Sunday, June 10, 2012

Upcoming robot changes

I feel really bad that I have neglected this blog so much.  Months ago I had intended to integrate a pair of VR glasses (head mounted displays) and a head tracking device to control the robot.  The robot head would have acted as an extension of your physical body allowing you to look around the room as if you were really there. But along the way I got drawn into a number of other VR projects instead.

Free motion VR game tracking
Virtual Desktop

I have managed to do some maintenance and experimentation with the robot - mostly in the form of wireless networking.  My office layout is such that it is almost exclusively walled offices instead of cubicles so it is impossible for me to rove from end to end on a single router.  I tried many different configurations of routers and antennas both in the office and on the robot and have never found a workable solution.  Instead I just put several access points and remotely signal the robot to switch access points when I move from area to area.  Annoying? Yes, but it's the only way I can maintain a good connection throughout the office.

Also, I still have a lot of problems with the robot "browning" out and the computer dying during the docking sequence.  I finally decided just to redesign the head and put the battery back in the computer.  I don't want the additional weight but its the only way I can stop the robot from losing power.  I'm working on that change in the next two weeks and will report on it.


  1. Hey I'd like to talk to you about your tech. I have been building something similar but with a different twist. I am actually making hardware from scratch. Contact me at

  2. Hi, excited by what you have been doing. I too was inspired by Johnny Lee a couple of months ago and wanted to take it to the next step, which I now find you have already done. Great work on the power and charging system. Love the neck joint. Any update on these hinted further power system changes since Part 9 (beyond adding the netbook battery back)?

  3. I tried adding the netbook battery but unfortunately it didn't help much of anything. My iRobot would still crash almost daily as it was entering the charging unit, leaving my netbook alive, but without a charging source so it would eventually just dwindle down and shut off. Eventually I just decided the base unit was flaky. So I removed the netbook battery again and put it back together like it was but changed the base iRobot unit out. That helped a lot and it's been almost rock solid for several months now.

    Those iRobot's are very finicky. I've been through 3 of them and the charging circuitry is different on each one. If you happen to get a bad one, it will crash on you all the time. If you get lucky it will stay rock solid ... until the charging circuit finally goes out on you after 6 months.

  4. Hopefully it just means they are upgrading it along with their consumer vacuum models and not that they have spastic supply channels. At least they sell the units separately from the batteries so it is not too expensive to replace.

    Good to hear that you got it stable, even if by luck. I had been wondering if a solution to the problems would have been to take the separation of the laptop/other components power to the next step by using something like the Energizer® XP8000A as an intermediary on the charging. It has pass through charging ability and it is supposedly smart enough to charge over three different voltages simultaneously. You could use the usb for the 5v usb, the 16v to 20v for the Create and if you had a netbook with only 12v requirements power it off that line. I see that some slightly older models like the Asus eee pc 900ha only require 12 volts but you would have to live without the wireless/n. I suppose you could also step it up to the higher voltage. I assumed something similar to Johnny's mod of the base station but with DC (both bricks) so that the base-station had its internal power needs met for the docking IR transmitter while passing through the Energizers needs. Or step down the energizers power to meet the base stations needs with the one brick. Then finally a change to the internal Create wiring so the charging contacts lead directly to the energizer input and then back out to the charging circuitry.

    1. I considered a few alternate power options like this, but they all require stacking additional weight on the base unit so I avoided them unless I absolutely needed them. Even Johnny's AC solution started to look better to me at one point. But now that the unit is stable, I stopped tinkering with it.

      I ran across a forum post once where somebody had managed to manually replace the charging circuit on the iRobot with a stronger version. I have a couple of "dead" units lying around now that I may try to upgrade in that way at some point.

  5. Have you thought about re-adding the battery to the netbook, and instead taping into the charging of the iRobot to charge the netbook battery in addition to the iRobot battery. It will take longer to fully charge both and I suspect the amp draw might be an issue that would need to be addressed, but it would create two separate power-sources that could in turn back each other up (or not). Also - I like the idea about detaching the screen from the netbook and mounting separately, but as you stated that could be hard to do.