Visual tactile fusion sensing technology is an integrated sensor biomimetic technology that combines visual camera image information and tactile sensor perception information of objects and environments, inspired by the coordinated regulation of human visual and tactile movements and behaviors. With the development of computer vision, neural networks, and tactile perception technology, the introduction of visual touch fusion sensing technology in the control and working process of intelligent robots has gradually become a research hotspot. The basic principles and development status of visual tactile sensors was summarized based on different tactile sensing modes, including diffuse reflection of light rays in reflective layers, tracking and detection of marked points/objects, optical waveguides, and bionics. At the same time, the application status of visual touch fusion sensors in object recognition and detection, object slip and grasping stability detection, and robot grasping fields was summarized. Finally, the following prospects were proposed for visual touch fusion technology: 1) development of multimodal perception technology; 2) innovation in human-computer interaction methods; 3) improvement of control accuracy for intelligent robots; 4) drive the consumption of electronic products.