Research On Collision Detection And Hole Sequence Planning Of Rock Drilling Robot

Rock drilling robots play a vital role in improving the efficiency and safety of tunnel operations.They are indispensable construction machinery in tunnel construction,and the control system is the core part of rock drilling robots.In order to realize the completely independent research and development of the rock drilling robot control system and improve the product competitiveness of domestic rock drilling robots.In this paper,the collision detection in the control system and the hole sequence planning of the working arm are taken as the main research contents,and the forward kinematics model of the full arm of the rock drilling robot is established.And based on this model,the drill boom collision detection algorithm was developed,the mathematical model of hole sequence planning was completed,and the rock drilling robot was tested on the relevant tunnel section operation.The main work is as follows:1.Firstly,it is necessary to carry out forward kinematics analysis of each working arm and auxiliary arm of the rock drilling robot.The DH modeling method is used here,and the XZ coordinate system transformation method is used to solve all joint coordinate transformation matrices in the forward kinematics model of the rock drilling robot.The simulation modeling of each manipulator of the rock drilling robot is carried out through the Robotic Toolbox in MATLAB software.At the same time,the forward kinematics algorithm program of the rock drilling robot is written,and the three-dimensional simplified model of all arms in the same Cartesian coordinate system is drawn.2.Simplify each joint model of the manipulator into a capsule model,and use the joint endpoint coordinates calculated by the rock drilling robot’s forward kinematics model to perform collision detection related calculations.Determine whether a collision occurs by calculating the distance between two simplified rods.Among them,the distance calculation method mainly adopts the vector calculation method from point to line segment,and the solution is more accurate and fast.At the same time,a multi-level collision detection algorithm is proposed by combining the vector distance calculation algorithm with the AABB bounding box algorithm.More accurate simplified member collision detection is no longer performed when the outer bounding box does not detect collisions.The detection speed requirement and accuracy of the hierarchical bounding box algorithm are verified by simulation analysis.This algorithm has greater advantages in discrete time sampling frequency while ensuring the detection accuracy.3.The hole sequence planning problem is transformed into a variant TSP model with a fixed starting point and no need to return to the starting point.At the same time,the sum of the joint parameter changes of all joints in the drill arm is used as the optimization target of the model,and the corresponding TSP model is established.In this model,a new improved ant colony algorithm is proposed,an adaptive improvement strategy for pheromone volatilization factor is proposed,and a new improved optimization scheme is proposed for the pheromone concentration increment model.Compared with the basic ant colony algorithm and other optimization algorithms of the same type,it is found that the improved ant colony algorithm has faster convergence speed and excellent global optimization performance.Using the improved TSP model algorithm,the dual-arm hole sequence planning simulation is carried out on the tunnel shot hole diagram,which verifies the rationality and safety of the hole sequence planning algorithm.4.Taking the dual-arm rock drilling robot SWDT62 GG as the experimental object,firstly,the experimental platform including the software and hardware platform is built and tested for the rock drilling robot.After the debugging of the test platform is completed,the relevant experiments of collision simulation are carried out on the rock drilling robot.The detection accuracy of the multi-level collision detection algorithm in the detection of the collision between each arm and its self-collision is verified,and the results show that the overall algorithm accuracy can reach 96.7%.At the same time,the tunnel operation was tested and tested to verify the double-arm hole sequence planning algorithm.The results show that the completion time after planning is shortened by 12% compared with the operation time without planning,and the operation efficiency after planning is significantly improved.