-
 

reach_ros package from reach_ros repo

reach_ros

Package Summary

Tags No category tags.
Version 1.4.0
License Apache 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/ros-industrial/reach_ros2.git
VCS Type git
VCS Version master
Last Updated 2024-11-21
Dev Status DEVELOPED
CI status No Continuous Integration
Released RELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

The reach_ros package

Additional Links

No additional links.

Maintainers

  • mripperger

Authors

  • mripperger

REACH ROS Plugins

Ubuntu

This package contains the ROS2-based plugin implemenations of REACH kinematics, evaluation, and display interfaces

REACH ROS

Installation

First, clone the repository into a colcon workspace

cd ~/reach_ws/src
git clone https://github.com/ros-industrial/reach_ros2.git
cd ..

Install the dependencies

vcs import src < src/reach_ros2/dependencies.repos
rosdep install --from-paths src --ignore-src -r -y

Build the repository

colcon build --symlink-install

Demo

A simple demonstration of the capability of this repository is provided in the demo sub-directory. See the instructions for details on how to launch the demo from ROS 2 and see the Python instructions to see how you can use the Python interface to launch studies.

Usage

Use the following steps to run a reach study with a robot using the ROS1 infrastructure and plugins.

  1. Create any files describing your robot system required by the REACH plugins (e.g., URDF, SRDF, kinematics file, joint limits file, etc.)
  2. Create a mesh model of the workpiece

    Note: the origin of this model should align with the kinematic base frame of the robot

  3. Create a point cloud of the target points on the workpiece
    • This point cloud can be generated using a command line tool from PCL 1.8:
      pcl_mesh_sampling <workpiece_mesh>.ply <output_cloud>.pcd -n_samples <number of samples> -leaf_size <leaf_size> -write_normals true
      
  1. Create a configuration YAML file defining the parameters of the reach study and the configuration of the interface plugins (see this demo example)
  2. Run the setup launch file
    ros2 launch reach_ros setup.launch robot_description_file:=<path_to_URDF>
    
  1. Run the reach study analysis
    ros2 launch reach_ros start.launch \
      robot_description_file:=<path_to_URDF> \
      robot_description_semantic_file:=<path_to_SRDF> \
      robot_description_kinematics_file:=<path_to_kinematics.yaml> \
      robot_description_joint_limits_file:=<path_to_joint_limits.yaml> \
      config_file:=<config_file.yaml> \
      config_name:=<arbitrary_config> \
      results_dir:=<arbitrary_results_directory>
    

Evaluation Plugins

Manipulability

This plugin uses MoveIt! to calculate the manipulability of a robot pose. Higher manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Manipulability Scaled

This plugin uses MoveIt! to calculate the manipulability of a robot pose divided by the characteristic length of the motion group. The characteristic length is computed by walking from the base link to the tip link of the motion group and summing the distances between adjacent links. Higher scaled manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only
  • excluded_links (optional)
    • The names of links contained in the motion group that should not contribute to the characteristic length

Manipulability Ratio

This plugin uses MoveIt! to calculate the manipulability of a robot pose and evaluate a score. The score is calculated as the ratio of the smallest manipulability value to the largest manipulability value. The larger this ratio, the more uniform the dexterity and the higher the score. Range [0, 1]

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Distance Penalty

This plugin uses the MoveIt! collision environment to calculate the distance to closest collision for a robot pose. That distance value is then used to score the robot pose. Larger distance to closest collision results in higher pose score. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning group to be used to solve the robot’s inverse kinematics
  • distance_threshold
    • The distance between 2 closest surfaces to collision under which an inverse kinematics solution will be considered invalid
  • collision_mesh_filename (optional)
    • The filename (in ROS package URI format) of the reach object mesh to be used to do collision checking
    • Example: `package:////.stl
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The names of the robot links with which the reach object mesh is allowed to collide
  • exponent
    • score = min(abs(closest_distance_to_collision / distance_threshold), 1.0)^exponent.

Joint Penalty

This plugin uses the MoveIt! robot model to calculate a robot pose score based on how much the pose deviates from the center of the joint range. Robot poses that are closer to the center of the joint range result in higher pose scores. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning_group with which to evaluate the joint penalty

IK Solvers

MoveIt! IK Solver

This plugin uses MoveIt! kinematics solvers and collision checking to produce collision aware IK solutions

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses

Discretized MoveIt! IK Solver

This plugin performs the same function as the MoveIt! IK solver plugin above, but calculates IK solutions for a target that has been discretized about its Z-axis by an input angle. The pose with the best score is returned.

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses
  • discretization_angle
    • The angle (between 0 and pi, in radians) with which to sample each target pose about the Z-axis

Display Plugins

ROS Reach Display

This plugin publishes joint state and visualization markers to display the results of the reach study. The markers are interactive:

  • Left-clicking on a marker will change the robot position to the IK solution
    • This works by publishing a sensor_msgs/JointState message to the reach_joints topic, which is a source topic for the joint_state_publisher
  • Right-clicking on the marker will bring up a context menu that displays the score of the reach target

The markers are colorized with a heat-map, where red represents the highest scores (i.e., hottest) and blue represents the lowest scores (i.e., coldest). Unreachable points are colorized black. There are two methods for computing the values for the heat map:

  1. By default, the values for the heat map are computed as the ratio of the individual scores to the maximum score. Thus if the scores of the reach study are all fairly consistent and close to the maximum, then the colors of the markers will be mostly red (with the target with the maximum score being the deepest hue of red). If the scores of the reach study targets differ greatly, more variation in color will be seen. Usually this method produces the most meaningful display.
  2. The values can also be computed using the “full color range”, where the scores are fully normalized before colorization. The lowest score (regardless of value) is always displayed as the deepest hue of blue and the the highest score is always shown as the deepest hue of red. This is valuable for highlighting differences in reachability but can be misleading due to the normalization of the scores.

Parameters:

  • collision_mesh_filename
    • The file path to the collision mesh model of the workpiece, in the package:// or file:// URI format
  • collision_mesh_frame
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to kinematic_base_frame
  • kinematic_base_frame
    • The base frame of the kinematic tree in which to display the interactive markers
  • marker_scale
    • The length (in meters) of the arrow markers representing the target Cartesian points
  • use_full_color_range (optional, default: False)
    • Colorize the heat map using the full range of colors (such that the target with the lowest score is the deepest hue of blue, and the target with the highest score is the deepest hue of red)
  • hue_low_score (optional, default: 270.0)
    • Allows changing the default heatmap style coloring to a different color scale in HSV space (e.g. hue_low_score: 0.0, hue_high_score: 180.0 will lead to a red to green color scale).
  • hue_high_score (optional, default: 0.0)
    • See hue_low_score for an explanation.

Target Pose Generator Plugins

Transformed Point Cloud Target Pose Generator

This plugin inherits reach::PointCloudTargetPoseGenerator, which creates target reach study waypoints from a point cloud file with normals, and transforms the point cloud points into a desired URDF frame (typically the kinematic base frame of the robot).

Parameters:

  • pcd_file
    • The path to the point cloud file (.pcd format), in the package:// or file:// URI
  • points_frame
    • The frame of the URDF to which the point cloud points are relative (i.e. the origin frame of the point cloud)
  • target_frame
    • The frame into which the point cloud points should be transformed for the reach study purposes. This should be the kinematic base frame of the robot
CHANGELOG

Changelog for package reach_ros

1.4.0 (2023-10-04)

  • IK Solver Update (#30)
    • Migrate utility function to reach
    • Replace deprecated function
    • Check that IK solver has been loaded for move group
    • Run format jobs on 20.04
    • Ran clang format
  • Contributors: Michael Ripperger

1.3.5 (2023-10-04)

  • Export dependency on reach (#29)
  • Contributors: Michael Ripperger

1.3.4 (2023-09-25)

  • Python CMake Update (#27)
    • Ensure PYTHON_INSTALL_DIR is set by ament
    • Replace ROS version compile definition with preprocessor check
  • CI Update (#26)
    • Push docker image on ref_type tag
    • Added AFTER_SCRIPT to CI
  • Contributors: Michael Ripperger

1.3.3 (2023-09-22)

  • Use YAML anchors in reach study config files to simplify (#24)
  • Run CI workflow on tag push and workflow dispatch (#25)
  • Contributors: Michael Ripperger

1.3.2 (2023-09-14)

  • Updated CMake version for if STRGREATER_EQUAL command (#23)
  • Contributors: Michael Ripperger

1.3.1 (2023-09-07)

  • Updated python module install (#22)
  • Contributors: Michael Ripperger

1.3.0 (2023-08-16)

  • Add customizable color scale for visualization (#18)
  • Contributors: Marc Bestmann

1.2.0 (2023-08-16)

  • Added Python equivalent of the c++ reach study node
  • Added a Python interface to allow usage of ROS2 plugins in Python
  • Revised ROS node singleton pattern
  • Remove the plugin library from ament_target_export so downstream packages cannot link against it
  • Link executable against main library instead of plugin library
  • Fixed typo in plugin export
  • Contributors: Marc Bestmann, Michael Ripperger

1.1.0 (2023-06-13)

  • Updated to REACH 1.3.0
  • Clean up config/launch files
  • Updated demo results
  • Updated distance penalty to cap at 1.0
  • Updated collision distance evaluator to optionally accept collision mesh; updated README
  • Added comments about collision_mesh_frame parameter
  • Fix deps install in README
  • Updated to reach 1.2.0
  • Updated README
  • Added transformed point cloud target pose generator
  • Exposed interface implementations separately from plugin library
  • Fixed collision mesh frame key
  • Fixed docker commit name in CI configuration
  • Updated node to join executor thread properly
  • Added reach plugin library name to environment variable for discovery by the plugin loader
  • Updated to reach 1.1.0
  • Updated to reach 0.1.1
  • Merge pull request #2 from marip8/ros2 ROS2 Support
  • Contributors: Marc Bestmann, Michael Ripperger

1.0.0 (2023-05-10)

  • Move ROS1 components (#44)
  • Contributors: Michael Ripperger

Wiki Tutorials

This package does not provide any links to tutorials in it's rosindex metadata. You can check on the ROS Wiki Tutorials page for the package.

Launch files

No launch files found

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged reach_ros at Robotics Stack Exchange

No version for distro iron. Known supported distros are highlighted in the buttons above.

reach_ros package from reach_ros repo

reach_ros

Package Summary

Tags No category tags.
Version 1.4.0
License Apache 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/ros-industrial/reach_ros2.git
VCS Type git
VCS Version master
Last Updated 2024-11-21
Dev Status DEVELOPED
CI status No Continuous Integration
Released UNRELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

The reach_ros package

Additional Links

No additional links.

Maintainers

  • mripperger

Authors

  • mripperger

REACH ROS Plugins

Ubuntu

This package contains the ROS2-based plugin implemenations of REACH kinematics, evaluation, and display interfaces

REACH ROS

Installation

First, clone the repository into a colcon workspace

cd ~/reach_ws/src
git clone https://github.com/ros-industrial/reach_ros2.git
cd ..

Install the dependencies

vcs import src < src/reach_ros2/dependencies.repos
rosdep install --from-paths src --ignore-src -r -y

Build the repository

colcon build --symlink-install

Demo

A simple demonstration of the capability of this repository is provided in the demo sub-directory. See the instructions for details on how to launch the demo from ROS 2 and see the Python instructions to see how you can use the Python interface to launch studies.

Usage

Use the following steps to run a reach study with a robot using the ROS1 infrastructure and plugins.

  1. Create any files describing your robot system required by the REACH plugins (e.g., URDF, SRDF, kinematics file, joint limits file, etc.)
  2. Create a mesh model of the workpiece

    Note: the origin of this model should align with the kinematic base frame of the robot

  3. Create a point cloud of the target points on the workpiece
    • This point cloud can be generated using a command line tool from PCL 1.8:
      pcl_mesh_sampling <workpiece_mesh>.ply <output_cloud>.pcd -n_samples <number of samples> -leaf_size <leaf_size> -write_normals true
      
  1. Create a configuration YAML file defining the parameters of the reach study and the configuration of the interface plugins (see this demo example)
  2. Run the setup launch file
    ros2 launch reach_ros setup.launch robot_description_file:=<path_to_URDF>
    
  1. Run the reach study analysis
    ros2 launch reach_ros start.launch \
      robot_description_file:=<path_to_URDF> \
      robot_description_semantic_file:=<path_to_SRDF> \
      robot_description_kinematics_file:=<path_to_kinematics.yaml> \
      robot_description_joint_limits_file:=<path_to_joint_limits.yaml> \
      config_file:=<config_file.yaml> \
      config_name:=<arbitrary_config> \
      results_dir:=<arbitrary_results_directory>
    

Evaluation Plugins

Manipulability

This plugin uses MoveIt! to calculate the manipulability of a robot pose. Higher manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Manipulability Scaled

This plugin uses MoveIt! to calculate the manipulability of a robot pose divided by the characteristic length of the motion group. The characteristic length is computed by walking from the base link to the tip link of the motion group and summing the distances between adjacent links. Higher scaled manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only
  • excluded_links (optional)
    • The names of links contained in the motion group that should not contribute to the characteristic length

Manipulability Ratio

This plugin uses MoveIt! to calculate the manipulability of a robot pose and evaluate a score. The score is calculated as the ratio of the smallest manipulability value to the largest manipulability value. The larger this ratio, the more uniform the dexterity and the higher the score. Range [0, 1]

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Distance Penalty

This plugin uses the MoveIt! collision environment to calculate the distance to closest collision for a robot pose. That distance value is then used to score the robot pose. Larger distance to closest collision results in higher pose score. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning group to be used to solve the robot’s inverse kinematics
  • distance_threshold
    • The distance between 2 closest surfaces to collision under which an inverse kinematics solution will be considered invalid
  • collision_mesh_filename (optional)
    • The filename (in ROS package URI format) of the reach object mesh to be used to do collision checking
    • Example: `package:////.stl
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The names of the robot links with which the reach object mesh is allowed to collide
  • exponent
    • score = min(abs(closest_distance_to_collision / distance_threshold), 1.0)^exponent.

Joint Penalty

This plugin uses the MoveIt! robot model to calculate a robot pose score based on how much the pose deviates from the center of the joint range. Robot poses that are closer to the center of the joint range result in higher pose scores. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning_group with which to evaluate the joint penalty

IK Solvers

MoveIt! IK Solver

This plugin uses MoveIt! kinematics solvers and collision checking to produce collision aware IK solutions

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses

Discretized MoveIt! IK Solver

This plugin performs the same function as the MoveIt! IK solver plugin above, but calculates IK solutions for a target that has been discretized about its Z-axis by an input angle. The pose with the best score is returned.

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses
  • discretization_angle
    • The angle (between 0 and pi, in radians) with which to sample each target pose about the Z-axis

Display Plugins

ROS Reach Display

This plugin publishes joint state and visualization markers to display the results of the reach study. The markers are interactive:

  • Left-clicking on a marker will change the robot position to the IK solution
    • This works by publishing a sensor_msgs/JointState message to the reach_joints topic, which is a source topic for the joint_state_publisher
  • Right-clicking on the marker will bring up a context menu that displays the score of the reach target

The markers are colorized with a heat-map, where red represents the highest scores (i.e., hottest) and blue represents the lowest scores (i.e., coldest). Unreachable points are colorized black. There are two methods for computing the values for the heat map:

  1. By default, the values for the heat map are computed as the ratio of the individual scores to the maximum score. Thus if the scores of the reach study are all fairly consistent and close to the maximum, then the colors of the markers will be mostly red (with the target with the maximum score being the deepest hue of red). If the scores of the reach study targets differ greatly, more variation in color will be seen. Usually this method produces the most meaningful display.
  2. The values can also be computed using the “full color range”, where the scores are fully normalized before colorization. The lowest score (regardless of value) is always displayed as the deepest hue of blue and the the highest score is always shown as the deepest hue of red. This is valuable for highlighting differences in reachability but can be misleading due to the normalization of the scores.

Parameters:

  • collision_mesh_filename
    • The file path to the collision mesh model of the workpiece, in the package:// or file:// URI format
  • collision_mesh_frame
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to kinematic_base_frame
  • kinematic_base_frame
    • The base frame of the kinematic tree in which to display the interactive markers
  • marker_scale
    • The length (in meters) of the arrow markers representing the target Cartesian points
  • use_full_color_range (optional, default: False)
    • Colorize the heat map using the full range of colors (such that the target with the lowest score is the deepest hue of blue, and the target with the highest score is the deepest hue of red)
  • hue_low_score (optional, default: 270.0)
    • Allows changing the default heatmap style coloring to a different color scale in HSV space (e.g. hue_low_score: 0.0, hue_high_score: 180.0 will lead to a red to green color scale).
  • hue_high_score (optional, default: 0.0)
    • See hue_low_score for an explanation.

Target Pose Generator Plugins

Transformed Point Cloud Target Pose Generator

This plugin inherits reach::PointCloudTargetPoseGenerator, which creates target reach study waypoints from a point cloud file with normals, and transforms the point cloud points into a desired URDF frame (typically the kinematic base frame of the robot).

Parameters:

  • pcd_file
    • The path to the point cloud file (.pcd format), in the package:// or file:// URI
  • points_frame
    • The frame of the URDF to which the point cloud points are relative (i.e. the origin frame of the point cloud)
  • target_frame
    • The frame into which the point cloud points should be transformed for the reach study purposes. This should be the kinematic base frame of the robot
CHANGELOG

Changelog for package reach_ros

1.4.0 (2023-10-04)

  • IK Solver Update (#30)
    • Migrate utility function to reach
    • Replace deprecated function
    • Check that IK solver has been loaded for move group
    • Run format jobs on 20.04
    • Ran clang format
  • Contributors: Michael Ripperger

1.3.5 (2023-10-04)

  • Export dependency on reach (#29)
  • Contributors: Michael Ripperger

1.3.4 (2023-09-25)

  • Python CMake Update (#27)
    • Ensure PYTHON_INSTALL_DIR is set by ament
    • Replace ROS version compile definition with preprocessor check
  • CI Update (#26)
    • Push docker image on ref_type tag
    • Added AFTER_SCRIPT to CI
  • Contributors: Michael Ripperger

1.3.3 (2023-09-22)

  • Use YAML anchors in reach study config files to simplify (#24)
  • Run CI workflow on tag push and workflow dispatch (#25)
  • Contributors: Michael Ripperger

1.3.2 (2023-09-14)

  • Updated CMake version for if STRGREATER_EQUAL command (#23)
  • Contributors: Michael Ripperger

1.3.1 (2023-09-07)

  • Updated python module install (#22)
  • Contributors: Michael Ripperger

1.3.0 (2023-08-16)

  • Add customizable color scale for visualization (#18)
  • Contributors: Marc Bestmann

1.2.0 (2023-08-16)

  • Added Python equivalent of the c++ reach study node
  • Added a Python interface to allow usage of ROS2 plugins in Python
  • Revised ROS node singleton pattern
  • Remove the plugin library from ament_target_export so downstream packages cannot link against it
  • Link executable against main library instead of plugin library
  • Fixed typo in plugin export
  • Contributors: Marc Bestmann, Michael Ripperger

1.1.0 (2023-06-13)

  • Updated to REACH 1.3.0
  • Clean up config/launch files
  • Updated demo results
  • Updated distance penalty to cap at 1.0
  • Updated collision distance evaluator to optionally accept collision mesh; updated README
  • Added comments about collision_mesh_frame parameter
  • Fix deps install in README
  • Updated to reach 1.2.0
  • Updated README
  • Added transformed point cloud target pose generator
  • Exposed interface implementations separately from plugin library
  • Fixed collision mesh frame key
  • Fixed docker commit name in CI configuration
  • Updated node to join executor thread properly
  • Added reach plugin library name to environment variable for discovery by the plugin loader
  • Updated to reach 1.1.0
  • Updated to reach 0.1.1
  • Merge pull request #2 from marip8/ros2 ROS2 Support
  • Contributors: Marc Bestmann, Michael Ripperger

1.0.0 (2023-05-10)

  • Move ROS1 components (#44)
  • Contributors: Michael Ripperger

Wiki Tutorials

This package does not provide any links to tutorials in it's rosindex metadata. You can check on the ROS Wiki Tutorials page for the package.

Launch files

No launch files found

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged reach_ros at Robotics Stack Exchange

reach_ros package from reach_ros repo

reach_ros

Package Summary

Tags No category tags.
Version 1.4.0
License Apache 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/ros-industrial/reach_ros2.git
VCS Type git
VCS Version master
Last Updated 2024-11-21
Dev Status DEVELOPED
CI status No Continuous Integration
Released UNRELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

The reach_ros package

Additional Links

No additional links.

Maintainers

  • mripperger

Authors

  • mripperger

REACH ROS Plugins

Ubuntu

This package contains the ROS2-based plugin implemenations of REACH kinematics, evaluation, and display interfaces

REACH ROS

Installation

First, clone the repository into a colcon workspace

cd ~/reach_ws/src
git clone https://github.com/ros-industrial/reach_ros2.git
cd ..

Install the dependencies

vcs import src < src/reach_ros2/dependencies.repos
rosdep install --from-paths src --ignore-src -r -y

Build the repository

colcon build --symlink-install

Demo

A simple demonstration of the capability of this repository is provided in the demo sub-directory. See the instructions for details on how to launch the demo from ROS 2 and see the Python instructions to see how you can use the Python interface to launch studies.

Usage

Use the following steps to run a reach study with a robot using the ROS1 infrastructure and plugins.

  1. Create any files describing your robot system required by the REACH plugins (e.g., URDF, SRDF, kinematics file, joint limits file, etc.)
  2. Create a mesh model of the workpiece

    Note: the origin of this model should align with the kinematic base frame of the robot

  3. Create a point cloud of the target points on the workpiece
    • This point cloud can be generated using a command line tool from PCL 1.8:
      pcl_mesh_sampling <workpiece_mesh>.ply <output_cloud>.pcd -n_samples <number of samples> -leaf_size <leaf_size> -write_normals true
      
  1. Create a configuration YAML file defining the parameters of the reach study and the configuration of the interface plugins (see this demo example)
  2. Run the setup launch file
    ros2 launch reach_ros setup.launch robot_description_file:=<path_to_URDF>
    
  1. Run the reach study analysis
    ros2 launch reach_ros start.launch \
      robot_description_file:=<path_to_URDF> \
      robot_description_semantic_file:=<path_to_SRDF> \
      robot_description_kinematics_file:=<path_to_kinematics.yaml> \
      robot_description_joint_limits_file:=<path_to_joint_limits.yaml> \
      config_file:=<config_file.yaml> \
      config_name:=<arbitrary_config> \
      results_dir:=<arbitrary_results_directory>
    

Evaluation Plugins

Manipulability

This plugin uses MoveIt! to calculate the manipulability of a robot pose. Higher manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Manipulability Scaled

This plugin uses MoveIt! to calculate the manipulability of a robot pose divided by the characteristic length of the motion group. The characteristic length is computed by walking from the base link to the tip link of the motion group and summing the distances between adjacent links. Higher scaled manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only
  • excluded_links (optional)
    • The names of links contained in the motion group that should not contribute to the characteristic length

Manipulability Ratio

This plugin uses MoveIt! to calculate the manipulability of a robot pose and evaluate a score. The score is calculated as the ratio of the smallest manipulability value to the largest manipulability value. The larger this ratio, the more uniform the dexterity and the higher the score. Range [0, 1]

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Distance Penalty

This plugin uses the MoveIt! collision environment to calculate the distance to closest collision for a robot pose. That distance value is then used to score the robot pose. Larger distance to closest collision results in higher pose score. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning group to be used to solve the robot’s inverse kinematics
  • distance_threshold
    • The distance between 2 closest surfaces to collision under which an inverse kinematics solution will be considered invalid
  • collision_mesh_filename (optional)
    • The filename (in ROS package URI format) of the reach object mesh to be used to do collision checking
    • Example: `package:////.stl
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The names of the robot links with which the reach object mesh is allowed to collide
  • exponent
    • score = min(abs(closest_distance_to_collision / distance_threshold), 1.0)^exponent.

Joint Penalty

This plugin uses the MoveIt! robot model to calculate a robot pose score based on how much the pose deviates from the center of the joint range. Robot poses that are closer to the center of the joint range result in higher pose scores. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning_group with which to evaluate the joint penalty

IK Solvers

MoveIt! IK Solver

This plugin uses MoveIt! kinematics solvers and collision checking to produce collision aware IK solutions

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses

Discretized MoveIt! IK Solver

This plugin performs the same function as the MoveIt! IK solver plugin above, but calculates IK solutions for a target that has been discretized about its Z-axis by an input angle. The pose with the best score is returned.

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses
  • discretization_angle
    • The angle (between 0 and pi, in radians) with which to sample each target pose about the Z-axis

Display Plugins

ROS Reach Display

This plugin publishes joint state and visualization markers to display the results of the reach study. The markers are interactive:

  • Left-clicking on a marker will change the robot position to the IK solution
    • This works by publishing a sensor_msgs/JointState message to the reach_joints topic, which is a source topic for the joint_state_publisher
  • Right-clicking on the marker will bring up a context menu that displays the score of the reach target

The markers are colorized with a heat-map, where red represents the highest scores (i.e., hottest) and blue represents the lowest scores (i.e., coldest). Unreachable points are colorized black. There are two methods for computing the values for the heat map:

  1. By default, the values for the heat map are computed as the ratio of the individual scores to the maximum score. Thus if the scores of the reach study are all fairly consistent and close to the maximum, then the colors of the markers will be mostly red (with the target with the maximum score being the deepest hue of red). If the scores of the reach study targets differ greatly, more variation in color will be seen. Usually this method produces the most meaningful display.
  2. The values can also be computed using the “full color range”, where the scores are fully normalized before colorization. The lowest score (regardless of value) is always displayed as the deepest hue of blue and the the highest score is always shown as the deepest hue of red. This is valuable for highlighting differences in reachability but can be misleading due to the normalization of the scores.

Parameters:

  • collision_mesh_filename
    • The file path to the collision mesh model of the workpiece, in the package:// or file:// URI format
  • collision_mesh_frame
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to kinematic_base_frame
  • kinematic_base_frame
    • The base frame of the kinematic tree in which to display the interactive markers
  • marker_scale
    • The length (in meters) of the arrow markers representing the target Cartesian points
  • use_full_color_range (optional, default: False)
    • Colorize the heat map using the full range of colors (such that the target with the lowest score is the deepest hue of blue, and the target with the highest score is the deepest hue of red)
  • hue_low_score (optional, default: 270.0)
    • Allows changing the default heatmap style coloring to a different color scale in HSV space (e.g. hue_low_score: 0.0, hue_high_score: 180.0 will lead to a red to green color scale).
  • hue_high_score (optional, default: 0.0)
    • See hue_low_score for an explanation.

Target Pose Generator Plugins

Transformed Point Cloud Target Pose Generator

This plugin inherits reach::PointCloudTargetPoseGenerator, which creates target reach study waypoints from a point cloud file with normals, and transforms the point cloud points into a desired URDF frame (typically the kinematic base frame of the robot).

Parameters:

  • pcd_file
    • The path to the point cloud file (.pcd format), in the package:// or file:// URI
  • points_frame
    • The frame of the URDF to which the point cloud points are relative (i.e. the origin frame of the point cloud)
  • target_frame
    • The frame into which the point cloud points should be transformed for the reach study purposes. This should be the kinematic base frame of the robot
CHANGELOG

Changelog for package reach_ros

1.4.0 (2023-10-04)

  • IK Solver Update (#30)
    • Migrate utility function to reach
    • Replace deprecated function
    • Check that IK solver has been loaded for move group
    • Run format jobs on 20.04
    • Ran clang format
  • Contributors: Michael Ripperger

1.3.5 (2023-10-04)

  • Export dependency on reach (#29)
  • Contributors: Michael Ripperger

1.3.4 (2023-09-25)

  • Python CMake Update (#27)
    • Ensure PYTHON_INSTALL_DIR is set by ament
    • Replace ROS version compile definition with preprocessor check
  • CI Update (#26)
    • Push docker image on ref_type tag
    • Added AFTER_SCRIPT to CI
  • Contributors: Michael Ripperger

1.3.3 (2023-09-22)

  • Use YAML anchors in reach study config files to simplify (#24)
  • Run CI workflow on tag push and workflow dispatch (#25)
  • Contributors: Michael Ripperger

1.3.2 (2023-09-14)

  • Updated CMake version for if STRGREATER_EQUAL command (#23)
  • Contributors: Michael Ripperger

1.3.1 (2023-09-07)

  • Updated python module install (#22)
  • Contributors: Michael Ripperger

1.3.0 (2023-08-16)

  • Add customizable color scale for visualization (#18)
  • Contributors: Marc Bestmann

1.2.0 (2023-08-16)

  • Added Python equivalent of the c++ reach study node
  • Added a Python interface to allow usage of ROS2 plugins in Python
  • Revised ROS node singleton pattern
  • Remove the plugin library from ament_target_export so downstream packages cannot link against it
  • Link executable against main library instead of plugin library
  • Fixed typo in plugin export
  • Contributors: Marc Bestmann, Michael Ripperger

1.1.0 (2023-06-13)

  • Updated to REACH 1.3.0
  • Clean up config/launch files
  • Updated demo results
  • Updated distance penalty to cap at 1.0
  • Updated collision distance evaluator to optionally accept collision mesh; updated README
  • Added comments about collision_mesh_frame parameter
  • Fix deps install in README
  • Updated to reach 1.2.0
  • Updated README
  • Added transformed point cloud target pose generator
  • Exposed interface implementations separately from plugin library
  • Fixed collision mesh frame key
  • Fixed docker commit name in CI configuration
  • Updated node to join executor thread properly
  • Added reach plugin library name to environment variable for discovery by the plugin loader
  • Updated to reach 1.1.0
  • Updated to reach 0.1.1
  • Merge pull request #2 from marip8/ros2 ROS2 Support
  • Contributors: Marc Bestmann, Michael Ripperger

1.0.0 (2023-05-10)

  • Move ROS1 components (#44)
  • Contributors: Michael Ripperger

Wiki Tutorials

This package does not provide any links to tutorials in it's rosindex metadata. You can check on the ROS Wiki Tutorials page for the package.

Launch files

No launch files found

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged reach_ros at Robotics Stack Exchange

reach_ros package from reach_ros repo

reach_ros

Package Summary

Tags No category tags.
Version 1.1.4
License Apache 2.0
Build type CATKIN
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/ros-industrial/reach_ros.git
VCS Type git
VCS Version master
Last Updated 2024-06-15
Dev Status DEVELOPED
CI status
Released RELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

The reach_ros package

Additional Links

No additional links.

Maintainers

  • mripperger

Authors

  • mripperger

REACH ROS Plugins

Focal Noetic

This package contains the ROS1-based plugin implemenations of REACH kinematics, evaluation, and display interfaces

REACH ROS

Installation

First, clone the repository into a catkin workspace

cd ~/reach_ws/src
git clone https://github.com/ros-industrial/reach_ros.git
cd ..

Install the dependencies

vcs import src < src/reach_ros/dependencies.repos
rosdep install --from-paths src --ignore-src -r -y

Build the repository

catkin build

Demo

A simple demonstration of the capability of this repository is provided in the demo sub-directory. See the instructions for details on how to run the demo.

Usage

Use the following steps to run a reach study with a robot using the ROS1 infrastructure and plugins.

  1. Create a URDF of your robot system
  2. Create a launch file to load the URDF, SRDF, and other required parameters (e.g. related to kinematics, joint, limits) to the parameter server (see this demo example file)
  3. Create a mesh model of the workpiece

    Note: the origin of this model should align with the kinematic base frame of the robot

  4. Create a point cloud of the target points on the workpiece
    • This point cloud can be generated using a command line tool from PCL 1.8:
      pcl_mesh_sampling <workpiece_mesh>.ply <output_cloud>.pcd -n_samples <number of samples> -leaf_size <leaf_size> -write_normals true
      
  1. Create a configuration YAML file defining the parameters of the reach study and the configuration of the interface plugins (see this demo example)
  2. Run the setup launch file
    roslaunch reach_ros setup.launch robot:=<load_robot_parameters>.launch
    
  1. Run the reach study analysis
    roslaunch reach_ros start.launch config_file:=<config_file.yaml> config_name:=<arbitrary_config>
    

Evaluation Plugins

Manipulability

This plugin uses MoveIt! to calculate the manipulability of a robot pose. Higher manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Manipulability Scaled

This plugin uses MoveIt! to calculate the manipulability of a robot pose divided by the characteristic length of the motion group. The characteristic length is computed by walking from the base link to the tip link of the motion group and summing the distances between adjacent links. Higher scaled manipulability results in a higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only
  • excluded_links (optional)
    • The names of links contained in the motion group that should not contribute to the characteristic length

Manipulability Ratio

This plugin uses MoveIt! to calculate the manipulability of a robot pose and evaluate a score. The score is calculated as the ratio of the smallest manipulability value to the largest manipulability value. The larger this ratio, the more uniform the dexterity and the higher the score. Range [0, 1]

Parameters:

  • planning_group
    • The name of the planning group with which to evaluate the manipulability of a given robot pose
  • jacobian_row_subset (optional)
    • The indices of the rows of the Jacobian to use when evaluating the manipulability. The row indices should be on [0, 6) and correspond to the output space [x, y, z, rx, ry, rz]
    • Ex. jacobian_row_subset: [0, 1, 2] # position manipulability only

Distance Penalty

This plugin uses the MoveIt! collision environment to calculate the distance to closest collision for a robot pose. That distance value is then used to score the robot pose. Larger distance to closest collision results in higher pose score. Range: [0, inf)

Parameters:

  • planning_group
    • The name of the planning group to be used to solve the robot’s inverse kinematics
  • distance_threshold
    • The distance between 2 closest surfaces to collision under which an inverse kinematics solution will be considered invalid
  • collision_mesh_filename (optional)
    • The filename (in ROS package URI format) of the reach object mesh to be used to do collision checking
    • Example: `package:////.stl
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The names of the robot links with which the reach object mesh is allowed to collide
  • exponent
    • score = min(abs(closest_distance_to_collision / distance_threshold), 1.0)^exponent.

Joint Penalty

This plugin uses the MoveIt! robot model to calculate a robot pose score based on how much the pose deviates from the center of the joint range. Robot poses that are closer to the center of the joint range result in higher pose scores. Range: [0, 1]

Parameters:

  • planning_group
    • The name of the planning_group with which to evaluate the joint penalty

IK Solvers

MoveIt! IK Solver

This plugin uses MoveIt! kinematics solvers and collision checking to produce collision aware IK solutions

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses

Discretized MoveIt! IK Solver

This plugin performs the same function as the MoveIt! IK solver plugin above, but calculates IK solutions for a target that has been discretized about its Z-axis by an input angle. The pose with the best score is returned.

Parameters:

  • planning_group
    • Name of the planning group
  • distance_threshold
    • The distance from nearest collision at which to invalidate an IK solution. For example, if this parameter is set to 0.1m, then IK solutions whose distance to nearest collision is less than 0.1m will be invalidated
  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to the kinematic base frame associated with planning_group
  • touch_links
    • The TF links that are allowed to be in contact with the collision mesh
  • evaluation_plugin
    • The name (and parameters) of the evaluation plugin to be used to score IK solution poses
  • discretization_angle
    • The angle (between 0 and pi, in radians) with which to sample each target pose about the Z-axis

Display Plugins

ROS Reach Display

This plugin publishes joint state and visualization markers to display the results of the reach study. The markers are interactive:

  • Left-clicking on a marker will change the robot position to the IK solution
    • This works by publishing a sensor_msgs/JointState message to the reach_joints topic, which is a source topic for the joint_state_publisher
  • Right-clicking on the marker will bring up a context menu that displays the score of the reach target

The markers are colorized with a heat-map, where red represents the highest scores (i.e., hottest) and blue represents the lowest scores (i.e., coldest). Unreachable points are colorized black. There are two methods for computing the values for the heat map:

  1. By default, the values for the heat map are computed as the ratio of the individual scores to the maximum score. Thus if the scores of the reach study are all fairly consistent and close to the maximum, then the colors of the markers will be mostly red (with the target with the maximum score being the deepest hue of red). If the scores of the reach study targets differ greatly, more variation in color will be seen. Usually this method produces the most meaningful display.
  2. The values can also be computed using the “full color range”, where the scores are fully normalized before colorization. The lowest score (regardless of value) is always displayed as the deepest hue of blue and the the highest score is always shown as the deepest hue of red. This is valuable for highlighting differences in reachability but can be misleading due to the normalization of the scores.

Parameters:

  • collision_mesh_filename (optional)
    • The file path to the collision mesh model of the workpiece, in the package:// or ‘file://’ URI format
  • collision_mesh_frame (optional)
    • The TF frame to which the collision mesh should be attached
    • If left unspecified, the collision mesh will be attached to kinematic_base_frame
  • kinematic_base_frame
    • The base frame of the kinematic tree in which to display the interactive markers
  • marker_scale
    • The length (in meters) of the arrow markers representing the target Cartesian points
  • use_full_color_range (optional, default: False)
    • Colorize the heat map using the full range of colors (such that the target with the lowest score is the deepest hue of blue, and the target with the highest score is the deepest hue of red)

Target Pose Generator Plugins

Transformed Point Cloud Target Pose Generator

This plugin inherits reach::PointCloudTargetPoseGenerator, which creates target reach study waypoints from a point cloud file with normals, and transforms the point cloud points into a desired URDF frame (typically the kinematic base frame of the robot).

Parameters:

  • pcd_file
    • The path to the point cloud file (.pcd format), in the package:// or file:// URI
  • points_frame
    • The frame of the URDF to which the point cloud points are relative (i.e. the origin frame of the point cloud)
  • target_frame
    • The frame into which the point cloud points should be transformed for the reach study purposes. This should be the kinematic base frame of the robot
CHANGELOG

Changelog for package reach_ros

1.1.4 (2024-06-15)

  • Merge pull request #20 from marip8/fix/visualization-speed Fixed speed of loading interactive markers
  • Fixed speed of loading interactive markers
  • Contributors: Michael Ripperger

1.1.3 (2024-06-10)

  • Take self collisions into account in distance_penalty
  • Run format jobs on 20.04
  • Check that IK solver has been loaded for move group
  • Replace deprecated function
  • Migrate utility function to reach
  • Contributors: Iñigo Moreno, Michael Ripperger

1.1.2 (2023-09-22)

  • Merge pull request #15 from ros-industrial/update/demo-yaml Use YAML anchors in reach study config files to simplify
  • Merge pull request #14 from ros-industrial/update/ci Updated CI to run on tag push and workflow dispatch
  • Updated CI to run on tag push and workflow dispatch
  • Use YAML anchors in reach study config files to simplify
  • Contributors: Michael Ripperger

1.1.1 (2023-09-07)

  • Merge pull request #10 from ros-industrial/updates Minor Updates
  • Use RICB in CMakeLists
  • Update function to pass by reference
  • Export reach_ros library target
  • Merge pull request #9 from ros-industrial/fix/plugin Fixed typo in plugin export
  • Fixed typo in plugin export
  • Contributors: Michael Ripperger

1.1.0 (2023-06-13)

  • Merge pull request #8 from marip8/update/reach Updated to REACH 1.3.0
  • Updated to REACH 1.3.0
  • Merge pull request #7 from marip8/update/collision-distance-eval Collision distance evaluator update
  • Regenerated demo results
  • Clipped collision distance penalty cost on [0, 1]
  • Updated collision distance penalty to optionally include collision mesh; updated README
  • Merge pull request #6 from marip8/update/README README updates
  • Added installation instructions
  • Added documentation for the collision_mesh_frame parameter
  • Merge pull request #5 from marip8/updates Updates
  • Updated to reach 1.2.0
  • Updated the README
  • Added transformed point cloud target pose generator
  • Exposed plugins in separate library
  • Fixed collision mesh frame key
  • Merge pull request #4 from marip8/updates Updates
  • Set environment variable so plugin loader can find reach_ros plugins
  • Fixed tf2 issue
  • Merge pull request #3 from marip8/update/deps Updated to reach 0.1.1
  • Updated to reach 1.1.0
  • Merge pull request #2 from marip8/update/changelog Cleared changelog
  • Cleared changelog
  • Merge pull request #1 from marip8/update Replace eigen_conversions dependency with tf2
  • Replace eigen_conversions dependency with tf2
  • Removed the Python tests
  • Added reach to the dependencies.repos
  • Install the demo files
  • Updated README to include documentation about demo and usage
  • Added GIF to README
  • Flattened reach_ros directory
  • Removed docs directory
  • Removed reach directory
  • Contributors: Michael Ripperger

1.0.0 (2023-05-10)

  • Move ROS1 components (#44)
  • Contributors: Michael Ripperger

Wiki Tutorials

This package does not provide any links to tutorials in it's rosindex metadata. You can check on the ROS Wiki Tutorials page for the package.

Launch files

  • demo/config/robot.launch
  • launch/setup.launch
      • rviz [default: true] — Launch Rviz
      • robot — Launch file that loads URDF, SRDF, and other kinematics-related parameters
  • launch/start.launch
      • config_file — YAML configuration file for the reach study
      • config_name [default: reach_study] — Arbitrary configuration name for the reach study
      • results_dir [default: /tmp] — Location in which reach study results will be saved

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged reach_ros at Robotics Stack Exchange

No version for distro ardent. Known supported distros are highlighted in the buttons above.
No version for distro bouncy. Known supported distros are highlighted in the buttons above.
No version for distro crystal. Known supported distros are highlighted in the buttons above.
No version for distro eloquent. Known supported distros are highlighted in the buttons above.
No version for distro dashing. Known supported distros are highlighted in the buttons above.
No version for distro galactic. Known supported distros are highlighted in the buttons above.
No version for distro foxy. Known supported distros are highlighted in the buttons above.
No version for distro lunar. Known supported distros are highlighted in the buttons above.
No version for distro jade. Known supported distros are highlighted in the buttons above.
No version for distro indigo. Known supported distros are highlighted in the buttons above.
No version for distro hydro. Known supported distros are highlighted in the buttons above.
No version for distro kinetic. Known supported distros are highlighted in the buttons above.
No version for distro melodic. Known supported distros are highlighted in the buttons above.