Skip to main content

Deploying AI Models on SenseCAP A1102 Using SenseCraft AI

Overview​

The SenseCAP A1102 is a powerful sensor device that combines advanced AI capabilities with ease of deployment. This guide will walk you through deployingofficial or custom AI models on the A1102 using the SenseCraft AI website, offering step-by-step instructions for pairing, configuration, and testing.

Prerequisites​

Required Hardware​

  • SenseCAP A1102 device
  • USB Type-C data cable
  • PC

Required Software​

AI Model Flashing Process​

Step 1. Open our platform and Select A1102 in Workspace​

Step 2. Connect to our Platform by serial port​

  • Use a Type-C cable to connect the device to your PC (make sure you are connecting to the upper Type-C port, which is the Himax serial port, as shown in the image below).
  • Click Connect
  • Select correct serial port and click Connect in the pop-up window

If the correct port is selected, you will be able to view the preview video stream through Invoke and proceed with selecting and flashing the model normally.

Step 3. Select the AI model you want to deploy to device​

  • Click Select Model

  • Choose from Public Models or your own trained models in SenseCraft AI

  • The flashing process will begin after you click the model

Upload your own AI model and deploy it to device​

note

Our AI model transmission is based on RS485, and it supports up to 9 recognition classes only.

  • Click Upload Model
  • Configure your model information

  • Model File: For detailed instructions on training and exporting a TFLite model, please refer to this guide

  • Object: The category name corresponding to the model’s label order (from 0 to 8). Only up to nine classes are supported — any additional classes will not be displayed.

  • Click Send and wait for flashing process

Result Analysis​

  • Each result contains 10 measurements, which correspond to the 10 RS485 channels:

  • The first nine measurements represent the confidence scores (measurementValue) of each class index (from 0 to 8).

  • The tenth measurement carries the model information, including metadata related to the deployed AI model.

Tech Support & Product Discussion​

Thank you for choosing our products! We are here to provide you with different support to ensure that your experience with our products is as smooth as possible. We offer several communication channels to cater to different preferences and needs.

Loading Comments...