I don’t usually do press releases, but as a big fan of M5Stack dev boards, I’d like to share a quick update. You probably read my articles about their M5Paper and Core devices, therefore you understand the foundation for my love. Anything that is ESP32 based and this polished is likely to get on my Radar. This time the latest offering steps away from the usual platform in favour of a Cortex-A7 chip.
M5Stack UnitV2
You can now read the review and see it in action in this post, I also took the M5Stack UnitV2 for a drive to see how it would handle “AutoPilot” challenge.
The team had been busy working on their camera module and the M5Stack UnitV2 is coming out soon. I hope to get my hands on one for a couple of reasons. M5Stack UnitV2 is a new take on AI and Edge computing using vision.
Cortex-A7 enabled dev board comes with pretty beefy for its size stats:
- SigmaStar SSD202D
- ARM Cortex-A7 Dual-Core 1.2GHz with Neon and FPU
- Embedded 128MB DDR3 Memory
- 512MB NAND Flash
- GC2145 1080P Colored Sensor
- Microphone, TF Card slot, UART port, USB-C
- Wi-Fi 2.4GHz
- Cooling Fan
- OS: Linux
All this wrapped into a form factor similar to the M5Stick series, with a built-in camera and battery for the freedom of use.
For some time now, I wanted to dive into AI image processing and M5Stack always found a way to make it easy and fun. Take a look at this promo video. It shows how streamlined the image processing is. Tinkering with M5Stack UnitV2 can be as simple as pointing the camera at something, selecting the operational mode on associated software and reading JSON-formatted information spit out by AI.
It feels like a perfect dive into the world of vision, image processing and Edge computing. Priced around $70 won’t break your bank either (AI and Vision boards are usually pricy!). I’m full of hope that just like other M5Stack products, M5Stack UnitV2 will deliver on quality and software support. With the promise of WiFi operation and wireless programming, M5Stack UnitV2 looks like an interesting take on AI, Edge and vision-based projects.
If you want to find out more about it, visit their page for more details and be sure to follow me on one of the social platforms to get updates when I’ll get to play with one myself. I already have an evil plan hatching up involving rodents and birds. Watch this space! Have you got any comments? Leave it in this Reddit thread.
