Design And Implementation Of Robot Guidance System

Modern face recognition system for identifying positive face efficient, but in the face inside a three-dimensional space, in addition to the positive face there are a variety of scenarios. Human face location, orientation and rotation in the image and video source have uncontrollable characteristics, these realistic conditions greatly increased the difficulty of recognition, and the identification of these scenarios has not yet a very effective method. The problem of this article will to solve is side face recognition accuracy in yaw direction of the case and tracking guidance based on the identification.Key technologies of Robot guidance system need to be solved was the face recognition base on direction, the relative position calculation, motion planning and so on. With some of the results of previous studies, based on optimization of the algorithm of face recognition base on direction, then established use Dlib+Caffe algorithm, at the same time ensuring final recognition rate is over 90%, with an accuracy of two decimal places; based on the robot forward kinematics, provided with the camera parameter matrix and relevant face information to complete the calculation of the relative position; translation and rotation based on coordinate transformation, combining the robot coordinate system with the data returned by the code wheel, then to complete the motion planning design.Actual verification results show that the paper proposed based on side face recognition yaw direction of the case reached 92%, effectively solve the original only being efficient face recognition, and to build on the support side face recognition robot guidance system, the system ultimately face recognition base on direction, motion planning and calculation of the relative position module, etc. In addition to meeting the identification of efficient, accurate, also has a module structure, security and reliability features.The entire contents of the paper scheduled to start environmental studies to lead to problems in order to identify problems, and then from the point of view to solve the problem, carried out preferably of the algorithm of face recognition base on direction to the analysis and definition of system requirements, system architecture design, the overall process design, design and implementation of functional modules to complete building the entire system, effective support solutions. Finally, the system validation and optimization, to determine the effectiveness of the solution, obtained the full text of the research results, at the same time to be promoted for some of the ideas proposed in this paper at the follow-up to be explored further.