG is a low-cost driver IC for brushed DC motors - popular with DIY enthusiasts to build small wheeled robots. One TB6612FNG can drive two motors with power supply voltage up to 15V, average output current up to 1.2A per motor and up to 3.2A peak current. I’ve used TB6612FNG it on many occasions to drive small, low-cost N20-size motors.

Hardware Wiring

Here is how you can connect a TB6612FNG motor driver IC to ESP32 to drive a couple of small brushed DC motors.

Kaia.ai firmware aims works with all kinds of boards and drivers, so you can reuse your components.

Here is a schematic connecting TB6612FNG to ESP32 30-pin dev board.

In the schematic

ESP32 GPIOs (configured as outputs) connect to TB6612FNG that drives two (left and right) DC brushed motors.
ESP32 GPIOs (configured as inputs) also connect to (quadrature) motor encoders. Each quadrature motor encoder has 2 signals called A and B (or C1, C2 or something like that).
resistors in series with GPIO inputs protect ESP32 from 5V encoder signal levels

The choice of GPIO numbers to connect TB6612FNG and encoders is up to you - subject to ESP32 GPIO own limitations. Read more about ESP32 GPIO limitations:

Firmware Configuration

Kaia.ai firmware has a special config.yaml file where you can specify your GPIO assignments. Kaia.ai firmware loads this file at boot time. Please see more config.yaml documentation here.

Edit config.yaml for TB6612FNG as follows:

set motor.driver.type to IN1_IN2
assuming TB6612FNG AO1 and AO2 outputs drive the left motor
- set motor.driver.left.gpio.in1 to GPIO number driving TB6612FNG AIN1 input
- set motor.driver.left.gpio.in2 to GPIO number driving TB6612FNG AIN2 input
assuming TB6612FNG BO1 and BO2 outputs drive the right motor
- set motor.driver.right.gpio.in1 to GPIO number driving TB6612FNG BIN1 input
- set motor.driver.right.gpio.in2 to GPIO number driving TB6612FNG BIN2 input

Upload the edited config.yaml to ESP32 as sketch data, see video.

Here is an example of TB6612FNG setup in config.yaml:

motor:
  driver:
    type: IN1_IN2
  left:
    driver:
      gpio:
        in1: 26
        in2: 25
  right:
    driver:
      gpio:
        in1: 23
        in2: 22

100K+ Members in My Facebook Group

Thu, 06 Mar 2025 00:00:00 +0000

I’ve been running several Facebook groups related to Arduino, robotics, DIY and programming. Today my Arduino and Robotics group has reached over 100K members 🥳🎉.

Also, my other Facebook group Arduino and DIY Robots that focuses on DIY robots in a more narrow fashion.

Please visit and/or join my groups to

learn Arduino and robotics and find free popular learning resources
find a project you like
get inspiration, motivation for your project
see what’s happening in Arduino and robotics around the world

Online Robotics Store is Open

Sun, 23 Feb 2025 00:00:00 +0000

We’re excited to announce the launch of our online store, where innovation meets education! Whether you’re a hobbyist, educator, or robotics enthusiast, our store offers educational robots designed to make learning fun and interactive.

What We Offer

DIY Kits – Assemble your own robot and learn hands-on.
Robot Parts – Including motor driver boards, LiDAR sensor accessories, and more.
TODO Ready-to-Use Robots – Pre-built and ready to go!

Our robots are built using Arduino, ROS2, micro-ROS, and ESP32, ensuring flexibility and expandability. We provide firmware and software to help you get started.

Why Choose Us?

Step-by-Step Instructional Videos – Learn with ease.
Open-Source Products – Modify and enhance as you like.
Worldwide Shipping (*) – Bringing robotics to you, no matter where you are!

Join us on this journey and start exploring the world of educational robotics today! 🚀

Visit Our Store Now

(*) see the Shipping Policy for exceptions