This article focuses on the research on the sensing algorithm and learning state recognition system of classroom intelligent monitoring robot. The purpose of this study is to propose an efficient system to help improve the teaching quality. Firstly, this article expounds the basis of the sensing algorithm of intelligent monitoring robot in class, including hardware composition and related theories, and shows the corresponding relationship between hardware equipment and sensing data. Then, the DL (Deep Learning) algorithm model which combines CNN (Convolutional Neural Network) and LSTM (Long-term and Short-term Memory Network) is introduced in detail. The model inputs multimodal data, and the image is processed by CNN, the voice is processed by LSTM, and the features are fused and classified. The performance of the model is verified by experiments, and the data is collected from the real classroom. After pretreatment, it is divided into training set and test set for training and testing. The results show that the accuracy is improved and the loss function value is decreased in model training, and the accuracy, recall and F1 value are considerable in different learning state categories, such as "concentration" state accuracy of 0.85, recall of 0.88 and F1 value of 0.86. The results show that the fusion model is effective and reliable, and can provide powerful intelligent assistance for classroom teaching.
Gui et al. (Sun,) studied this question.