Wild camera trap will classify images captured outdoors. The camera does not require WiFi and can be trained.
The UART port has been used to control the Bluetooth device. So the Gertbot will not work as a addon board. To resolve this you must complete two steps:
Using sudo nano - add to the /boot/config.txt file: dtoverlay=pi3-disable-bt enable_uart=1
Using sudo nano - remove "console=serial0,115200 console=tty1" from the /boot/cmdline.txt file
Now you can install the Gertbot GUI for testing the board. The picture above shows the board wired to 4 brushed motors and two 9 volt batteries.
You can also control the Gertbot with a simple python program.
Software can be uploaded to the Maix Go using Arduino ! I had a version already loaded but it did not work. I downloaded the Windows Store version using this link .
I found steps to add the Maix Boards to the IDE. The steps are well written and easy to follow. The steps can be found here.
The next step is to set up mobile net on a SD card.
Download model here: * http://dl.sipeed.com/MAIX/MaixPy/model/mobilenet_0x300000.kfpkg * Unpack it(zip format), get m.kmodel, change name to a short name "m" for example, * put it in SD card at root path
Place the SD card and reboot.
The best instructions is a video made by Sipeed:
Of course I paused quite a lot to follow the steps. I did not install the WiFi antenna . There was a nice screw driver provided for those tiny screws.
I will build a separate log for:
1. Assembly of Maix Go Kit
2. Programming Arduino with Maix Boards and compiling MobileNet for Image Processing
3. Storage of Data Collected on SD Card
4. Taking it Outside
There will also be a log on setting up the Gertbot on Raspberry Pi 3