The Research On Coordination Planning And Control Of Mobile Manipulator

With the development of industrialization and intelligent manufacturing,most traditional robots that are only fixed in one place and complete repetitive tasks in a fixed space have become increasingly unable to meet the needs of robot applications.In order to better expand the potential of robots,more and more people focus on the technical research of mobile manipulator.The mobile manipulator system designed in this paper is composed of Kinova’s six-degree-of-freedom lightweight bionic manipulator,the Navigator Q2 omnidirectional mobile platform and Kinect2.0 depth camera.The main contents are as follows:First,the mobile manipulator system is built and the kinematic model is established.The mobile manipulator system was built,the performance parameters of the mobile platform and the manipulator were described in detail,the kinematics model of the omnidirectional mobile platform was established,and the forward and inverse kinematics of the manipulator were analyzed,laying a foundation for the subsequent coordinated planning and control.Secondly,design the vision system of the mobile manipulator.The camera model is determined to be the Kinect2.0 depth camera,the camera installation position is determined to be eye to hand,the camera imaging model is analyzed and established,and the internal parameter calibration and distortion correction are performed on the RGB camera and the depth camera of the Kinect2.0 depth camera.The image registration of the two and the depth correction of the depth camera are completed.We can obtain the coordinate transformation from the pixel coordinate system to the camera coordinate system.Hand-eye calibration is completed to obtain the coordinate transformation between the camera coordinate system and the manipulator base coordinate system.Then,the coordinated planning algorithm of the mobile platform and the manipulator is designed.The target recognition algorithm is determined to be the BRISK algorithm,and the target recognition and positioning are performed on this basis.The working space of the mobile manipulator is analyzed using Monte Carlo method,and then the best stopping position of the mobile platform is obtained by statistical analysis.The RRT*algorithm is determined as the path planning algorithm of the mobile platform through the simulation comparison analysis of the RRT algorithm,the RRT Connect algorithm and the RRT*algorithm.Plan the grasping trajectory of the manipulator,including polynomial interpolation and linear function fitting to parabola in joint space,linear interpolation and arc interpolation in Cartesian space,and using OPML motion planning library for obstacle avoidance planning,and simulated experiments were done separately.Finally,a comprehensive experiment of the mobile manipulator is carried out.Introduced the ROS system,including the ROS system framework,commonly used ROS components and Movelt!for the development of robotic arms platform.By analyzing the control mode of the Navigator Q2 omnidirectional mobile platform and the manipulator under ROS,the controller of the mobile manipulator is designed.Finally,a real experimental environment is built,and the coordinated planning and grasping experiment of the mobile manipulator is carried out to verify the feasibility.