Study On Hand-Eye Calibration And Error Compensation For Ore Crushing Unmanned System

With the continuous advancement of the "Industry 4.0" era,to build "green mines" and "smart mines",the intelligent development of mines has gradually become a new development trend in the field of mineral technology development.In the process of ore mining and transportation,it is necessary to crush the bulk ore at the wellhead.At present,the main way of crushing operation is a mainly manual operation.Miners operate hydraulic crushing hammer to crush and treat large ore.The degree of automation is low and there are some safety hidden dangers.To realize the unmanned operation of ore crushing and improve the intelligentization of ore crushing,this subject combines machine vision technology and ore crushing operation to conduct the application research of hand-eye calibration and depth measurement error compensation for the visual system composed of Kinect camera and mechanical arm.The work of this paper mainly includes the following four aspects:(1)To obtain the mechanical coordinates of the target point required for hand-eye calibration,first,analyze and model the hydraulic mechanical structure of the breaker.According to the joint structure of the hydraulic mechanical arm,it is equivalent to a four-degree-of-freedom robot,and the model between the joint of the mechanical arm and the connecting rod is established by the D-H modeling method.The kinematics analysis of the hydraulic manipulator after modeling is performed to obtain the homogeneous transformation relationship between the coordinate system where the target execution point is located and the base coordinate system of the manipulator.The mechanical coordinates of the target point can be calculated by measuring the stroke of the hydraulic mechanical arm cylinder.(2)The hand-eye calibration technology is applied and realized in the visual system composed of a hydraulic mechanical arm.Set up an eye-to-hand visual system according to the working environment and working process.The checkerboard grid was used to calibrate the Kinect depth camera and the internal and external parameters of the camera were obtained.The hand-eye calibration of the eye-to-hand visual system was carried out through the designed 3D calibration pieces,and the transformation relationship of target points from the camera coordinate system to the manipulator base coordinate system was obtained.(3)Analyze and compensate for the depth error in-depth measurement.Obtained according to the principle of TOF called depth camera imaging analysis of error related to the same plane of the different optical path of flight time,through the establishment of the error lookup table and curve-fitting method for error compensation model,and then using depth imaging pixel rotational symmetry relations,when the equivalent distance between pixels on compensation model optimization.The maximum error and average error after compensation shall not exceed 5mm and 1mm,which is significantly higher than before compensation and the measurement accuracy is significantly improved.(4)Design the upper layer software interface of ore crushing unmanned system based on.NET platform according to research results and field operation process.