With the development and application of intelligent nervous system, intelligent control system represented by machine vision has been promoted and used in various fields, including manipulator control system. Mechanical arm is the main actuator of robot and other mechanical components, but the application environment of mechanical arm is extremely complex, and with the development of industrial process, the sensitivity of technology and operating system for its arbitrary posture operation is also steadily improved. Based on the above problems, this paper designs a manipulator control system based on machine vision theory and orthogonal parallel six-dimensional force sensor to meet the working requirements of manipulator in high-precision environment. Systems hardware are consisted by machine vision structure and six dimensions of force sensor. The practical application effect of the model was verified by collision detection experiment. The relevant research can provide theoretical guidance and practical reference for the research of manipulator control field. Dance and practical application reference for the research of manipulator control field.