Study On Road Information Identification Technology Of All Terrain Crane Based On Data Fusion

All terrain crane has been widely used in urban construction and infrastructure construction due to the large lifting tonnage and excellent off-road performance.As the main incentive in its driving process,the road directly affects various aspects such as the ride comfort,stability and safety,so precisely identifying the road is particularly critical for improving the performance of all terrain crane.In this research,the road identification technology based on the all terrain crane is systematically studied,wherein the five axis all terrain crane is taken as the research object,the whole vehicle dynamics model including the hydro-pneumatic suspension system is established,the road classification identification method and the forward road identification system based on the data fusion technology are proposed,and the road identification test system of the all terrain crane is designed and built.On the basis of literature review of all terrain crane and the road recognition method,the advantages and disadvantages of the research on the hydro pneumatic suspension system related to all terrain crane,the data fusion technology related to the road identification method and the neural network method are analyzed and discussed,and a research scheme of road identification method based on all terrain crane is proposed.Based on the structure and working principle of hydro-pneumatic suspension,a mathematical model of hydro-pneumatic suspension system of five axis all terrain crane considering the influence of actual gas state change,temperature and other factors is established,and three interconnection schemes of hydro-pneumatic suspension system are proposed,and the stiffness and damping characteristics of the three schemes under vertical,roll,pitch and torsional vibration are analyzed by applying AMESim simulation.On this basis,the applicability of different schemes is analyzed according to the structural characteristics and working characteristics of all terrain crane,and the interconnection scheme in which the independent front and rear parts and cross connection between left andright parts is preferred.The dynamic model of all terrain crane including road excitation and hydro pneumatic suspension system is established.Based on the simulation analysis of the dynamic model by Matlab/Simulink,the response of suspension and body under different road excitation is obtained.In order to verify the reliability of the whole vehicle dynamics model,the driving response test of the five axis all terrain crane is carried out.The correctness of the dynamic model is verified.Based on the data fusion technology,a method of road classification identification is proposed,in which the correlation function method is used to fuse the pressure signals of the inner cavity of multiple cylinders,the feature extraction is performed on the fused signals,the relative displacement signals of the cylinders and the acceleration signals of the center of mass based on the wavelet decomposition,the dimensions of the features reduction is performed using the supervised local tangent space arrangement algorithm,and finally,the feature level fusion is performed using the neural network method to complete the road classification identification.Based on the simulation results of the whole vehicle dynamics model,it is found that the method can accurately identify the road classification,which verifies the effectiveness of the method.The road classification evaluation test platform and road classification identification test system are designed and built to evaluate and analyze the typical road classification,and collect the vehicle response signal and road classification identification results,the results show that the proposed road classification identification method can accurately identify the road classification.A road sensing system based on data fusion technology for all terrain crane locomotive is designed,which uses a lidar,an inertial measurement unit and a stereo camera to recognize the forward road information.Laser radar and inertial measurement unit are used to carry out joint external parameter calibration.Based on ORB algorithm,the feature points for the image collected by stereo camera are extracted and described,and then the pose transformation of camera is solved,and the pose fusion of inertial measurement unit and stereo camera is carried out based on multi-sensor data fusion technology.Based on the recursive algorithm,lidar is used to collect the elevation information of the road,and the collected point cloud data is filtered,then the probability guided random samplingconsistency algorithm is used to reconstruct the road.In order to verify the effectiveness of this method,the forward road sensing system base on all terrain crane is tested,and the forward road point cloud data is collected and the test results verified the effectiveness of the forward road sensing system.To sum up,this research establishes a dynamic model of five axis all terrain crane including hydro-pneumatic suspension system.Based on the data fusion technology,a road classification identification method and a forward road sensing system are proposed,and the road identification test system is designed and built.The test data obtained by the system verifies the effectiveness of the road classification identification method and the forward road sensing system.In this research,a study of road identification technology is provided,which provides a basis for the active control of the ride comfort of all terrain crane and is of great significance for the intelligent development of all terrain crane.