Many PCB vendors now provide flexible PCB manufacturing services, but Murata goes further with stretchable PCB technology that’s not only bendable but can be twisted and stretched to better fit on the body for bio-sensing and medical applications even on parts such as an elbow. Traditional bio-monitoring sensors have some limitations. For example, they can become unstuck when the body moves, damage the delicate skin of infants and the elderly, data may be distorted due to body movement, and Murata explains “there is a risk of a decline in insulation and the occurrence of ion migration when using a thermoplastic polyurethane elastomer (TPU), known as a stretchable base material, in a high humidity environment”. The company’s stretchable printed circuits ((SPC) are supposed to solve or at least mitigate all those issues. The technology is under development, but Murata still shared some highlights of the technology: Stretchable electrode printing in compliance [...]
FlexiPi is a bendable Raspberry Pi RP2040 board made of flexible PCB with the same layout as the original Raspberry Pi Pico, but featuring a USB-C port instead of a micro USB port on the official board. This follows the Flexduino flex PCB clone of the Arduino UNO made by “EDISON SCIENCE CORNER”, but the smaller design of “TOP Gadgets” FlexiPi may make it potentially more useful since it could be inserted into tight or round enclosures. FlexiPi specifications: MCU – Raspberry Pi RP2040 dual-core Cortex-M0+ microcontroller @ 48 MHz (overclockable to 133 MHz) with 264KB SRAM Storage – 2MB QSPI flash USB – 1x USB Type-C 1.1 port used for power and programming Expansion 2x 20-pin 2.54mm pitch header and castellated holes with 26 GPIOs, 3x 12-bit ADC up to 500 Kbps, 2x UART, 2x I2C, 2x SPI, 16x PWM, 2x programmable high-speed I/O 3.3V I/O voltage Sensor – [...]
PCBWay has launched its 7th project design contest in collaboration with Mouser with three categories, namely electronic project, mechanical project, and STM32 project. PCBWay’s design contests aim to encourage participants to engage in open-source innovation projects and inspire more people to join the electronics community. This year’s contest is no different and even adds the new STM32 project category to attract even more entrants. There are three phases in PCBWay’s 7th design contest: Project Release – September 2, 2024 – January 19, 2025 Project Review – January 20, 2024 – February 28, 2025 Result Announcement – March 10, 2025 The contest started last month, but you still have plenty of time to enter before January 19, 2025. As noted in the introduction three categories are available. Here are a few more details about these: Electronic project – Everything about electronic design, from simple circuits to advanced MCU and IoT projects, [...]
The world of technology is ever-changing, and young minds prove time and again that age really is just a number when it comes to mastery and innovation. One of the most interesting stories of youth combined with innovation belongs to a 9th-grade student named Akash Muthukumar, whose workshop on deploying TinyML using the XIAO ESP32S3 Sense stirred waves in both the world of machine learning and embedded systems.
A Passion for Technology
Early in his childhood, Akash’s journey in the world of technology begins. As a child, Akash was fascinated by gadgets and how they work. Middle school saw Akash deep-diving into programming, robotics, and machine learning while playing with platforms like Arduino and TensorFlow Lite. And here came his curiosity and drive to learn about TinyML-a nascent field where Akash deployed machine learning models on microcontrollers and embedded systems.
Why TinyML?
TinyML, short for Tiny Machine Learning, is a revolution within the field of artificial intelligence; it’s extending the power of machine learning into the smallest and most power-efficient kinds of devices-microcontrollers. These are just the kind of devices now being used everywhere these days in things like IoTs, where there is every need to perform intelligently locally-for instance, speech recognition, anomaly detection, and gesture recognition-without being remotely dependent on the cloud for end-to-end processing.
For Akash, the coolness factor about TinyML was its ability to take already ‘smart’ devices to the next level. The deployability of machine learning models on tiny devices opened up a world of possibilities, creating innovative projects such as real-time object detection and predictive maintenance systems.
Workshop on TinyML and XIAO ESP32S3 Sense
Akash’s workshop focused on TinyML deployment on the XIAO ESP32S3, a powerful Seeed Studio microcontroller for edge AI applications. The Xiao ESP32S3 is compact, powerful, and affordable, thus ideal for students, hobbyists, and developers interested in exploring TinyML.
Akash took participants through the whole process, from training a model to deploying it on the microcontroller. Here is a breakdown of what Akash covered:
1. Intro to TinyML Akash introduced the concepts of TinyML – what it is, why it is needed, how it works, and how it differs from normal machine learning. He noted that edge AI gets more relevant every day, and TinyML fared well in resource-constrained applications.
2. Introduction to XIAO ESP32S3 Then Akash presented the Xiao ESP32S3 Board: its features, specifications, and why it was a great platform for TinyML. Further, he presented the onboard Wi-Fi and Bluetooth capabilities, the low-power consumption, and compatibility with various sensors.
3. Building a Machine Learning Model Akash then walked them through building a machine-learning model on Edge Impulse, one of the most popular platforms for TinyML model development. Next, train your model on any simple dataset like a gesture or keyword recognition dataset.
4. Deployment of Model on XIAO ESP32S3 Deployment Process: The heart of the workshop was the deployment process. First, Akash showed how one could convert a trained model to a deployable format on the Xiao ESP32S3 using TensorFlow Lite for Microcontrollers; then, he uploaded the model onto the board and ran inferences directly on the device.
5. Real Time Demonstration The workshop concluded with a very exciting live demo: Akash showed how, in real-time, Xiao ESP32S3 was able to recognize hand gestures or detect certain sounds using the deployed TinyML model. This left the participants aghast and proved that even the most minute devices could do complex tasks using TinyML.
Empowering Next-Generation Innovators
Akash’s workshop was not limited to teaching some particular technology, but to inspire others. Being a 9th grader, he proved that none should be barred due to age factors from working on advanced fields like TinyML and can always contribute something meaningful. He made the workshop quite interactive, explaining each complex thing in an easy manner such that all participants, irrespective of age and skills, enjoyed it.
Akash has become a young leader in the tech community through his passion for teaching and deep knowledge of TinyML and embedded systems. This workshop reminded us that the future of technology rests in the hands of such young innovators who push beyond the edge of what’s possible.
Looking Ahead
And that is just where Akash Muthukumar gets started. With interests in TinyML, embedded systems, and machine learning, he definitely is going to keep making his presence known in the tech world. And as he does so, deeper into it all, it’s a dead giveaway that Akash is not only learning from the world but teaching too.
Akash’s workshop on deploying TinyML using the Xiao ESP32S3 is another good example of how the young mind has embraced technology and is showing the way. The world of TinyML is big, and with innovators like Akash at the helm, the future looks bright.
It is a story that inspires both novice and experienced developers alike to understand that with curiosity and passion in their hearts, commitments can help achieve great things even at a tender age!
Login
