I recorded ~13 minutes of training data and ~3 minutes of testing data, with 4 different types of motion classes: updown, wave, snake, and idleĬreate your impulse & classify motion data on the TI MSP432™ MCU.Select the Data acquisition tab on your Edge Impulse project dashboard and start collecting motion samples under 'Record new data':.Select the Devices tab on your Edge Impulse project dashboard and you should see your TI MSP432 board connected with a green dot.Open a command line window and run edge-impulse-data-forwarder -frequency 50, select your board and project settings in your Edge Impulse account:.Compile and flash the program onto your board.Select the RED LaunchPad w/ msp432p401r EMT (48MHz) board and port from the 'Tools' menu.Open the file get_msp432_motion_data/get_msp432_motion_data.ino in your Arduino IDE.Clone this repository locally: git clone įollow the instructions below to record motion data samples from the TI MSP432/Educational BoosterPack accelerometer and upload them directly into your Edge Impulse project.Īlternatively, you can instead use data that has already been collected from the Continuous motion recognition tutorial by cloning the Edge Impulse tutorial project into your account (select the Clone this project button on the top right).Select 'Tools' > 'Board' > 'Boards Manager.', search for and install Energia MSP432 EMT RED boards.Select 'Preferences' > 'Additional Boards Manager URLs', add the following URL.

Install the Arduino IDE - and install the TI Energia board package:.Purchase a SimpleLink™ MSP432P401R high-precision ADC LaunchPad™ Development Kit & Educational BoosterPack MKII (the BoosterPack is required to obtain access to accelerometer data).This repository contains an example for the TI MSP432™ MCU with Educational BoosterPack MKII for motion recognition with Edge Impulse. TI MSP432™ MCU Motion Recognition Demo with Edge Impulse

Msp432 tutorial