Blog 4: Week of 6/29/18
Andrew Brown here again. Last week left off with high hopes as we planned to try a couple of different computers to mitigate the latency which we believe is the root of the drone’s failure to hover in place while under computer control. We got my computer to work easily enough, but we had the exact same problem as with the other computer. The next step was loading Ubuntu onto the old “Hover Control 4” computer. The lab used to use this computer with the CubeSat but has not done so for some time. The computer has an atom processor. According to my understanding this processor came before the i5. It is really really slow. The hope, however was that the operating system on to the “Hover Control 4” computer would be able to send and receive information through the usb ports faster than my computer because the operating system would be native and not on a virtual machine.
However, in order to use the slow computer to fly the drone we first had to load it with ROS indigo. In order to download ROS indigo, we had to download ubuntu 14.04. Each of these steps took a really long time because the computer is so slow.
Along with using “Hover Control 4” computer, we also decided to use the Vicon system in the smaller room. We did this for two reasons. The first is that the Vicon cameras are closer together in the small room, giving this system higher resolution that the one in the big room. The second is that Siya often has participants in the large room. These meetings take a lot of time, so we will be able to work with less interference if we use the Vicon system in the smaller room.
Unfortunately, hooking up to Vicon in the smaller room included many hurdles. We tried to hook up first via wifi. The wifi was too spotty so we decided to hook up via ethernet. I was not sure how to go about this. I tried for a while by myself and then got AJ to help me. He worked on it for a little bit but did not have enough time to figure it out. I worked on it some more and then got AJ to help me again. He got the slow computer to communicate with the virtual machine on the Vicon computer, but then he noticed that while I was trying to get the ethernet connection to work, I broke the Vicon connection. It took him about an hour to reconnect Vicon. He was very gracious and did not get very upset with me for causing him so much extra work. I am truly grateful to him. AJ also explained to me what he did to fix Vicon and showed me how to make the ethernet connection work. I learned a lot.
On Friday we finally got the drone to fly in the small room. However, it still would not hover. This is probably because the slow computer does not have the proper graphics drivers. I am not sure its hardware can support the nessicary drivers, so we may have to get a new computer.
Socially things are good. We, the undergrads, still get along well, and the grad students are still awesome. Also, we won this week’s intermural volleyball game. We are not 2 and “o”!
Overall it was a good week. I really hope we get a drone to hover next week.