Abstract

This paper proposes a method of vehicle in-cabin occupancy detection using millimeter-wave (mmWave) radar and machine learning scheme. We develop a simple approximate model of extreme gradient boosting (XGBoost) which improves explainability and simplicity of the method. XGBoost is a model composed of a huge number of decision trees, which can be approximated into a single decision tree unlike neural network (NN). The proposed simple model can easily be implemented on a low-cost firmware which is installed with radar modules. The performance of the proposed model is evaluated in comparison with NN and XGBoost-based model.

Pages (from-to)	1-6
Publication status	Published - 2023