| The infrastructure construction in our country gradually perfect,and the mileage of roads in service is increasing year by year,the attendant task of road maintenance is also increasingly arduous.Pavement crack is the most common early road disease,timely completion of pavement crack repair work,can prevent further erosion of the road,thus improving the service life of the road.At present,artificial crack irrigation is the main crack repair method in our country,this labor-intensive road maintenance method is low in efficiency and has potential safety hazards,automated pavement crack repair is the development trend of road maintenance,with the rapid development of cutting-edge technologies such as robotics and machine vision,developing a robot system suitable for pavement crack repair provides a new idea for automatic pavement crack repair.In this paper,an automatic pavement crack repair system is designed and developed based on a visual inspection system and a robotic arm.The multi-adaptive path planning algorithm suitable for various crack distribution states and the automatic capture and tracing technology of crack trajectory are mainly studied.By extracting the skeleton of the binarized crack image,the coordinate information of crack pixels is obtained,and the key parameters such as crack number,starting point,intersection point and trajectory are identified.On this basis,the fracture distribution pattern is divided into independent distribution and cross distribution,and the cross fracture is transformed into independent fracture by dividing the cross points,the minimum weight perfect matching method is used to connect the independent cracks,and then the edge with the largest weight is removed to generate the crack repair path.The effectiveness of the designed crack repair path planning algorithm is verified by the example analysis of two different crack distribution types.The hand-eye system is composed of the manipulator and the monocular industrial camera,the kinematics of the manipulator is modeled by the D-H parameter method,and the forward and inverse kinematics analysis is carried out to obtain the motion relationship between the end position of the manipulator and the rotation angle of each joint.The checkerboard calibration method is used to correct the distortion of industrial cameras,and the hand-eye matching is carried out by the way of eye out of hand,according to the requirements of the working space and degrees of freedom of the pavement crack repair system,the hand-eye system is calibrated by using the plane nine-point calibration method,the transformation relationship between the image pixel coordinate system and the manipulator workspace coordinate system is obtained,and the hand-eye system fusion control algorithm is developed.The test platform of automatic pavement crack repair system was built,and the test method and process were developed,the fracture trace tracing test was carried out for independent fracture distribution pattern,cross fracture distribution pattern and composite fracture distribution pattern respectively,the experimental results show that the pavement crack repair system designed and studied in this paper can effectively solve the problem of capturing all the crack trajectories in the working area,and automatically plan the optimal path to complete the trajectory traversal,the robotic arm can automatically trace the fracture trajectory under the requirements of fracture repair accuracy. |
