Meeting Notes, ROS Industrial, Gyro Mounted

I forgot to mention in my last post that I have been doing some reading on ROS Industrial, which Wyatt pointed me to last semester. Basically, the Southwest Research Institute is trying to create a standard ROS interface for industrial arms. They’ve just put out their first prerelease code, and it looks like they’ve done some good stuff. They plan to release an interface for an ABB arm, but by the looks of it, all they have in the ABB section are stubs. Their documentation is nearly non-existent, but I am going to read through their code and decipher what they’re doing. I think it would be a good idea for me to use their structure (messages, nodes, etc) for my arm control. This would not only save me the trouble of recreating everything they’ve done, it would also make my work part of a larger, more significant project.

I’ve also been doing some reading on the Robot Reference Interface (RRI), which is ABB’s fastest protocol for TCP/UDP communication with the IRC5 controller. This is how I’m going to send commands to the arm. By the looks of it, it’s a pretty basic interface that allows the IRC5 to act as a server that listens for and publishes data. The data is then transparently written to and read from RAPID variables. The messages and all of the configuration files are written in XML, so it should be pretty easy to implement the ROS side and integrate it with ROS Industrial. Russell may also have some code written for this that I could leverage, but I hear he hasn’t been working in ROS, so I don’t know how much of his work is relevant.

My DigiKey order came yesterday. The order contained some 2MM headers for the XBEE modules (although it turns out I accidentally bought right angle headers. duh) and an enclosure for the gyro board. I spent part of this afternoon modifying the gyro board to fit in the enclosure and mounting it. I decided to get the enclosure so that I could mount the gyro hanging from the bottom plate, which would have required me to mount it upside down. With the enclosure, I just mounted the gyro board to the lid. This also protects the gyro board from contaminants and damage. Hopefully, the lab has some 2MM headers in stock. If not, I’ve got another DigiKey order to make.

The new inverter also came today. This inverter is a 2kW (3kW surge) 24vDC->220vAC/50Hz pure sine wave inverter. I was able to test it, and I can move the arm with it. It doesn’t seem to have any difficulty powering the controller and arm. It’s significantly larger than the previous inverter, so I need to rearrange the bottom plate a bit to make it fit, but it will fit. This inverter also appears to be in factory condition (unlike the previous one, which was obviously modified for 24v operation) and came with a small instruction manual. Overall, this seems like a much better product. I haven’t confirmed whether it actually puts out a pure sine wave, but I did quickly check the voltage with a multimeter. It was reading around 226vAC, which is within spec for the IRC5. The output connectors on this inverter are NEMA 5-15 plugs (which are standard US plugs for 110v outlets). I hacked together a cable for testing using an old IEC cable, but I think I’m going to order a proper plug for the final cable.

I brought my drill home last week, so I can’t mount the inverter (and remount some other things on the bottom plate to make room) until next week. In the mean time, I’m going to try to get Gazebo up and running on my laptop and finish reading up on RRI and ROS Industrial.

Meeting Notes

• Wyatt is out of town, giving a talk in Hong Kong
• Greg wants a demonstration/information about using SimMechanics link to turn CAD files into SimMechanics XML files. He'll probably have to do this in Solidworks. I'll figure it out and show him.
• Greg would like me to break out the wheels and castors in Abby's CAD file so they're articulable in the URDF model. I'm going to do this some time next week.
• We discussed ROS Industrial. He agrees with me that I should see if I can take advantage of it.
• There are some paper deadlines approaching.
  • ICRA is in Germany next year, and the theme is “Anthropomatics - Technologies for Humans”, highlighting the paramount role of Robotics and Automation in improving human quality of life. The submission deadline is September 17. I may actually try to get something together for this. There's a lot of things around here that could make good papers.
  • There is a security robotics conference with an approaching deadline. It's in the US, but my work probably isn't relevant.
• We discussed RRI briefly, and Greg mentioned Russell's work.