This document discusses the simulation of a robotic arm in ROS. It aims to simulate grasping, trajectory planning, and pick and place operations. The simulation will model the robot's kinematics and dynamics using URDF files in Gazebo. It will generate waypoints and trajectories for the arm using RRT path planning and MoveIt. Current work involves communication between ROS packages and simulating the arm. Remaining tasks include hardware interfacing, grasping simulation, and applying dynamic loads in Gazebo.
Download as PPTX, PDF, TXT or read online on Scribd
0 ratings0% found this document useful (0 votes)
141 views
Robot Simulation
This document discusses the simulation of a robotic arm in ROS. It aims to simulate grasping, trajectory planning, and pick and place operations. The simulation will model the robot's kinematics and dynamics using URDF files in Gazebo. It will generate waypoints and trajectories for the arm using RRT path planning and MoveIt. Current work involves communication between ROS packages and simulating the arm. Remaining tasks include hardware interfacing, grasping simulation, and applying dynamic loads in Gazebo.
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13
Robotic Simulation
Software basis of the robot
Introduction This sub-module of the project deals with the simulation of motion of robot in the presence of static object placement. The module primarily focusses on using pre-existing solutions for implementation, being applied to custom design. Kinematics and dynamics of the robot is of primary concern. Motivation Advantages of ROS: Provides abstraction hence shifting from one design to a different one is easier. Conventions and protocols help in keeping the workspace of robot modular in functionality. Softwares such as RViz, MoveIt and Gazebo have extensively developed API for interfacing with robot. Literature Survey: Mastering ROS for Robotics Programming : This book aids in fundamental understanding of ROS architecture and subsequent understanding of various above mentioned softwares. Introduction to Robotics: Mechanics and Control by John J Craig : For understanding the DH convention representation of Robot and forward kinematics and inverse kinematics. Objectives of the exercise The simulation has three fold objectives: The purpose is to simulate the processes of grasping and creating trajectories for sliding and pick and place. The simulation produces the planning path and joint accelerations and velocities for calculation of effort. The simulation also provides packages for real hardware interfacing. The processing of joint values and raw images from the object identification algorithm. Process (Visualisation of Robot) The first job was to choose the platform for creating simulations and simultaneously create backup for kinematics of the robot. The URDF of the robot was created using only primitives and the URDF was then used to generate additional files for kinematic solution planning. Matlab interface was tried to provide testing platform for controls and kinematic solutions. Currently a 3 DOF arm is produced with links of different inertia. A snip of the code is shown in next slide: Kinematics and dynamics of Robot Analytical solution for 3-DOF arm has been generated and tested using: Waypoint Generation Details of the trajectory path can be obtained for a translating to a particular, a snapshot of a waypoint description of movement of robot from its original position to a position of :0.16,0.12,2.33 in the frame of reference of base link. positions: [0.10014551221842302, 0.2178672469882867, 1.7660553833418764] velocities: [0.015222228822740835, 0.033214828587172576, 0.2697640434502649] accelerations: [-0.05600594596520207, -0.1222046992433586, -0.9925215693945678] time_from_start: secs: 3 nsecs: 805271578
The target of the robot is 0.16,0.12,2.33
RRT Connect Algorithm(for navigation and local control) Current Work Status Communication within ROS and its packages is of prime importance and the several messages that are sent by softwares have a degree of abstraction. With our current capabilities we can:
1 2 3 RQT Graph with MoveIt and RViz Glimpse of Robotic arm reaching the box position Tasks yet to complete
Hardware Interfacing: Creating Grasping messages and
Configuration files of MoveIt will gripper: have to interact with data from For gripping operation and pick sensors. and place, the points where robot grasps the object have to be identified. Final Simulation
Applying dynamic loads on robot Receiving feedback from real time
in gazebo simulator. system and creating nodes and topics for typical feedbacks.