ros-industrial / fanuc_experimental Goto Github PK
View Code? Open in Web Editor NEWExperimental packages for Fanuc manipulators within ROS-Industrial (http://wiki.ros.org/fanuc_experimental)
License: Other
Experimental packages for Fanuc manipulators within ROS-Industrial (http://wiki.ros.org/fanuc_experimental)
License: Other
As per subject.
I'm trying to connect the physical lrmate200id to my computer and issue commands to it. I'm running the command
roslaunch fanuc_lrmate200id_moveit_config moveit_planning_execution.launch sim:=false robot_ip:=10.0.0.93
which results in the error
[FATAL] [1498826383.877209818]: Parameter '~moveit_controller_manager' not specified. This is needed to identify the plugin to use for interacting with controllers. No paths can be executed.
As a side note, the robot state is loaded into RViz correctly, with all the joint positions matching the physical robot's state
I'm opening moveit setup with
roslaunch moveit_setup_assistant setup_assistant.launch
and clicking edit existing package. I load fanuc_lrmate200id_moveit_config. Then I click Planning Groups, then manipulator, then edit selected, and choose fanuc_lrmate200id_manipulator_kinematics/IKFastKinematicsPlugin for the Kinematic Solver. I then press Save, go to Configuration Files, and hit Generate Package. I run catkin_make, and open RViz with
roslaunch fanuc_lrmate200id_moveit_config demo.launch
which results in the robot being loaded on my screen normally, but I can't move it by clicking it or by running code to move it. Using KDLKinematicsPlugin works fine for me, but I would like to use IKFast
fi fanuc_lrmate200ic_moveit_plugins
does not list cmake_modules
as a build_depend
, causing rosdep
to not install it.
Related to moveit/moveit_ikfast#27?
As per subject: when loading the M-20iA MoveIt config the IKFast plugin xml files cannot be found (but only in devel space).
Hello ,
I am using this repository to control fanuc robot. I have found out that the position of end-effector is different in move_group and original robot data. In the below picture at right hand-side I get the end effector pose with ;
move_group.getCurrentPose().pose;
end show it for debugging. Although the rotations seems to match the positions are off the real robot data.
.
I went to see the details of URDF file regarding to joints and links connections. I think I need to modify some places wrt to technical specs of this model as below;
so far my attemps failed and I am quite not sure if it is the reason for the problem.
Could you please give me some feedback on this ?
Edit;
The changes I made for the URDF is here; https://gist.github.com/jediofgever/d8ffdbfce308f24ea6da138d416e6013
after the change the results seems even more off
The axis definition of joint 6 in lrmate200id_macro.xacro is wrong so that joint 6 rotates in the wrong direction. It is currently <axis xyz="1 0 0"/>
but should be <axis xyz="-1 0 0"/>
.
All MoveIt configuration packages have been generated using the Groovy MoveIt setup assistant. All MoveIt cfg packages should be updated using the newer Hydro version of MoveIt.
I'm having connection issues while connecting my laptop running ROS/MoveIt to the physical robot to run movement commands. I'm not sure whether I've configured something wrong or there is a network issue. Attempts to connect to the physical robot only go through about 1/3 of the time, and there's a small chance I'll get disconnected while running my movement commands. Sometimes the robot seems to connect, and RViz will update to match the physical robot's position, but move commands don't reach the robot. Other times, the connection isn't made at all.
Here's the output from the teach pendant, I'm not sure how useful this is:
I also occasionally get this error in my console
Failed to connect to server, rc: -1. Error: 'Transport endpoint is already connected'
This is / was a bug in the moveit_ikfast
package generation templates.
Travis has announced they're migrating everyone to their .com
domain -- and we should all start using their Github App.
In an effort to get migration for this repository going, Travis integration (the old, services based, one) was removed, but migration to the new system is not available yet for fanuc_experimental
.
This is a reminder to setup Travis again if/when the migration becomes possible.
I'd like to use fanuc_lrmate200id on gazebo, but it requires setting inertial value on urdf file.
Are there any data sheets for this?
@marip8: you have files for the 125L
variant in your fork in the model/r2000ic
branch.
Would you be willing to merge those with the fanuc_r2000ic_support
package in this repository?
I'd be mostly interested in the current state, not so much the complete commit history (to avoid repository bloat).
Would it be beneficial to store OPW kinematic parameters in a file in the support packages for robot variants? They're not necessarily hard to set up (and the MoveIt OPW kinematics plugin has some helpful sanity checks for the parameters now), but it can be a bit confusing for those not familiar with the parameters or the means of figuring out what the parameters should be.
Hello is it possible to add the transmission files in the urdf to allow ROS Control to function correctly in gazebo. I know alot of the open PR regarding gazebo did not pass, so is there any push to allow for that.
As needed.
I am having issues with moving my fanuc lrmate 200id with code. I can use the gui in rviz and move the six axis ball which gives me the coordinates and orientation. If i then put those same exact coordinates to compute a Cartesian path. I get multiple errors and warnings.
[ WARN] [1527256660.896593985]: ParallelPlan::solve(): Unable to find solution by any of the threads in 0.000141 seconds
[ERROR] [1527256660.896706601]: manipulator[RRTConnectkConfigDefault]: Insufficient states in sampleable goal region
[ERROR] [1527256660.896736589]: manipulator[RRTConnectkConfigDefault]: Insufficient states in sampleable goal region
[ERROR] [1527256660.896770195]: manipulator[RRTConnectkConfigDefault]: Insufficient states in sampleable goal region
[ERROR] [1527256660.896804013]: manipulator[RRTConnectkConfigDefault]: Insufficient states in sampleable goal region
[ WARN] [1527256660.896847235]: ParallelPlan::solve(): Unable to find solution by any of the threads in 0.000149 seconds
[ INFO] [1527256660.896914156]: Unable to solve the planning problem
A few observations I've made, the planner seems to take a longer path when using RViz, and it tries to plan to a shorter path with the Cartesian path plan.
CR-35iA introduced in 82951cb is current 'fanuc yellow', but it needs to be green (the new yellow).
Hi,
Is it possible to get initial parameters of CRX-10iA/L, I want to simulate the robot in Gazebo.
Thanks
I'm pretty sure that the M-710iC robot definition is wrong because axis 6 is reversed.
The following RViz captures are connected to the real robot with ROSSTATE
Fetched value from /joint_states
corresponds to the value I read on the teach pendant so the problem must be that the axis is reversed in the XACRO/URDF definition of the robot.
I have submitted a pull request to add support for the crx10ial collaborative robot.
Hi,
This is more a question than an issue,
when I set a goal pose for end-effector, the robot cannot be at the goal in precision manufacturer promises. For instance in below picture;
I set goal pose for end effector as ;
robot_x; 400.0
robot_y; 0.0
robot_z; 700.0
robot_rx; 180.0
robot_ry; 0.0
robot_rz; 0.0
However the end effector ends up at ;
robot_x; 400.085405
robot_y; 0.042173
robot_z; 699.966790
robot_rx; 179.953275
robot_ry; 0.008988
robot_rz; -0.021190
is this expected behavior? if not what could be possible reasons for this behavior ?
According to manufacturer's claim the precision is 3 decimals.
They're failing due to the fanuc
package not having been released yet.
The CI job http://jenkins.ros.org/job/devel-hydro-fanuc_experimental/ARCH_PARAM=amd64,UBUNTU_PARAM=precise,label=devel/3/console fails because of the missing dependency on fanuc_m10ia_support
.
As per subject. This would cause the buildfarm to build these packages N times for N architectures, while one package would work for all.
On the teach pendant:
J3: -33.342 (deg)
Using robot_state_visualize
(indigo-devel branch):
$ rostopic echo /joint_states
header:
seq: 4802
stamp:
secs: 1462191929
nsecs: 471022367
frame_id: ''
name: ['joint_1', 'joint_2', 'joint_3', 'joint_4', 'joint_5', 'joint_6']
position: [0.2743533253669739, 0.5467115640640259, -0.035214900970458984, 1.2852824926376343, -1.7649667263031006, -0.5436538457870483]
velocity: []
effort: []
-0.035214900970458984 rad
= -2.017 deg
The current collision meshes of the LR mate 200id are too restrictive.
In particular, I'm talking about the mesh for link2.
Link 2 is supposed to have a bigger opening such that link 4 is partly inside of the current link 2 collision mesh (example of such a joint configuration in radians: [0, 0.3, -1, 0, 1.3, 0]).
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.