nisur's profile - activity

System: Ubuntu 16.04 / ROS Kinetic / Nuvo-5095GC-022 / Universal Robots UR10 Packages: freenect_stack, geometry2, libfreenect, perception_pcl, trac_ik, universal_robot, ur_modern_driver

Hi,

Some background: As part of a dynamic collision avoidance project I'm trying to achieve full visibility of a live scene using multiple Kinect cameras. In the current setup I have two Xbox Kinect V1s, each producing a working point cloud. The cameras are positioned with a 90deg offset and the frames have been linked using tf2, so the points match up relatively well in 3D (cameras are intrinsically calibrated). From there i have used the two point clouds to create an octomap that the robot plans to avoid.

Now that i've got the live octomap sorted i'm running into problems with the integrated self_filter function inside Moveit!, in that only the gripper is being filtered from the scene regardless of the padding/offset values used. If i view the filtered_cloud topic as a point cloud 2, it clearly illustrates the problem.

It's almost like the robot urdf is being ignored due to the addition of the end effector?

UPDATE (01/11/2018). Observations:

1. Removal of the second Octomap updater plugin does not resolve the problem
2. * I've noticed that the Octomap is not clearing correctly. If I move the Kinect, the map is 'filled' with new blocks but old blocks are not removed, meaning that they have to be cleared manually. This is likely to be the cause but I have no idea why this might be happening, i'll look into it when I can.
3. If I copy the UR10 sensor.yaml and launch files into the UR5 package, the Octomap behaves as expected (clearing correctly, self_filter operational). I think the only difference is that the UR10 has a gripper attached with all gripper joints set as passive, the UR5 has no gripper attached.

Any ideas?

My sensors.yaml file looks like:

sensors:
  - sensor_plugin: occupancy_map_monitor/PointCloudOctomapUpdater
    point_cloud_topic: /camera/depth_registered/points
    max_range: 5.0
    point_subsample: 1 # As expected. Every 1 in point_subsample read
    padding_offset: 0.1
    padding_scale: 1.0 # Scale of voxels
    filtered_cloud_topic: filtered_cloud
    shape_padding: 0.5
    shape_scale: 1.0
    self_see_default_padding: 0.1
    self_see_default_scale: 1.0

  - sensor_plugin: occupancy_map_monitor/PointCloudOctomapUpdater
    point_cloud_topic: /camera2/depth_registered/points
    max_range: 5.0
    point_subsample: 1 # As expected. Every 1 in point_subsample read
    padding_offset: 0.1
    padding_scale: 0.1 # Scale of voxels
    filtered_cloud_topic: filtered2_cloud
    shape_padding: 0.3
    shape_scale: 1.0
    self_see_default_padding: 0.5
    self_see_default_scale: 1.0

And my urdf:

<?xml version="1.0"?>
<robot xmlns:xacro="http://ros.org/wiki/xacro">

<!--
  Author: Kelsey Hawkins
  Contributers: Jimmy Da Silva, Ajit Krisshna N L, Muhammad Asif Rana
-->


  <xacro:include filename="$(find ur_description)/urdf/ur.transmission.xacro" />
  <xacro:include filename="$(find ur_description)/urdf/ur.gazebo.xacro" />

<!-- Robotiq --> 
  <xacro:include filename="$(find robotiq_description)/urdf/robotiq_85_gripper.urdf.xacro" />
  <xacro:robotiq_85_gripper parent="ee_link" prefix="gripper">
    <origin xyz="0 0 0" rpy="0 0 0"/>
  </xacro:robotiq_85_gripper>

  <xacro:macro name="cylinder_inertial" params="radius length mass *origin">
    <inertial>
      <mass value="${mass}" />
      <xacro:insert_block name="origin" />
      <inertia ixx="${0.0833333 * mass * (3 * radius * radius ...

Fairly new to this so let me know if there's a better way of doing things!

I'm attempting dynamic collision avoidance with a Kinect V1 and the ur5 moveit package on ROS kinetic. The PC is running Ubuntu 16.04 and i have the following packages installed:

• camera_pose
• octomap_mapping
• octomap_ros
• OpenNI
• openni_camera
• perception_pcl
• robotiq
• trac_ik
• universal_robot
• ur_modern_driver
• ur_scripts

I've been following the ROS tutorial on perception/configuration (link below) in an effort to have moveit process the point cloud into an octomap and display the occupied voxels: http://docs.ros.org/kinetic/api/movei...

And am also aware of this other thread which is extremely similar, however i wasn't able to achieve a working result using just the information here: https://answers.ros.org/question/1957...

I have Openni working so can visualise the pointcloud2 okay in moveit and have completed all the steps in the above, including altering the sensor_manager.launch.yaml files. However when i echo the planning scene topic, /move_group/monitored_planning_scene, i am getting nothing. This is despite trying a number of static transform publishers to arrange everything with respect to /world (I would post images of my tf trees but i dont have enough karma).

The moveit! config files used can be found at: ( https://github.com/nis-ur/ur_kinect_m... )

As requested, the terminal log when i launch ur5 planning:

nvidia@tegra-ubuntu:~/catkin_ws$ roslaunch ur5_moveit_config ur5_moveit_planning_execution.launch sim:=true limited:=true
... logging to /home/nvidia/.ros/log/6ae42ada-891a-11e8-b1d3-62c9d7ce9d78/roslaunch-tegra-ubuntu-7323.log
Checking log directory for disk usage. This may take awhile.
Press Ctrl-C to interrupt
Done checking log file disk usage. Usage is <1GB.

started roslaunch server http://tegra-ubuntu:38298/

SUMMARY
========

PARAMETERS
 * /move_group/Manipulator/longest_valid_segment_fraction: 0.005
 * /move_group/Manipulator/planner_configs: ['SBLkConfigDefau...
 * /move_group/Manipulator/projection_evaluator: joints(shoulder_p...
 * /move_group/allow_trajectory_execution: True
 * /move_group/controller_list: [{'action_ns': 'f...
 * /move_group/jiggle_fraction: 0.05
 * /move_group/max_range: 5.0
 * /move_group/max_safe_path_cost: 1
 * /move_group/moveit_controller_manager: moveit_simple_con...
 * /move_group/moveit_manage_controllers: True
 * /move_group/octomap_resolution: 0.025
 * /move_group/planner_configs/BFMTkConfigDefault/balanced: 0
 * /move_group/planner_configs/BFMTkConfigDefault/cache_cc: 1
 * /move_group/planner_configs/BFMTkConfigDefault/extended_fmt: 1
 * /move_group/planner_configs/BFMTkConfigDefault/heuristics: 1
 * /move_group/planner_configs/BFMTkConfigDefault/nearest_k: 1
 * /move_group/planner_configs/BFMTkConfigDefault/num_samples: 1000
 * /move_group/planner_configs/BFMTkConfigDefault/optimality: 1
 * /move_group/planner_configs/BFMTkConfigDefault/radius_multiplier: 1.0
 * /move_group/planner_configs/BFMTkConfigDefault/type: geometric::BFMT
 * /move_group/planner_configs/BKPIECEkConfigDefault/border_fraction: 0.9
 * /move_group/planner_configs/BKPIECEkConfigDefault/failed_expansion_score_factor: 0.5
 * /move_group/planner_configs/BKPIECEkConfigDefault/min_valid_path_fraction: 0.5
 * /move_group/planner_configs/BKPIECEkConfigDefault/range: 0.0
 * /move_group/planner_configs/BKPIECEkConfigDefault/type: geometric::BKPIECE
 * /move_group/planner_configs/BiESTkConfigDefault/range: 0.0
 * /move_group/planner_configs/BiESTkConfigDefault/type: geometric::BiEST
 * /move_group/planner_configs/BiTRRTkConfigDefault/cost_threshold: 1e300
 * /move_group/planner_configs/BiTRRTkConfigDefault/frountier_node_ratio: 0.1
 * /move_group/planner_configs/BiTRRTkConfigDefault/frountier_threshold: 0.0
 * /move_group/planner_configs/BiTRRTkConfigDefault/init_temperature: 100
 * /move_group/planner_configs/BiTRRTkConfigDefault/range: 0.0
 * /move_group/planner_configs/BiTRRTkConfigDefault/temp_change_factor: 0.1
 * /move_group/planner_configs/BiTRRTkConfigDefault/type: geometric::BiTRRT
 * /move_group/planner_configs/ESTkConfigDefault/goal_bias: 0.05
 * /move_group/planner_configs/ESTkConfigDefault/range: 0.0
 * /move_group/planner_configs/ESTkConfigDefault/type: geometric::EST
 * /move_group/planner_configs/FMTkConfigDefault/cache_cc: 1
 * /move_group/planner_configs/FMTkConfigDefault/extended_fmt: 1
 * /move_group/planner_configs/FMTkConfigDefault/heuristics: 0
 * /move_group/planner_configs/FMTkConfigDefault/nearest_k: 1
 * /move_group/planner_configs/FMTkConfigDefault/num_samples: 1000
 * /move_group/planner_configs/FMTkConfigDefault/radius_multiplier: 1.1
 * /move_group/planner_configs ...