diff --git a/domestic_robotics/images/hsr_sim_hbrs_c069.png b/domestic_robotics/images/hsr_sim_hbrs_c069.png new file mode 100644 index 0000000..3349d8a Binary files /dev/null and b/domestic_robotics/images/hsr_sim_hbrs_c069.png differ diff --git a/domestic_robotics/manipulation.ipynb b/domestic_robotics/manipulation.ipynb new file mode 100644 index 0000000..507bf55 --- /dev/null +++ b/domestic_robotics/manipulation.ipynb @@ -0,0 +1,374 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "import IPython" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Manipulation" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This tutorial describes how manipulation is performed by our domestic robots. All examples in this tutorial will refer to the Toyota Human Support Robot." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## List of contents\n", + "\n", + "1. [Manipulation: The `move_arm` Action](#Manipulation:-The-move_arm-Action)\n", + "2. [`move_arm` in Action](#move_arm-in-Action)\n", + "3. [Dynamic Motion Primitives](#Dynamic-Motion-Primitives-%28DMPs%29)\n", + " 1. [DMP Primer](#DMP-Primer)\n", + " 2. [Learning Motion Primitives](#Learning-Motion-Primitives)\n", + " 1. [Trajectory Demonstration](#Trajectory-Demonstration)\n", + " 2. [Learning the Weights of the Motion Primitive](#Learning-the-Weights-of-the-Motion-Primitive)\n", + " 3. [Using a DMP](#Using-a-DMP)\n", + " 4. [Tuning the DMP Parameters](#Tuning-the-DMP-Parameters)\n", + "5. [Arm Motion Use Cases](#Arm-Motion-Use-Cases)\n", + " 1. [Picking Up Objects](#Picking-Up-Objects)\n", + " 2. [Placing Objects](#Placing-Objects)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Manipulation: The `move_arm` Action\n", + "\n", + "To perform manipulation and in line with our skill-based domestic robot architecture, we use a `move_arm` action that exposes a high-level interface for controlling a robot arm (under the assumption that there is a single robot arm). The `move_arm` action allows controlling the arm in three different ways:\n", + "1. the arm joints can be sent to a predefined named configuration (e.g `folded`) using MoveIt!\n", + "2. the end effector can be sent to a desired pose with respect to a given coordinate frame; the arm motion can be performed either using MoveIt! or with a dynamic motion primitive\n", + "3. the joint positions can be set arbitrarily; once again, the motion will be performed using MoveIt!\n", + "\n", + "The way in which the arm is controlled can be selected by setting the `goal_type` parameter of the action, such that the additional parameters that have to be passed to the action depend on the goal type. Please refer to the [documentation](https://github.com/b-it-bots/mas_domestic_robotics/tree/kinetic/mdr_planning/mdr_actions/mdr_manipulation_actions/mdr_move_arm_action) of the action for the action specification.\n", + "\n", + "Note that the named joint configurations need to be specified in the MoveIt! configuration file for a specific robot (for the Toyota HSR, they are in the `hsrb.srdf` file in the `hsrb_moveit_config` package)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## `move_arm` in Action\n", + "\n", + "We can now illustrate how arm motions can be performed. Let us first start the simulation of the Toyota HSR:\n", + "\n", + "```\n", + "roslaunch mas_hsr_sim hbrs-c069.launch\n", + "```\n", + "\n", + "This should result in the following Rviz output:" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "IPython.core.display.Image('images/hsr_sim_hbrs_c069.png', embed=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Once the simulation has started, we need to start the server for the `move_arm` action:\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_arm_action move_arm.launch \n", + "```\n", + "\n", + "Let's now set up a client for the action:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import rospy\n", + "import actionlib\n", + "from mdr_move_arm_action.msg import MoveArmAction, MoveArmGoal\n", + "\n", + "# we set an arbitrary timeout for the action\n", + "action_timeout = 15.\n", + "\n", + "rospy.init_node('move_arm_client_test')\n", + "\n", + "client = actionlib.SimpleActionClient('/move_arm_server', MoveArmAction)\n", + "client.wait_for_server()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's assume that we want to send the arm to the $(x,y,z) = (0.4, 0.078, 0.8)m$ with orientation $(x,y,z) = (0, 0, 0)$ with respect to the `base_link` frame using MoveIt! The following code snippet (adapted from the [move_arm_action_client_test](https://github.com/b-it-bots/mas_domestic_robotics/blob/kinetic/mdr_planning/mdr_actions/mdr_manipulation_actions/mdr_move_arm_action/ros/scripts/move_arm_action_client_test) script) will do just that:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from geometry_msgs.msg import PoseStamped\n", + "\n", + "goal = MoveArmGoal()\n", + "goal.goal_type = MoveArmGoal.END_EFFECTOR_POSE\n", + "\n", + "pose = PoseStamped()\n", + "pose.header.frame_id = 'base_link'\n", + "\n", + "pose.pose.position.x = 0.4\n", + "pose.pose.position.y = 0.078 # hsr-specific arm offset\n", + "pose.pose.position.z = 0.8\n", + "\n", + "pose.pose.orientation.x = 0.\n", + "pose.pose.orientation.y = 0.\n", + "pose.pose.orientation.z = 0.\n", + "pose.pose.orientation.w = 1.\n", + "\n", + "goal.end_effector_pose = pose\n", + "\n", + "# empty if we want to use MoveIt! rather than a DMP for moving the arm\n", + "goal.dmp_name = ''\n", + "\n", + "client.send_goal(goal)\n", + "client.wait_for_result(rospy.Duration.from_sec(int(action_timeout)))\n", + "\n", + "result = client.get_result()\n", + "if result and result.success:\n", + " print('Arm moved successfully')\n", + "else:\n", + " print('Could not move the arm')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If we want to send the arm to a predefined joint configuration instead of sending the end effector to a given pose, we can call the action as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "goal = MoveArmGoal()\n", + "goal.goal_type = MoveArmGoal.NAMED_TARGET\n", + "goal.named_target = 'go'\n", + "\n", + "client.send_goal(goal)\n", + "client.wait_for_result(rospy.Duration.from_sec(int(action_timeout)))\n", + "\n", + "result = client.get_result()\n", + "if result and result.success:\n", + " print('Arm moved successfully')\n", + "else:\n", + " print('Could not move the arm')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Dynamic Motion Primitives (DMPs)\n", + "\n", + "For the purpose of easier setup and portability, our robots have traditionally used [MoveIt!](https://moveit.ros.org) as a manipulation framework. As described above, our `move_arm` action uses MoveIt! as well, but not exclusively, such that we additionally make use of dynamic motion primitives [1].\n", + "\n", + "MoveIt! generally uses randomised planners for finding motion trajectories for a robot manipulator. This offers flexibility in the motion planning process; however, randomised planners often result in trajectories that are unpredictable or suboptimal (e.g. with respect to the distance travelled). Dynamic motion primitives on the other hand encode trajectories that a manipulator will then try to reproduce during its motion; one benefit of this is that resulting trajectories are more predictable and often more natural.\n", + "\n", + "### DMP Primer\n", + "\n", + "In the DMP framework [1], trajectories are represented by a second-order differential equation of the form\n", + "\n", + "\\begin{equation*}\n", + " \\tau \\ddot{\\mathbf{y}} = \\alpha(\\beta(\\mathbf{g} - \\mathbf{y}) - \\dot{\\mathbf{y}}) + \\mathbf{f}\n", + "\\end{equation*}\n", + "\n", + "In this equation, $\\mathbf{y}$ is the state of the system (usually the 3D Cartesian pose of the robot), $\\mathbf{g}$ is the desired end effector pose, $\\tau$ is a parameter that adjusts the trajectory duration, while $\\alpha$ and $\\beta$ are positive constants (please consult [these very nice lecture notes](http://tutorial.math.lamar.edu/Classes/DE/Vibrations.aspx) if you need a refresher on second-order differential equations and spring-mass-damper systems in particular).\n", + "\n", + "A common use for DMPs is to encode trajectories that have been demonstrated to a robot, such that the goal in this case is finding a representation for $\\mathbf{f}$ that will represent the demonstrated trajectory as closely as possible; the forcing term should however eventually vanish so that the trajectory can converge to the desired goal. In [1], which is what our implementation is based on, the forcing term is represented as\n", + "\n", + "\\begin{equation*}\n", + " f_j(t) = \\frac{\\sum_{i=1}^{N}\\Psi_{i,j}(t)w_{i,j}}{\\sum_{i=1}^{N}\\Psi_{i,j}(t)}\n", + "\\end{equation*}\n", + "\n", + "where the $\\Psi_{i,j}$ are exponential basis functions and the $w_{i,j}$ are weighting terms for the individual basis functions.\n", + "\n", + "We demonstrate trajectories by moving a marker array and recording the observations of the marker using a robot's camera (a video illustrating the demonstration process can be found at https://www.youtube.com/watch?v=jEtlm96KAbA); given a demonstration of the trajectory, we can learn the trajectory weights and then use them for execution. The learning process is described next.\n", + "\n", + "[1] A. J. Ijspeert, J. Nakanishi, H. Hoffmann, P. Pastor, and S. Schaal, “Dynamical Movement Primitives: Learning Attractor Models for Motor Behaviors,” Neural Computation, vol. 25, no. 2, pp. 328-373, 2013. Available: https://homes.cs.washington.edu/~todorov/courses/amath579/reading/DynamicPrimitives.pdf" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Learning Motion Primitives\n", + "\n", + "The process of learning a DMP involves two steps, namely\n", + "\n", + "1. a trajectory demonstration needs to be performed and\n", + "2. given the demonstrated trajectory, the weights of the DMP need to be learned\n", + "\n", + "#### Trajectory Demonstration\n", + "\n", + "To learn a motion primitive, we first need to demonstrate a trajectory and record it; the [`demonstrated_trajectory_recorder`](https://github.com/abhishek098/demonstrated_trajectory_recorder/tree/master) package has been designed for that purpose. For recording a trajectory, we need to launch the trajectory recording node:\n", + "\n", + "```\n", + "roslaunch demonstrated_trajectory_recorder demo.launch\n", + "```\n", + "\n", + "The command line instructions can then be followed for starting the recording. After that, the trajectory needs to be demonstrated by moving the marker rray in front of the robot's camera (it is clearly important that the marker array remains visible throughout the demonstration). Once the demonstration is over, the recording can be stopped; the trajectory will then be saved to a YAML file with a name specified on the command line under `/data/file_name.yaml`.\n", + "\n", + "#### Learning the Weights of the Motion Primitive\n", + "\n", + "To learn the weights of the motion primitive, we need to use the functionalities in the [`ros_dmp`](https://github.com/abhishek098/ros_dmp/blob/whole_body_motion/src/learn_motion_primitive.py) package - currently on the `whole_body_motion` branch). Before proceeding with the learning step, we need to copy over the demonstrated trajectory to `/data/trajectories/old_trajectories/`. We can now learn the weights by running the learning script (which is under `ros_dmp/src`:\n", + "\n", + "```\n", + "python learn_motion_primitive.py\n", + "```\n", + "\n", + "The learning script will ask for the name of the trajectory file (just the name of the file, not the path); the DMP weights will then be saved to a YAML file under `/data/weights/weights_`.\n", + "\n", + "#### Using a DMP\n", + "\n", + "To use a DMP in the `move_arm` action, the path to the DMP should be passed as value for the `dmp_name` parameter of the action goal. By passing this parameter, MoveIt! will not be used for motion, but the trajectory represented by the DMP will be executed instead. As mentioned at the beginning of this tutorial, DMPs can only be used when moving the end effector to a specified pose.\n", + "\n", + "**Note 1**: The implementation of the component that takes care of executing a trajectory represented by a DMP is [inside the `move_arm` action](https://github.com/b-it-bots/mas_domestic_robotics/blob/devel/mdr_planning/mdr_actions/mdr_manipulation_actions/mdr_move_arm_action/ros/src/mdr_move_arm_action/dmp.py); however, as mentioned before, the trajectory represented by the DMP is for the end effector and not in the joint space. The implementation of the inverse kinematics solver that actually allows a manipulator to follow the trajectory can be found in the [`mcr_arm_cartesian_control` package](https://github.com/b-it-bots/mas_common_robotics/tree/kinetic/mcr_manipulation/mcr_arm_cartesian_control).\n", + "\n", + "**Note 2**: Due to limitations in the current implementation, we can only reliably reproduce trajectories with a sideways approach vector; top-down approach vectors are not guaranteed to work well.\n", + "\n", + "#### Tuning the DMP Parameters\n", + "\n", + "When using the `move_arm` action with a learned DMP, users can only set the $\\tau$ parameter - `dmp_tau` is one of the action goal parameters - which controls the duration of motion, but also the accuracy of execution: a value of $1$ means that the velocity will match the velocity of the demonstration, while higher values will increase the velocity, but may cause oscillations and inaccuracies in reaching a goal. For the HSR robot, $\\tau = 30$ has been found to provide a good balance between accuracy and speed.\n", + "\n", + "During learning, another important parameter is $N$, the number of basis function used for representing the DMP forcing term. A small number of basis function generally leads to a crude representation of the demonstrated trajectory, but reduces the computational cost during reproduction. On the other hand, a large number of basis functions increases the accuracy of the representation, but also the complexity of the reproduction; in addition, a value of $N$ that is very high will lead to learning the noise in the demonstrated trajectory. For these reasons, the value of $N$ is best determined experimentally and may need to vary from trajectory to trajectory, such that values $N = 300$ or $N = 500$ may be good initial guesses." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Arm Motion Use Cases\n", + "\n", + "\n", + "### Picking Up Objects\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_arm_action move_arm.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_base_action move_base.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_arm_joints_action move_arm_joints.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_forward_action move_forward.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_pickup_action pickup_action.launch\n", + "```\n", + "\n", + "```\n", + "rosrun mdr_pickup_action pickup_action_client_test\n", + "```\n", + "\n", + "### Placing Objects\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_arm_action move_arm.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_base_action move_base.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_arm_joints_action move_arm_joints.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_move_forward_action move_forward.launch\n", + "```\n", + "\n", + "```\n", + "roslaunch mas_hsr_place_action place_action.launch\n", + "```\n", + "\n", + "```\n", + "rosrun mdr_place_action place_action_client_test\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## About This Tutorial\n", + "\n", + "**Author(s)**: Alex Mitrevski\n", + "\n", + "**Last update**: 23.02.2019" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.5.2" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}