brawner's profile - activity

For a fixed joint between two components, you need to fully define the joint type and the joint coordinate system throug

For a fixed joint between two components, you need to fully define the joint type and the joint coordinate system throug

Unfortunately, the URDF specification itself does not allow for kinematic structures with cycles, like a parallelagram.

Thanks for posting about the issue. If it is something that has cropped up in the 2021 version, then it's quite possibly

Thanks for posting about the issue. If it is something that has cropped up in the 2021 version, then it's quite possibly

It's unfortunate that this bug occurs with the precompiled installer. It's most likely that the compiled version of the

Which version are you using? For SW 2019, I believe 1.5.1 works better than 1.6

When you first click "Export as URDF", it should open a plugin manager window to the left-hand side to begin initial set

Which plugin version are you using? 1.6 should probably only be used with 2020, and 1.5.1 with 2019.

There is always a possibility you are experiencing a different issue or a better answer can be provided, but this questi

This sometimes happens if you do not have full administrator access to the computer. Is this your own machine, or is it

If the RPY values are incorrect and the axis is incorrect, I would first verify the coordinate system transforms are as

If you update your model in the dropbox folder with the single-assembly-per-link version, I'll take a look at it this we

Sorry, to clarify, the coordinate system in the mass properties dialog should be the coordinate system of the joint. Not

Thank you for moving from the github issues over here, and for your very detailed question. It helps out everyone who ha

Sorry I missed responding to this earlier, but I think the primary issue is using SolidWorks 2018. There is a bug in ser

I've downloaded the model you gave me, exported it with version 1.5.1 of the exporter in SolidWorks 2019 and have found

What service pack of SolidWorks are you using? SW 2018 SP 1-4 has known issues with mesh exports. You need SP 5. It's di

This is a known issue experienced by a few people. I'm not entirely sure of why this is caused. What version of Windows

Unfortunately this is a known bug and it is being tracked in several places. I would recommend you open rqt_plot as a st

Does this link not work? I can download it without issue. https://bitbucket.org/brawner/sw2urdf/raw/2f613f8f2a9e9de4983

Which version of SolidWorks are you using? SolidWorks 2018 has had a bug for most of the year, and they just issued a ne

It's a plugin to SolidWorks, which currently only runs on Windows, so there are no plans to get it to run on any other s

You don't need to compile the source code, if you don't intend to modify the code itself. Since the installer worked jus

Unfortunately, the automatic detection isn't working correctly right now under the latest versions of SolidWorks. Something that I mean to get to. In the meantime, you'll need to insert the reference geometry manually and configure your links to reference the manually inserted reference geometry.

I have been able to get it to work this way with SW 2015

Not all values are saved, yet. The configuration file saves coordinate reference information. In essence, the information that the exporter needs in order to extract proper info from SW each time.

I believe limit mates are respected in the latest version of the exporter. If you want to create an easy export, add limit mates to each joint that represent the limits of the joint itself. I think the soft limits are set to the same thing as the hard limits in the urdf. Create an configuration solely for export with these limit mates unsuppressed if you don't want to deal with them in the future.

Do you mean to say that they are not exporting the limits, or they are not saving them to the configuration inside of SW?

I want to find partial overlaps between meshes in the household object database and pointcloud sensor data. My method is to convert the arm_navigation_msgs::Shape to a pcl::PointCloud and then find clusters that overlap through pcl::IterativeClosestPoint. I have all pieces working, I just think my function for converting the shape meshes to pointclouds is inefficient. Is there a way to take advantage of the underlying structure to speed up the conversion?

This is my current implementation

pcl::PointCloud<pcl::PointXYZ> PCLMeshOverlap::convertMeshToPointcloud(arm_navigation_msgs::Shape mesh)
{
    ROS_INFO("Converting mesh to pointcloud");
    pcl::PointCloud<pcl::PointXYZ> cloud_out;
    pcl::PointXYZ point;
    for (int i = 0; i < mesh.vertices.size(); i++)
    {
        float isNew = true;
        for (int j = 0; j < cloud_out.points.size(); j++)
        {
            if (    mesh.vertices.at(i).x == cloud_out.points.at(j).x &&
                    mesh.vertices.at(i).y == cloud_out.points.at(j).y &&
                    mesh.vertices.at(i).z == cloud_out.points.at(j).z   )
            {
                isNew = false;
                break;
            }
        }
        if (isNew)
        {
            point.x = mesh.vertices.at(i).x;
            point.y = mesh.vertices.at(i).y;
            point.z = mesh.vertices.at(i).z;
            cloud_out.points.push_back(point);
        }
    }
    cloud_out.width = cloud_out.points.size();
    cloud_out.height = 1;
    cloud_out.is_dense = true;
    cloud_out.header.frame_id = "/base_link";
    return cloud_out;
}

*Edited because the original code had some typos