Microchip Edge Ai, Smart Home, Predictive Maintenance, Polarfire Fpga Vision, Sama7g54

Microchip’s new Edge AI business unit brings together a broad portfolio of 16-bit dsPIC33 Digital Signal Controllers, 32-bit Arm-based microcontrollers and microprocessors, and low-power PolarFire FPGAs into a coherent edge machine learning stack. The goal is to let designers deploy sensor analytics, computer vision and predictive maintenance workloads directly on small embedded targets instead of depending on the cloud, while still integrating with existing industrial networks and management platforms. https://www.microchip.com/en-us/solutions/technologies/machine-learning

—
HDMI® Technology is the foundation for the worldwide ecosystem of HDMI-connected devices; integrated with displays, set-top boxes, laptops, audio video receivers and other product types. Because of this global usage, manufacturers, resellers, integrators and consumers must be assured that their HDMI® products work seamlessly together and deliver the best possible performance by sourcing products from licensed HDMI Adopters or authorized resellers. For HDMI Cables, consumers can look for the official HDMI® Cable Certification Labels on packaging. Innovation continues with the latest HDMI 2.2 Specification that supports higher 96Gbps bandwidth and next-gen HDMI Fixed Rate Link technology to provide optimal audio and video for a wide range of device applications. Higher resolutions and refresh rates are supported, including up to 12K@120 and 16K@60. Additionally, more high-quality options are supported, including uncompressed full chroma formats such as 8K@60/4:4:4 and 4K@240/4:4:4 at 10-bit and 12-bit color.
—

In the smart home demo, an 11-inch HMI is driven by a SAMA7 Cortex-A7 MPU running up to 1 GHz, which is well suited for Linux-based GUIs, graphics acceleration and camera input. The system controls lights and fans, aggregates door and environmental sensors and runs a fully local face-recognition pipeline based on the open-source SCRFD detector and a Facenet-style embedding model, so authorized users can be recognized at the door without streaming frames to the cloud. The same platform can host keyword spotting, IMU-based activity recognition and other edge ML workloads using the same Cortex-A7 compute budget and camera or sensor interfaces.

For industrial users, Microchip highlights predictive maintenance on a 16-bit dsPIC33 DSC motor-control board with swappable DIM controller modules. Engineers can log current, RPM and vibration data via MPLAB Data Visualizer and then use the MPLAB Machine Learning Development Suite to train anomaly detectors or regression models that run entirely on the DSC in real time. A complementary retrofit reference design uses accelerometers, a MEMS microphone and a temperature sensor, along with a custom low-power power-management subsystem, to turn existing motors and machines into monitored assets without redesigning the drive electronics.

Both approaches emphasize an edge-first architecture where inference runs locally and connectivity is used mainly for configuration, fleet monitoring and model lifecycle management. Wired Ethernet networks of microcontroller nodes can coexist with Wi-Fi based connectivity to a cloud partner platform such as aet IoT, so operators see consolidated dashboards while devices continue to make millisecond-level decisions at the edge. This balance between local autonomy and cloud coordination is key to scaling condition-based maintenance in factories without over-provisioning bandwidth or compute in the data center.

On the vision side, a PolarFire-class FPGA demo runs face recognition and pose estimation over multiple CSI camera streams, with the option to offload further processing to an attached NVIDIA accelerator when workloads grow. Microchip has been positioning PolarFire FPGAs as low-power, deterministic fabrics for edge AI and has even introduced an Ethernet Sensor Bridge that feeds multi-protocol sensor data into NVIDIA Holoscan and Jetson platforms for real-time robotics and medical imaging workloads. ([microchip.com][4]) This interview, filmed at Embedded World North America 2025 in Anaheim, captures how those FPGA capabilities, dsPIC33 DSCs and SAMA7 MPUs are being combined into a practical edge ML roadmap spanning smart home, factory and embedded vision system.

I’m publishing about 90+ videos from Embedded World North America 2025, I upload about 4 videos per day at 5AM/11AM/5PM/11PM CET/EST. Join https://www.youtube.com/charbax/join for Early Access to all 90 videos (once they’re all queued in next few days) Check out all my Embedded World North America videos in my Embedded World playlist here: https://www.youtube.com/playlist?list=PL7xXqJFxvYvjgUpdNMBkGzEWU6YVxR8Ga

source https://www.youtube.com/watch?v=EURnUDsizGw