Obstacle Crossing And Anti-slide Control Of High Voltage Transmission Line Inspection Robot

Ground line of various grades of high voltage power line share the same disirable traits: similarity in structure, open viewing angle, few electro-magnetic disturbance, so that research of inspection robot along earth line has been gaining more and more importance. Nevertheless, obstruction crossover has become a difficulty because no magnetic field exists around earth line and that is what inspection robot navigation bases on. Measuring distance between unit and obstacles so that robot will not interfere with obstacle, controlling the body pose when crossing obstacles, increasing obstacle crossing efficiency through dual-axis coordinationand anti-sliding during robot patrolling are the4key techniques that calls for desperate effort. And this page is comprised of4parts accordingly:1)A contact-free navigation tactic is composed based on obstacle location and coordinate transformation method. A two stage approach by first examining obstacles with known line information followed by precise location method is studied. To be specific, location point is used as base point, on which a special coordinate is constructed.Lastly homogeneous transformation space reconstruction is used to locate obstacle and robot. Tactics, which guarantee robot will not crash on obstacle,of crossing over several typical obstacles is studied down to each step, after robot to obstacle space relationship is measured. All these put together can enable robot to cross obstacle autonomously without trouble.2)Reason of rotation of line-grabbing arm around line under condition of robot crossing obstacle in a reverse pose is analyzed. With knowledge of rotation angle changing behavior, relationship between rotation angle of grabbing arm and maneuvering distance of crossing arm are decided as linear relationship, which set the corner stone for crossing arm propelling system modeling. Then crossing arm propelling system is modelled. And dynamic function of crossing arm movement is presented. System state space modelling is constructed after selection of state criteria. State space model clearly depict the dynamic changing condition between different state criteria, reflecting system dynamic behavior. In order to obtain better controlling performance, feedback control method based on state criteria is designed. Observation instrument is designed for the system to tackle situation when state condition cannot be directly observed. Target function that simultaneously meet two standards, minimizing state criteria deviation square sum integration and minimizing power consuming is presented.3)Importance of dual-axis coordination is introduced; DOF number when crossing obstacle is decided; General dynamic model when crossing obstacle is modelled; a general dual-axis coordination dynamic function construction method is proposed, further, a PD controlling method based on gravity model is designed, whose overall stability is validated by Lyapunov law; After analyzing obstacle crossing condition and motor to propeller matching condition, coordinating articulation which might present in practical study is found. Finally, matlab is used to simulate the2groups of coordinating articulation, and experiment is carried out for validation. The result shows that, the dual-axial controlling method, once applied to robot, can save time in crossing obstacle, increase obstacle crossing efficiency, enhance autonomy of robot, fulfill practical patrolling need.4)Reasons for sliding have been studied. It is pointed out that dune angle is the major criteria that affect robot's normal moving.Robot dune climbing ability is then studied, maximum climbing angle is then given, theoretical evidence of squeezing wheel improving robot dune climbing ability is obtained. Sliding model of robot is constructed, a slide detection method using squeezing arm is proposed. Sliding controlling method is studied based such foundation, which controls sliding by controlling squeezing arm moving distance. Sliding gain and sliding degree are analyzed, relationship between which is also learned. A fuzzy controlling method using sliding degree and dune degree as input is designed to cope with difficulty in precise mathematical modelling of sliding. Lastly, experiments are made to validate the proposed slide controlling method. Experiment result proves the validity of the proposed fuzzy controlling method, which can effectively eliminate sliding phenomenon presented in robot, fulfilling practical needs.