| Industrial robots are widely used in handling tasks as key equipment in the field of intelligent manufacturing.At present,in the production of new energy vehicle batteries,most of the robots use offline programming or online point-by-point teaching methods.With the acceleration of the iterative upgrade speed of automobiles and the increasing demand for personalized customization,the manual teaching method of transporting batteries has the disadvantages of long debugging cycle,poor applicability,and high cost.This method has been difficult to meet the needs of new energy vehicle production lines for flexibility and intelligence.Aiming at the problem of poor adaptability of manual teaching,this thesis focuses on the autonomous transportation of batteries on the intelligent flexible production line,and conducts research from three aspects: battery pose estimation,robot hand-eye calibration,and robot transportation planning.The main research contents of this thesis are as follows:(1)By calibrating the internal and external parameters of the depth camera,the color camera and infrared camera of the depth camera are registered;based on the parameters of the depth camera calibration,an improved Linemod template matching algorithm based on the random sampling consensus algorithm is obtained.The 6-DOF posture expressed in camera coordinates is obtained under the algorithm.(2)Aiming at the problem of posture conversion between camera coordinate and robot tool coordinate,namely hand-eye calibration problem.According to the actual camera assemble method,the mathematical model of robot hand-eye calibration is established.The objective function to be optimized is established by decoupling handeye calibration mathematical model.Based on the objective function,an algorithm based on manifold space optimization is obtained.The algorithm also optimizes the rotation and translation equations to obtain the conversion relationship between the camera coordinate and the robot tool coordinate.(3)After determining the position and posture of the battery expressed under the robot tool coordinate,the autonomous battery transportation planning problem of the robot is studied.Aiming at the transportation sequence of multiple batteries,the transportation sequence of multiple batteries is modeled as a traveling salesman planning problem,and the traveling salesman planning problem is solved by a dynamic programming algorithm,and a better battery transportation sequence is obtained.Aiming at the problem of collision-free path planning when a single object is transferred by a robot,a mathematical model of robot motion planning is established based on the Gaussian process.A nonlinear iterative optimization algorithm is used to solve the mathematical model of motion planning,and the collision-free path of the robot is obtained.(4)Based on the Gazebo simulation platform of the robot operating system,a digital model of the robot battery handling station is built,and the handling planning algorithm is used to simulate the new energy vehicle battery handling process.Simulation experiments show that the transportation sequence based on the dynamic programming algorithm and the motion planning algorithm based on the Gaussian process are highly efficient and can be used for autonomous transportation planning of the battery,thereby the intelligence and flexibility of the new energy vehicle production line improved. |
