Aiming at the urgent need for human and remote unmanned devices to cooperate precisely and accurately, a remote situational awareness system of human-machine integration was proposed based on the robot operating system, and experiments and analysis were carried out. Based on visual positioning technology, with the integration of human-machine perception as the breakthrough point, through real-time 3D scene reconstruction technology and scene consistency fusion method, the environment and target information detected by unmanned equipment were 3D reconstructed. The result was consistent and fused with the human visual information, and displayed by the augmented reality device, realizing the coordinated positioning between the remote unmanned device and the augmented reality device worn by the person without GPS. The experimental results show that the system has better performance at close range. The accuracy of human-machine coordinated positioning is gradually reduced as the distance increases. The proposed system makes the unmanned device an extension of the human eye, realizing the ability to penetrate obstacles, and crosses the sight distance without interfering with the normal movement of personnel. It can play an important role in future information operations.