The board allows users to evaluate the company’s Imagimob Studio, deploy ML models and test Edge AI applications. (Image credit: Infineon Technologies)
The AI boom has given rise to task-specific hardware capable of handling all the heavy loads that applications such as machine learning and Edge computing require. Manufacturers are keeping up with demand to get the latest AI-processing hardware to developers looking to create the next generation of AI applications. To that end, Infineon Technologies recently released its PSoC 6 AI Evaluation Kit (CY8CKIT-062S2-AI), which is designed for Edge AI and Machine Learning (ML) applications.
The kit allows users to create and test their applications, evaluate the company’s Imagimob Studio AI platform, deploy ML models and test Edge applications. The SoC comes equipped with a host of sensors, including a DSP368 barometric pressure sensor, BMI270 IMU (Inertial Measurement Unit), magnetometer, a XENSIV MEMS microphone for sound detection and an ENSIV (60 GHz) radar sensor for object detection, counting, tracking, gesture sensing, and motion detection.
On the hardware end, the PSoC 6 comes outfitted with Infineon’s 62S2 MCU that features a dual-core Arm Cortex-M4F and Cortex-M0+ processors with up to 1 MB Flash and 288 KB SRAM. Connectivity is handled by a Murata Electronics LBEE5KL1YN-814 module with integrated Wi-Fi 4 (802.11b/g/n) and Bluetooth 5.2 BR/EDR/LE with speeds up to 65Mbps (Wi-Fi) and 3Mbps (Bluetooth) respectively. Interfaces include a USB Type-C port (power/data), a pair of Pmod connectors for expansion, 5x CapSense buttons and 2x CapSense sliders and Arduino Uno R3 headers for add-on boards.
The PSoC 6 also comes heavy with security features, including a hardware-based RoT (Root of Trust), cryptographic accelerators, secure onboarding (boot, key storage, and firmware updates) and trusted FW-M security services. It also comes equipped with a KitProg3 onboard debugger and programmer, a pair of user LEDs, a user button, and a reset button.
Infineon’s ModusToolbox platform provides a collection of development tools, libraries and runtime assets to help develop AI Edge and ML applications and deploy them on the PSoC 6 MCU. (Image credit: Infineon Technologies)
The evaluation kit comes with ready-to-use ML models, making it easy for users to become familiar with AI and to develop applications tailored to specific needs. It also comes with Infineon’s ModusToolbox, a collection of development tools, libraries, and embedded runtime assets to make development easier and more efficient. The toolbox includes the company’s Imagimob AI platform for creating machine-learning models, which provides an end-to-end machine learning toolchain that makes it a bit easier to create production-grade ML models for a wide range of use cases, including intelligent consumer, smart home, and IoT applications.
A few lucky Element 14 readers have a chance to roadtest Infineon’s PSoC 6 AI Evaluation Kit and write a review based on a series of tests. For each test, reviewers will need to provide text, images, videos, and any other media so that the readers will understand what is involved in using the kit and potential results. The tests are broken down as follows:
Test the out of box experience
- Working with the documentation, open the box and test the out-of-the-box experience of the Infineon PSoC 6 AI Eval Kit.
- Explore the board, and then follow the “CY8CKIT-062S2-AI Getting Started Guide” and use Imagimob Studio and the onboard sensors to collect data in preparation for creating your model. (You are not required to create a model.)
- Flash one of the Imagimob Studio ready models.
Develop code using ModusToolbox
- Explore how to develop code with the PSoC 6 AI kit.
- Explore available code examples (from “hello world” to connectivity and ML examples, and more).
- Debug development code directly with the KitProg3 on-board debugger.
Learn ML models on the PSoC 6 can be easily connected to the cloud with Avnet’s IoTConnect (Optional)
- By following a step-by-step guide on GitHub, you can create an interactive dashboard leveraging the Imagimob Sound Classification model without any software development! To discover how this experience was made, check out the IoTConnect libraries in ModusToolbox and start using them today with Project Creator!
- If applicable, your roadtest can include the use of Avnet’s IoTConnect software by either using IoTConnect libraries within ModusToolbox or using a pre-build binary made available through GitHub from Avnet.
- Test out the IoTConnect dashboards for easy visualization of the Imagimob-ready models.
Those who decide to participate in the roadtest will receive a PSoC 6 AI Evaluation Kit, ModusToolbox software (v3.2 or later), Imagimob Studio and Imagimob Ready Models and Avnet’s IoTConnect for the ModusToolbox and framework. More information on Element 14’s PSoC 6 Evaluation Kit road test can be found here.
Have a story tip? Message me at: http://twitter.com/Cabe_Atwell
