Deploy the CapSense Demo Project on the Kit

We’ll walk through deploying the demo project on PSoC4™ 4100S Max pioneer kit using ModusToolbox™. You can use the same procedure to deploy your own project—just follow the steps below and substitute your project folder.

1. Open ModusToolbox™> Eclipse IDE for ModusToolbox™ from the Windows Start menu. The Eclipse IDE for ModusToolbox window appears.

2. Browse and select the workspace directory for your project and click Launch to open the ModusToolbox™ workspace.

3. Select Import Existing Application In-Place from the Quick Panel or navigate to File> Import> Modus Toolbox™ Application> Import Existing Application In-Place to import the CapSense Demo Project.

4. In Project Location, click Browse to navigate Capsense-Project\data_processing and select the mtb_nn_demo_project folder and click Finish.

5. Right-click the mtb_nn_demo_project and select Build Project or click Build Application in the Quick panel to build the project. Execute “make getlibs” in MTB terminal if needed

6. Connect the Flat Flex Cable (FFC) connector (J9) on the pioneer board to the Slider (CSS1) in the expansion board the Flat flex cable. Now, connect the KitProg3 USB connector (J8) on the pioneer board to the PC or laptop using USB cable.

7. In Quick Panel> Launches, click the program. The demo project is deployed on the board.

Demonstration of Neural Network based Swipe Detection Mode

The video below demonstrates the neural-network based algorithm for detecting left and right swipe gestures running on the board. The neural network approach work reliably with similar to classic algorithm performance. The model detects swipe gestures only and not the simple touches. The neural network approach is more flexible, especially when working with non-standard sliders.

For the Slider Position Calculation and Button Touch Detection, the demo example runs both the classic algorithm and the neural-network-based algorithm in parallel.

You can view the demo using either of the following interfaces: On-board LEDs or MTB CapSense Tuner. Use the controls as follows:

• Touch the left button to turn on two LEDs: the left LED reflects the neural-network (NN) output, and the right LED reflects the classic algorithm output.
• Touch the right button to switch between slider position mode and slider gesture mode.
• In slider position mode, sliding a finger across the slider changes LED brightness: the left LED uses the NN position algorithm, and the right LED uses the classic position algorithm.
• In slider gesture mode, the LEDs indicate the swipe direction using the NN algorithm only.

In case of MTB CapSense Tuner usage, the algorithms results are shown in the Tuner User Interface:

• Two sliders representing the classic and NN position algorithms
• Two buttons representing the classic and NN touch algorithms