Abstract: In view of the covert demand for information transmission when the squad and platoon combat teams cooperate with unmanned equipment to carry out combat tasks in urban combat scenarios, a covert gesture interaction method based on surface electromyography signals is proposed, which is verified by the simulation application experiment of bomb group command and control. Specifically including, an offline and online acquisition system for surface EMG signals is developed, and an end-to-end surface EMG signal classification algorithm based on convolutional neural network is proposed, a covert gesture interaction paradigm is formed. On such basis, the command and control system of loitering munitions is constructed relying on the simulation environment, 5 participants are recruited to carry out the experiment. The experimental results show that the covert gesture interaction technology based on electromyography signals not only improves the concealment of information transmission, but also ensures the accuracy of information interaction, which is of positive significance for improving the ability of the off-duty platoon team to cooperate with unmanned equipment in urban combat scenarios.