Official OpenROV Dev Call: Feb 2 2015


#1

Attendies:
Ben M.
@Eric_Stackpole
Dominik
Lawrence T

Video Conferencing Setup

Dogfooding. We are working on some tele-robotics solutions that extend existing tele conferencing solutions. We are hopeful that future calls can run on the network we are setting up for tele-robotics. Today we are using bluejeans.com which uses some of the same technology.

Potential Conerns to watch out for

  • Don’t let it become a distraction for the person controlling the ROV

Ben: Shared the use of AUVs in search and rescue. Need systems that allow video to be distributed to lots of people, but want that to be independent of the operator who is focused on just driving. Don’t want the operator to be pulled in to trying to micro-analyze the video while driving. Might be a good use case if you have a driver, narrator, and lots of watchers.

Eric: Referenced out youtube video from the search for the SS Tahoe that has a flow similar to when Ben mentioned:

Ben: Feature Request Allow someone to click on the screen at some place and tag it with “Something interesting here” (CH: real-time annotations).

CH: Discuss how this is a similar concept to Microsoft’s marketing video on their Holo glasses, where someone is remotely annotating where on the pipe for someone to cut when doing a plumbing task.

Erik: One of the problems is how to quickly edit hours of raw video in to a highlight real people want to watch.

Ben: Interest indicator: I have 16 people watching online, I have an area with lots of spot lights on the screen that shows where people’s focus is to inform the pilot on where to move the ROV.

Lawrence: Recounts his experience doing scientific dives using video for transects.

Dom: Recounts a discussion with the navigator from last years Nautilus project. They used 2 ROVs working in tandem.

Eric: Would be nice to talk to the people who created and design NASA mission control. We should use it as a solid starting point.

CH: NASA as a starting point yes, but our use case is probably way more casual. The strictly enforced roles NASA uses may not be a great fit. Will need some trial and error to dial in. Needs to be flexible enough to handle just looking around with your friends, educational use, expensive scientific missions.

Dom: Be sure to talk to @erika_bergman about her experience.

Call for project updates

Dom: 2.7 is coming along. Hopefully get it in the water next week. Going to get a 2.6 ROV from the Sydney Harbor Lab as well. Discussed some of the experience of building 2.7 vs prior versions. Working on a new UI as envisioned by Eric. Taking some time as we are turning everything in to HTML5 web components. Will allow for easily popping key UI widgets off of the driver screen on to a second screen.

Lawrence: Shared some of the things that interest him about the ROV. Discussed counting small sea creatures the size of a thumbnail and having to determine if they were alive of dead while doing so. There is hope that computer vision might help for these quantitative type of studies. Eric shared some of his stories as well. Lawrence mentioned the plankton species identification system. Eric shared thoughts on fluidic well plates with an embedded usb microscope to do mass collection of specimens. Lawrence: Knowing depth and distance is important as well.

Eric: Call for action. There is only so much we can do directly at OpenROV, we want to get the community excited and working on projects that benefit everyone.

Lawrence: Wonder if the community would be interested in converting old underwater camera and video housings with those nice lenses built in in to ROVs…


#2

I listened to the Dev call and have an idea with the observation mode. If you Place 360 degree cameras above and below the Rov and have the moderator software interact with those cameras for highlighting what observers are interested in. This way the pilot of the Rov isn’t distracted by the observers. You would only have the moderator(s) able to directly interact with the Pilot or group running the Rov. Just a thought.