When it comes to interesting robotics projects that showcase innovation, our community at Petoi leads the way! From blending advanced automotive tech architecture with Petoi Bittle to designing custom code to move the robots, from building your own voice command AI robot pet to breaking the boundaries in autonomous movement. In this article, we will take a look at a few robotics projects created by our community that shed light on the fusion between sophisticated quadruped robotics applications and Raspberry Pi running on Petoi open source quadruped robotics framework and hardware.
Run And Control A Petoi Bittle Stem Robot On SOAFEE Open-source Automotive Architecture With Raspberry Pi
Starting things off we have GitLab Field Chief Cloud Architect Darwin Sanoy. His amazing robotics Raspberry Pi project Petoi Bittle Robot Dog Running SOAFEE(Scalable Open Architecture for the Embedded Edge) is at the intersection of STEM education and Software Defined Vehicle architecture with smart robot that acts like a test prototype for the automotive industry.
First-time embedded systems developers enjoy seeing their code interact with the real world, but this experience is limited in software-defined vehicle development due to the impracticality of having a car at every desk. To address this, the idea of using robotics in STEM education as a substitute for cars in automotive development is being explored. Darwin and his collogue James Moverley have developed an initial solution using a Raspberry Pi 5 with an Arm processor and k3s by SUSE's Rancher, managed by GitLab's AgentK. This setup enables the beginnings of the SOAFEE open-source automotive architecture to run on and control a Petoi Bittle STEM coding robot.
We're lucky to have Darwin document his journey on GitHub and share everything with us through a fantastic informative video! He also posts regular updates on his Linkedin profile like this.
Here is the link to the video: Darwin's AI Robot project shaping the future of the automotive tech education!
Bittle Robot Dog Teleoperation with ROS and Raspberry Pi: Custom Driver, Using Twist Messages
We take a quick trip down memory lane, taking a look at the fantastic work done by Dmitry Maslov on his Youtube channel Hardware.ai. Back in 2020, when Petoi Bittle robot dog was still on kickstarter, Dmitry combined Petoi Bittle robotics kit with ROS (Robotic Operating System) and explained 2 options on how to write custom drivers for ROS to interact with the robot's hardware and control Petoi Bittle's movements.
Dmitry's first option was using Bittle coding robot dog's existing microcontroller and running an ROS controller directly on this chip. The second option also dealt with a more advanced method where Dmitry used a Raspberry Pi to communicate with the microcontroller.
This in depth article on the Hardware.ai article gives us an insight on both of these options and how to start writing custom code. You can also watch the video "Bittle Teleoperation with ROS and Raspberry Pi: Custom driver, using Twist messages" to actually see this in action:
Build Your Own Alexa-like Voice-controlled Using Raspberry Pi
For our third robotics project, we can take a look at how Friende Peng, a summer intern at Petoi, built his own Alexa-like voice-controlled robot dog with AI! This video demos the voice control capability with a small speech recognition model and Voice Activity Detection algorithm running on a Raspberry Pi. By going through the open source robotics project code repository for this, you will learn about the tradeoff and experiments Friende made to enable speech recognition and voice command capability on a small robot.
It's a valuable lesson for anyone that develops robotics AI application under physical constraints.
Friende also did a voice control robot cat demo with Petoi open source quadruped robot cat Nybble:
Did you know: Petoi Bittle X robotics kit also comes with in-built voice control? Find Bittle X here!
Explore Various Ways To Run the Camera Module On Bittle To Adjust Its Behaviors
Simon from Malta experimented with Petoi Bittle as he was reviewing its capabilities. Simon, who has a Master degree in AI, used Bittle coding robot for this AI project. In a 3 part blog series which Simon documented, he wrote about his full interaction with the smart robot dog and also explored the scope of the camera and movement tracking with the Intelligent Camera Module. He was amazed to see how the Petoi Bittle not only recognized him but also tracked his movements (just like you can see in the video above).
After exploring the tracking behavior of Bittle, naturally, Simon wanted to continue his research and see how he could run an ML model and begin to print some kind of output from the camera. He went through a system design exercise, theorizing different ways he could potentially leverage the camera to adjust the Bittle’s behaviour and the various trade-offs each setup would have, including control everything locally on a Raspberry Pi. You can read more about his findings in this blog here.
Bittle Robot Dog Sees and Walks Toward Me
Bittle Robot Dog Autonomous Movement & Object Detection with Raspberry Pi
Want to experiment with artificial intelligence and build your own AI Robotics projects? Get your own Petoi Bittle robot dog for Raspberry Pi quadruped application development!
