ROS-based Motion Planning Research For Indoor Mobile Manipulator

Currently,the development of robot technology has become a focus of attention in various countries around the world.Mobile manipulators have attracted attention due to their combination of the operability of a robotic arm and the mobility of a mobile robot,and their motion planning has always been a key research area in the field of robotics.This thesis takes the indoor environment as the application background,and uses the Turtlebot2 mobile platform and a DIY six-degree-of-freedom(DOF)robotic arm as the mobile manipulator system to analyze and integrate motion planning from two angles: obstacle avoidance of the mobile platform and obstacle avoidance of the robotic arm during object grasping.The thesis uses ROS as the core to build a robot simulation model,calculates and analyzes obstacle avoidance algorithms,and conducts experiments in an indoor scene.The main contents of this thesis are as follows:Firstly,the obstacle avoidance motion planning of the entire process of the robotic arm grasping an object is studied.The DIY robotic arm is modeled in the ROS platform,the motion parameters of the robotic arm model are configured,and collision detection is completed.A robotic arm obstacle avoidance motion scene is set up,and the parameters of the target object and the end effector of the robotic arm are set in a Python program.RRT-Connect is used as the obstacle avoidance motion planning algorithm,and the obstacle avoidance motion planning of the robotic arm grasping target objects of different shapes and sizes is studied through simulation.Secondly,in response to the problems of slow search speed and easy local optimization that traditional A~* algorithm faces,its heuristic function is improved.Two parameters,the ratio of the distance between the starting point,the current node,and the target point,and the ratio of the obstacle occupancy of the overall map,are introduced into the heuristic function to optimize it.Based on the ROS system,a map is constructed in the simulation environment built in Gazebo,and obstacle avoidance navigation experiments of the mobile platform subsystem are conducted.The results prove that compared with the traditional A~* algorithm,mobile robots can more effectively avoid surrounding obstacles and reach the target point.Finally,obstacle avoidance motion planning experiments of the robot system are conducted in an indoor environment.The Trans Form Frame(TF)coordinate transformation relationship is used to solve the problem of map displacement in the mapping process,and Hector＿mapping is used as the algorithm for building the map in this system.The operability value of the robotic arm and the reachability of the obstacle avoidance grasp are used as the evaluation criteria for determining the stopping position of the mobile manipulator system,and obstacle avoidance and grasping experiments of the robotic arm subsystem are conducted in a real environment.The experimental results show that the mobile manipulator can achieve various functions such as autonomous localization and mapping in indoor environments,navigation from the starting point to the target object area,and obstacle avoidance and grasping.