| In recent years,with the enhancement of awareness of people's requirements for the safety of life and property,aviation safety problem has been widespread concern.The wall climbing robots with movement speed,stable movement,reaction flexible has broad application prospects in the detection of aircraft surface and anti-hijacking missions.In the aircraft surface detection,it can reduce the labor intensity of the operation personnel,improve the fault detection rate.In the application of the anti-hijacking,it can maximize the assurance against the personal safety.The research in this paper is supported by the National High Technology Research and Development Program of China(863 Program)under Grant 2009AA04Z2171,the National Natural Science Foundation of China under Grant 61179049.In this paper,researches are started with the climbing robot obstacle planning methods based on laser point cloud.Mainly include the following four aspects:First,in order to realize the wall climbing robot to a smooth,fast movement on the aircraft surface and realize the function of autonomous negotiating obstacles.It combines with the requirements of the controller with small size,low power consumption and light weight of the actual prototype of the wall climbing robot and it determines the controller of the wall climbing robot.In this paper,the selection of the main control chip,the hardware design scheme of the system and the resource allocation of the main control chip are introduced in detail.And combined with the hardware circuit of the wall climbing robot to explain the details of the various parts of the hardware system.Second,for the wall climbing robot autonomous obstacle negotiation demand combines with the wall climbing robot's working environment,we proposed the use of 2D laser range finder to achieve the wall climbing robot environment modeling.By building a 3D laser scanning system,the correctness of the proposed method is verified by the upper computer.The content and the process of the algorithm are described.By building a 3D laser scanning system,the correctness of the triangular mesh construction based on point cloud matrix and edge feature extraction algorithm is verified on the host computer.Third,movement pattern of the wall climbing robot should be executed by analysis how the wall climbing robot can successfully climb obstacles.The first the climbing robot executes the wheeled movement and then the climbing robot implements the motion model of biped locomotion to complete the obstacle negotiation.It focuses on the design of fuzzy controller of the wall climbing robot in the wheeled movement model.The kinematic equation of the wall climbing robot under the wheeled movement model and the possible movement of the wall climbing robot on the plane surface are used to design the robot's fuzzy controller.The fourth part is the experimental study in this paper.First,the laser point cloud data is processed.The proposed algorithm can effectively achieve the triangle mesh construction of the wall climbing robot motion environment.And it can accurately extract the boundary feature information of the front environment.It completes the environmental modeling.Secondly,the fuzzy controller is designed for simulation experiments.Determine the wall climbing robot in wheeled locomotion mode,can use the designed fuzzy controller to realize the wall climbing robot from the attachment surface to move anywhere to ideal obstacle position.At last,the robot prototype of the biped robot is carried out to verify the ability of the wall climbing robot.Secondly,the fuzzy controller is designed for simulation experiments to conform that the wall climbing robot in wheeled movement model can achieve the position which is the ideal obstacle negotiation position from anywhere on the surface by using the fuzzy controller.Finally,the wall climbing robot prototype takes experiments under the bipedal motion model and verified the wall climbing robot prototype has a better obstacles negotiation ability. |
PDF Full Text Request