Research On Key Technologies Of Intelligent Positioning Of Rock Drilling Robot

Rock drilling robot is a very important drilling equipment in tunnel drilling and blasting construction.At present,foreign enterprises occupy the main share of China’s rock drilling equipment market,and the relevant technical data are kept confidential.China has a late start in the intelligent research of rock drilling equipment,and there is a great demand for tunnel construction.The intelligent positioning accuracy of rock drilling robot is affected by many factors,and its intelligent positioning result will affect the positioning accuracy of gun hole,thus affecting the drilling accuracy and construction efficiency of tunnel.Therefore,the intelligent positioning problem of rock drilling robot needs to be solved urgently.In order to accurately realize the intelligent positioning of rock drilling robot,the key technologies of intelligent positioning of rock drilling robot were studied.D-H method was used to analyze the forward kinematics of the manipulator based on the physical structure of the manipulator,and the position and pose of the manipulator end in the base coordinate system were obtained,jacobian matrix iterative method was used to solve the inverse kinematics of the manipulator,and the kinematics model of the manipulator was established.Three attitude angles were used to describe the position and pose of the vehicle body in geodetic coordinate system.Based on the transformation principle of space coordinate system,the position and pose matrix between geodetic coordinate system and tunnel coordinate system was solved to determine the position and pose of the rock drilling robot in the tunnel,which solved the positioning problem of the vehicle body.Based on the cylindrical envelope box technology,the collision detection model of the robotic arm of rock drilling robot was established.By calculating and judging the distance between the cylindrical envelope box and the projection of the cylindrical envelope box and the tunnel contour line,the collision detection between the robotic arm of rock drilling robot and the robotic arm and the tunnel was realized.The joint data and the actual position coordinates of the end of the manipulator were collected to calculate the positioning error of the end of the manipulator,and the joint data and positioning error data were used as the input and output data of BP neural network for training and prediction.The weights and thresholds of BP neural network were optimized by genetic particle swarm optimization,and the error compensation model based on GAPSO-BP neural network was obtained,and the positioning error compensation of manipulator end was realized.The experimental results showed that the maximum positioning error of the manipulator end before compensation was 167.280 mm,and the maximum positioning error of the manipulator end after compensation was 52.964 mm.The positioning accuracy of the manipulator end was improved by 68%.The end positioning error compensation results based on GAPSO-BP neural network meet the practical engineering requirements.