Research On Visual Detection And Trajectory Planning And Control Method Of Rotor Flying Robot For Eliminating Foreign Matter In Power Supply And Distribution Network

The attachment of foreign objects in the power supply and distribution network is an important factor that threatens the safety of rail transit,so it is necessary to make accurate identification and judgment on foreign matters and deal with them in time.The traditional operation and maintenance methods of rail transit have high cost,low intelligence and low efficiency,which can no longer meet the requirements of the development of the industry.At present,most UAV patrol inspections can only carry out passive environmental monitoring,but cannot directly interact effectively with the outside and cannot deal with problems in a timely manner.Therefore,the rotor flying robot equipped with an active working mechanism has been paid more and more attention by researchers.In this paper,a rotor flying robot combining the rotorcraft and the multi-joint manipulator is studied,and is equipped with a visual servo system,so that it can meet the requirements of the task of removing foreign matters in the power supply and distribution network.To this end,from the complexity of the foreign matter removal task,the limitations of the traditional operation and maintenance mode of rail transit,and the advantages that the rotor flying robot has in solving these deficiencies,a model for the problem of foreign matter invasion and removal in power supply and distribution network is constructed.Aiming at the problem of intelligent identification and detection of suspended foreign matters in power supply and distribution network,combining with the characteristics of the image and the characteristics of foreign matters,a robust detection algorithm for moving objects under dynamic background based on dense optical flow method is proposed.The research focuses on three aspects of catenary image preprocessing method,image enhancement and foreign matter detection,and realizes the task of object detection in complex environment.In order to effectively track the detected target area,an improved CAMSHIFT target tracking algorithm based on Kalman filter is proposed,and the tracking template of the original algorithm is improved.Aiming at the intelligent control problem of multi-joint manipulator’s grasping operation,by analyzing the structure and motion characteristics of typical multi-joint manipulator,the number of the manipulator’s joints is determined,and its kinematics is analyzed.A multiconstrained multi-joint trajectory planning method based on quintic polynomials is proposed.Then for the problem of visual servo control of the manipulator,a visual servo system based on the model predictive control of the double-layer structure is established,and a position-based visual servo controller is designed.This dissertation is an exploration of the research direction of object grabbing for rotor-flying-robots,which realizes the intelligent recognition and detection of foreign matters suspended in the power supply and distribution network of rail transit,as well as the intelligent control of multi-joint manipulator.Simulation experiments have verified The algorithm proposed in this paper has achieved good results in solving the problem mentioned above.The research has certain reference significance for solving problems in the fields of intelligent transportation,powerlines inspection,and intelligent control.