Small Tabletop Telepresence Robot


When [Peter] saw the Sparkfun Magician robot chassis in a recent new product post, he knew instantly that he had to have one for a telepresence project that had been kicking around in his head for a while.

Onto the robot chassis, he added an Arduino to provide the brains of the bot, an Adafruit motor shield for controlling the wheels, and a Pololu Wixel for wireless communications. An iPhone is mounted on the top of the robot, which communicates with his laptop using Apple’s Facetime app. The robot is controlled from his laptop as well using the Wixel, which enables him to direct the Magician chassis as if it was attached via USB.

While he thinks the robot is pretty neat and that it works well, [Peter] already has improvements in mind. The robot chassis is a bit weak on anything but smooth surfaces, so a new set of motors and wheels are likely the first changes he’ll make. He wants to add a servo-based aiming mechanism for the phone’s camera, as well as some sensors to prevent the bot from taking a nosedive off his table.

iPhone aside, this is probably one of the cheaper mobile telepresence setups we’ve seen, so we can’t wait to hear how the improvements work out, and how much they add to the robot’s cost.

12 thoughts on “Small Tabletop Telepresence Robot

    1. Replace iPhone with the cheapest Android that is still powerful enough to do two-way video.

      Have the phone control the wheels directly (either through a cheap microcontroller or simply using two hacked servos as motors and controlling directly from the audio out, though that would require the VoIP to be routed elsewhere).

      If you’re wanting to do this for the tightest budget possible, you could literally just tape the two hacked micro servos to the bottom of an extremely cheap handset and run them off the phone’s battery. Total parts list would be the handset, two micro servos, two wheels, and some sort of adhesive (say, tape).

      1. You could go even cheaper than an iPod Touch. With a DTMF decoder chip, you don’t even need a smartphone – any old 3G video phone would work (and you could use a dumb videophone as the client at the other end, too).

  1. Seems a cheap WiFi camera would make a whole lot more sense than strapping a smartphone onto the front of this; especially since the smartphone isn’t even helping with the processing. Could even get one with pan & tilt.

    Otherwise, I would look into replacing the Arduino with the phone itself. At least it would justify it’s presence then.

  2. Nice build, gives a use for that upgraded from phone.

    As for EoT sensing- mechanical probes with enough outreach for lag time seem prudent.

    There used to be some “rotary drive wheel” toys fitted with a Sled Probe that lifted the motor from it’s friction drive and engaged a rear drive wheel motor to pull back from edges. A mechanical wheel brake lever thru a slip clutch= back off only, as failure mitigation might be a Very Good Idea..

  3. It seems a lot of you are missing a vital point: chances are this can be solved with a software update. If you don’t like it its your own fault for being early adopters. I’m waiting for all these problems to be ironed out before I get mine, which probably won’t take too long.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.