ignacio's profile - activity

Try using the robot_state_publisher tutorial. You will be able to load the URDF so ROS can use and gazebo

You can convert to sensor_msgs::PointCloud and work with the XYZ coordinates that you wanted or to pcl::PointCloud. But

what happens if you use ROS_INFO_STREAM("[IntensityCorrection] got riegl pointcloud with frame id: " << msg->po

Thanks for providing the branches with the fixes! I just managed to have it running and I had to alternate between updat

I had a similar issue when using catkin_tools, which was solved by updating the python3-catkin-tools package

can you try apt install python3-rospkg? The ros2 distributions rely on python3.

if you don't have pip installed, you can also use apt install python-rospkg

In my case, the nodes are created with private namespaces and with different naming, so the topics should be different

Testing node throws 'ros::serialization::StreamOverrunException' BufferOverrun I am trying to have a pcl::PointCloud<

I will add more info to the description

Is there any reason for that to happen exclusively with pcl::PointCloud<pcl::PointXYZ>? Because the test works wit

I am running that in a test scenario. For that I am processing 5 points only and using ros::AsyncSpinner spinner(0);. I

I tried adding the NodeHandle as a class attribute. It still doesn't work, unfortunately

nowhere. After the SetUp() I use the fixture's subscriber and publisher to check the information from my node

Testing node throws 'ros::serialization::StreamOverrunException' BufferOverrun I am trying to have a pcl::PointCloud<

This question came up for not knowing when I should run spin using node or node_base_interface. Perhaps answering when to use node_base_interface might help too.

In C++, it seems that I may use spin with:

RCLCPP_LIFECYCLE_PUBLIC
rclcpp::node_interfaces::NodeBaseInterface::SharedPtr
get_node_base_interface();

as in

rclcpp::executors::SingleThreadedExecutor exe;
std::shared_ptr<rclcpp_lifecycle::LifecycleNode> node = std::make_shared<SomeClass>();
exe.spin_node_some(node->get_node_base_interface());

or I could also use

std::shared_ptr<rclcpp::Node> node = std::make_shared<SomeClass>();
exe.spin_node_some(node);

Now it also works with sensor_msgs::PointCloud2 using bool concatenatePointCloud (const sensor_msgs::PointCloud2 &cl

I ended up in this thread looking for a solution for this problem. But my issue was that I assigned a time to stamp pc.h

"launch it once the rest of gazebo is up and running for a little bit just to be sure it has time to launch controller_m

Thanks for the fast reply. It does answer my question and I will close this one

Should I remap a topic or change as a paremeter I am confused about which approach is better. When I want to either subs

In my case, the issue was caused by calling the shutdown() method from the publisher

this is likely to be a CMakeLists.txt problem, where you could have forgotten to include the required directories

one example could also be args="pub /my_controller/command std_msgs/Float64 50". In this case I am publishing an std_msg

one example could also be args="pub /my_controller/command std_msgs/Float64 50". In this case I am publishing an std_msg

one example could also be args="pub /my_controller/command std_msgs/Float64 50". In this case I am publishing an std_msg

is that a typo or are you missing the </robotNamespace>?

Hi all, I have a car simulation in ROS Melodic with 4 wheels and a joint on top of it and I am getting behavior that I don't know how to solve even after searching that in the forum.

The wheels are actuated according to the following plugin

<!-- Add Skid Steer Plugin -->
    <gazebo>
        <plugin name="skid_steer_drive_controller" filename="libgazebo_ros_skid_steer_drive.so">
            <updateRate>100.0</updateRate>
            <robotNamespace>/my_car</robotNamespace>
            <commandTopic>cmd_vel</commandTopic>
            <odometryTopic>odom</odometryTopic>
            <odometryFrame>odom</odometryFrame>
            <covariance_x>0.0001</covariance_x>
            <covariance_y>0.0001</covariance_y>
            <covariance_yaw>0.0001</covariance_yaw>
            <leftFrontJoint>base_wheel_front_left</leftFrontJoint>
            <rightFrontJoint>base_wheel_front_right</rightFrontJoint>
            <leftRearJoint>base_wheel_rear_left</leftRearJoint>
            <rightRearJoint>base_wheel_rear_right</rightRearJoint>
            <wheelSeparation>2.11</wheelSeparation>
            <wheelDiameter>0.71</wheelDiameter>
            <robotBaseFrame>base_footprint</robotBaseFrame>
            <torque>1000000</torque>
            <topicName>cmd_vel</topicName>
            <broadcastTF>false</broadcastTF>
        </plugin>
    </gazebo>

While the joints on top of the car are actuated using the transmission element like this one:

        <transmission name="chassi_boom_transmission">
            <type>transmission_interface/SimpleTransmission</type>
            <joint name="base_mount_joint">
                <hardwareInterface>hardware_interface/EffortJointInterface</hardwareInterface>
            </joint>
            <actuator name="chassi_boom_actuator">
                <hardwareInterface>hardware_interface/EffortJointInterface</hardwareInterface>
                <mechanicalReduction>1</mechanicalReduction>
            </actuator>
        </transmission>

The problem is... Whenever I spawn the joint_state_publisher, both gazebo and joint_state_publisher publish to the topic joint_states. However, the gazebo never updates its joint_state causing my simulation model to flicker.

$ rostopic info /my_car/joint_states 

Type: sensor_msgs/JointState

Publishers: 
 * /gazebo (http://ps-imr:42743/)
 * /joint_state_publisher (http://ps-imr:38597/)

Subscribers: 
 * /robot_state_publisher (http://ps-imr:38075/)

I currently have one launch file to start the simulation and another one to start the ros_control layer.

part of My_car.launch:

<node name="joint_state_publisher" pkg="joint_state_publisher" type="joint_state_publisher">
        <remap from="/joint_states" to="/my_car/joint_states" />
    </node>
        <!-- Spawn the example robot -->
    <arg name="my_car_model" default="tesla" />
    <param name="robot_description" command="$(find xacro)/xacro --inorder '$(find my_car_description)/urdf/tesla_my_car.urdf.xacro' my_car_model:=$(arg my_car_model)" />
    <node pkg="gazebo_ros" type="spawn_model" name="spawn_model" args="-urdf -param /robot_description -model my_car" />

    <!-- Publish TFs -->
    <node pkg="robot_state_publisher" type="robot_state_publisher" name="robot_state_publisher" output="screen">
        <param name="publish_frequency" type="double" value="50.0" />
        <remap from="/joint_states" to="/my_car/joint_states" />
    </node>

And that is part of my control.launch

<!-- Load joint controller configurations from YAML file to parameter server -->
<rosparam file="$(find belt_loader_control)/config/belt_loader_control.yaml" command="load" />

<!-- load the controllers -->
<node name="controller_spawner" pkg="controller_manager" type="spawner" respawn="false" output="screen" ns="/my_car" args="my_car_controller flex_ramp_controller joint_state_controller" />

Any thoughts about why that could be happening?

I just realized I didn't add the transmission to the wheels <transmission name="wheel_transmission_${position}">

I mean ... they are not event shown in my Robot Model. It states No transform from [wheel_front_left] to [base_footprint

because otherwise I won't see the wheels in place in RViz. Even though I am still able to move it by publishing a messag

Multiple publishers for joint_state cause RViz to flicker Hi all, I have a car simulation in ROS Melodic with 4 wheels a

Multiple publishers for joint_state cause RViz to flicker Hi all, I have a car simulation in ROS Melodic with 4 wheels a

I had a similar issue and adding rospy.Duration(4) was essential

remember to run sudo apt update, just in case

just updated. Thanks guys!

You may need to add two CallbackGroups and pass then as arguments while calling create_subscription() or create_publishe