doughe@bamboo.WV.TEK.COM (Douglas E Helbling) (01/29/91)
I am working toward putting together a robotic competition for small rover-like vehicles. The contest would take place on the Oregon desert at the Oregon Moonbase, an earth analog of a lunar lavatube. The Oregon Moonbase is a project of The Oregon L-5 Society, Inc., a chapter of the National Space Society. The robotics competition, if it comes to reality, will probably take place in late 1992 or early 1993. If you are not interested in any of the above, now's the time for a kill. The basic goals of the contest: 1) Demonstrate the feasibility of usings small, relatively inexpensive robotic vehicles on productive off-planet missions. 2) Provide a forum for academic and industrial robotic enthusiasts to meet and exchange ideas and information. 3) Further demonstrate the feasibility of lavatubes as useful for the eventual proliferation of humans in lunar and martian settlement. The basic goals of the robots in the contest: 1) Deploy (from contained) a robot in the proximity of a lavatube. 2) Send the robot out to locate a lavatube (or a specific lavatube among several at the site). 3) Have the robot navigate the terrain between the surface and the interior of the tube; in other words, have it get safely inside. 4) Have the robot reach a particular location within one or more of the lavatubes, said location to be marked by concentrated metal (tin foil) or mineral deposit (powdered sugar) markings detectable with some form of simple sensor capability. The framework for the contest currently allows for four categories: Category One: Autonomous (non-telemetry) Operation, Custom Hardware Essentially, this is for self-contained rovers of unique physical construction. Category Two: Autonomous Operation, Off-the-Shelf Hardware This is for self-contained rovers of commercially available design (hobbyist or pro), with custom programming to meet the objectives of the contest and the site. Category Three: Remote Operated Operation Custom Hardware This is for remote-operated rovers of custom design. Category Four: Remote Operated Operation, Off-the-shelf Hardware This is for remote-operated rovers of commercially available design. What I would like to get feedback on is: 1) What is the existing level of interest? Is there enough interest out there to warrant this kind of competition? If so, is it likely to come from industry, academia, or where? 2) What sort of incentives would be required to get you to participate? I'm still working on acquiring funding for some type of prize. A fabulous cash award (likely to be between $1K and $5K) or generic scholarship funds (same general dollar figures) are likely to be the best that this not-for-profit organization can muster without a very generous individual or corporate sponsor in private industry. 3) What are the technical constraints? Here's a starting list, gleaned in part from knowledge of the sight, my own goals for lunar analog demonstrations of robotic systems, and from chatter on this newsgroup: a) Propulsion - since this is a lunar analog, all forms of air-consuming propulsion (i.e. internal combustion) are out. Because humans will be in proximity, and because we wish to avoid federal supervision and control, nuclear systems are out. b) Size - the current limit on physical size for the "unextended unit" is two feet square. (Must fit in a box two by two by two feet.) The notion that a robot could "expand" out of its shipping crate after delivery (particularly for units basing their design on insect models from the natural world) is acceptable. c) Attendance/Remote operation limitations - since some participants in a competition such as this might well spend their entire funds on the project itself, they may not be able to attend personally. In the case of robots that are fully self-operating, this should not be an issue. So long as the unit can be started with minimal instructions (the limitations of which are yet to be determined), it should be safe to expect that you could mail (or UPS, etc.) your entry to the contest site, and have it entered for you by volunteers at the site. If your entry is indeed remotely operated, your presense at the site during the contest may be required, unless the unit can be teleoperated through a cellular phone/modem connection or via HAM radio. The latter alternative may require that some properly licensed volunteer be on the sight. As with most similar competitions, the contestants would retain all rights to their designs and entries. The sponsors will probably request rights to any video or audio records of the contest as conditions for entry. If you have any feedback on this idea, critical or otherwise, I'd like to hear from you, here on the net (comp.robotics), via email (doughe@bamboo.cax.tek.com), or by paper mail to: Robotics Contest c/o The Oregon L-5 Society, Inc. P.O. Box 86 Oregon City, OR 97045 Thanks in advance. Doug Helbling doughe@bamboo.cax.tek.com Disclaimer: Tektronix, Inc., is not a sponsor of this proposed contest, of Oregon L-5, or Oregon Moonbase. Opinions, solicitations, etc., are the work of this individual. Doug Helbling (503) 591-1696
smith@sndpit.enet.dec.com (Willie Smith) (01/29/91)
In article <9972@orca.wv.tek.com>, doughe@bamboo.WV.TEK.COM (Douglas E Helbling) writes... > > I am working toward putting together a robotic competition for small > rover-like vehicles. The contest would take place on the Oregon > desert at the Oregon Moonbase, an earth analog of a lunar lavatube. > The Oregon Moonbase is a project of The Oregon L-5 Society, Inc., a > chapter of the National Space Society. The robotics competition, > if it comes to reality, will probably take place in late 1992 or > early 1993. The Sunswick Engineering team will be there with the Tycho and Waldo vehicles if this happens. Just what we've been looking for! > b) Size - the current limit on physical size for the "unextended unit" > is two feet square. (Must fit in a box two by two by two feet.) > The notion that a robot could "expand" out of its shipping crate > after delivery (particularly for units basing their design on > insect models from the natural world) is acceptable. Aw, c'mon, make it a three-foot cube, so we can bring Waldo (the real thing) , instead of just Tycho (the toy). A few more things I'd like to see thrown in for good measure: 1) All teleoperated vehicles should properly emulate the 3-second speed-of-light communications delay. 2) Make the contest a little more useful, and ask that some useful work be performed. F'rinstance, clearing a landing field, digging habitat foundations, covering habitat modules, etc. Sounds like a blast, we'll be there! Willie Smith smith@sndpit.enet.dec.com smith%sndpit.enet.dec.com@decwrl.dec.com {Usenet!Backbone}!decwrl!sndpit.enet.dec.com!smith
bleck@mobot.ai.mit.edu (Olaf Bleck) (01/29/91)
I like it! My spiel: In general, when asked about such contests, we try to encourage robotics contests to have open-ended goals, perhaps more like a talent show, because it allows additional creativity. I would say the MIT Mobile Robotics Lab could be expected to show up at such an event...our current loosely defined long range goals are to have some sort of space qualified version of "Attila", our six-legger, in the works by then, along with some idea of a mission date. "Attila's" will also be on the market, certainly be then, for about $50k, as an AI research tool. Anyone interested in more details about that can contact me directly. On a different note, is anyone out there participating in the AUVS contest in July (autonomous flying robots)? Cheers! -Olaf bleck@ai.mit.edu
bleck@mobot.ai.mit.edu (Olaf Bleck) (01/30/91)
>Could you post any info on the Flying Robotic Vehicle Contest >in JULY ? AUVS (Association for Unmanned Vehicle Systems) is hosting an Aerial Robotics Competition in July. Site, Georgia Tech. The task is as follows: The venue is tennis court size with a 3 ft. barrier across the center. There are two 6ft round pedestals, one on each side of the "fence", each with a xenon stobe in it. The task is two move six disks from one pedestal to the other without touching ground anywhere but on the pedestals, and obviously staying within the confines of the court. 3 min. time limit. The prize is up to at least 10k right now, and teams can be student, student/faculty, student/industrial, etc. Technically you were supposed to have registered by Dec. 31. Contact Robert Michelson, Chairman AUVS Technical Committee c/o Georgia Tech Aerospace Science & Technology Lab Atlanta, GA 30332 USA (404) 528-7568. RMICHELS%GTRI01@vm1.gatech.edu -Olaf
winans@sirius.mcs.anl.gov (John Winans) (02/06/91)
I have been reading this thread with some interest. And have wondered about the remote-controlled "categories". The origional note mentioned the use of cellular phones and ham radio to communicate with the robots. Since we all know that radio works and we COULD use it if we wanted to pay for it, why not allow (or create a category) that would allow a looooooooonnggg wire to be used instead? Sure it's sort of a wimpy way out, but I'm sure there are others like myself that would like to enter this thing, but don't feel like dealing with (and paying for) the radio stuff. (Those others might even have time to finish building their machine in time as well :-( ) This would also allow us poor folks to use video. And it would also eliminate any argument for letting people within direct sight of their vehicle when controlling it. Hmmm If there was SOME standardization in the format used, the sponsor of the event could also provide an error generator that could inject errors into the communication system (like interrupt video signals, pop bit errors into the standard serial communications formats etc...). This might even up the competition a little bit as well. -- ! John Winans Advanced Computing Research Facility ! ! winans@mcs.anl.gov Argonne National Laboratory, Illinois ! ! ! !"The large print giveth, and the small print taketh away"-- Tom Waits !
smith@sndpit.enet.dec.com (Willie Smith) (02/06/91)
In article <1991Feb5.202021.2886@mcs.anl.gov>, winans@sirius.mcs.anl.gov (John Winans) writes... > >I have been reading this thread with some interest. And have wondered about >the remote-controlled "categories". The origional note mentioned the use >of cellular phones and ham radio to communicate with the robots. Since we >all know that radio works and we COULD use it if we wanted to pay for it, >why not allow (or create a category) that would allow a looooooooonnggg wire >to be used instead? Not a bad idea! I figure that by the end of this year, my costs for radio gear ought to be about half of the total vehicle costs. I personally don't want to deal with tethers, but to each his own. >Hmmm If there was SOME standardization in the format used, the sponsor of >the event could also provide an error generator that could inject errors into >the communication system (like interrupt video signals, pop bit errors into >the standard serial communications formats etc...). > >This might even up the competition a little bit as well. Not really, as the AX.25 data that runs over ham radio packet links is error corrected - either you get all the data error-free or the link goes down and you get nothing. I suppose you could mess up the video, but that's going to be about the same effect, either it's usable and the Waldo team wipes the floor with you all :+) or it's not and no-one gets any work done.... Willie Smith smith@sndpit.enet.dec.com smith%sndpit.enet.dec.com@decwrl.dec.com {Usenet!Backbone}!decwrl!sndpit.enet.dec.com!smith
larson@snmp.sri.com (Alan Larson) (02/07/91)
If you are going to require small size for the lunar robot contest, and limit the power sources to lunar capable ones, allowing wire connections or local visual observations is a bit imbalanced. If you are talking about the 3 second delay from earth based teleoperation, why aren't you including the difficulty of radio communications back to earth. A quick check will discover that the video transmitter will suck a fair chunk of power over that range. It is probably more reasonable to simplify the requirements to only test the robotics (and possibly the AI) part, as the communications part could overwhelm the efforts of individuals or small groups. Alan
yamauchi@cs.rochester.edu (Brian Yamauchi) (02/07/91)
In article <20916@sri-unix.SRI.COM> larson@snmp.sri.com (Alan Larson) writes:
If you are going to require small size for the lunar robot contest,
and limit the power sources to lunar capable ones, allowing wire
connections or local visual observations is a bit imbalanced.
I agree -- if you really want to demonstrate the feasibility of
*small* lunar probes, you should prohibit either tethers or local
observations. All necessary processing and/or communications hardware
should be onboard.
It would be particularly unfair to allow teleoperated vehicles to use
tethers or local observations and not to allow autonomous robots to do
their processing on external hardware (connected via tethers).
On the other hand, if what you're interested in is demonstrating the
capabilities of AI algorithms and/or teleoperation for lunar
exploration, then you could remove all restrictions on the the
hardware (size, power, tethers, etc.) Though, I still think a 2-3
second delay should be implemented on the teleoperated systems and
local observations should be prohibited.
--
_______________________________________________________________________________
Brian Yamauchi University of Rochester
yamauchi@cs.rochester.edu Department of Computer Science
_______________________________________________________________________________smith@sndpit.enet.dec.com (Willie Smith) (02/07/91)
In article <20916@sri-unix.SRI.COM>, larson@snmp.sri.com (Alan Larson) writes... > > If you are going to require small size for the lunar robot contest, >and limit the power sources to lunar capable ones, allowing wire >connections or local visual observations is a bit imbalanced. >If you are talking about the 3 second delay from earth based teleoperation, >why aren't you including the difficulty of radio communications back >to earth. A quick check will discover that the video transmitter will >suck a fair chunk of power over that range. > > It is probably more reasonable to simplify the requirements to only >test the robotics (and possibly the AI) part, as the communications >part could overwhelm the efforts of individuals or small groups. > > Alan Oh, I don't think it's that difficult... First off, each individual vehicle wouldn't have to have the power and antenna (and antenna pointing) capabilities to reach the earth. They would only have to transmit as far as the nearest base, which could be as simple as the ship they landed in. Thus, each vehicle has to have a couple of watts each for the telemetry link and any number of TV channels. Other repeaters could be scattered around the site of a lunar colony or teleoperations area, to cover areas in radio shadow or extend range. these could even be small enough to be carried and emplaced by the teleoperated vehicles themselves. In fact, with the addition of recievers, each vehicle could act as a repeater for others that got lost. The only high-power links are those at the main base, or built into the lander. Willie Smith smith@sndpit.enet.dec.com smith%sndpit.enet.dec.com@decwrl.dec.com {Usenet!Backbone}!decwrl!sndpit.enet.dec.com!smith
winans@sirius.mcs.anl.gov (John Winans) (02/08/91)
Hmmm... I think my point was not very clear (about the comm-link via wire instead of radio.) First of all it was intended to replace the RF transmission link ONLY. Not to supply power or the ability to remove the need for on-board computing devices. The 3 second delay should be required on both the video and the data lines. And can be implemented the SAME way that it will have to be done for the RF links. The general purpose for this sugestion was to allow student groups studying areas like robotics for manufacturing and such (where RF is might not be used and therfore not learned by the individual) participate in an event like this. As far as radio *simulation* goes, (as I stated earlier) the bit stream(s) can be messed with as needed to generate *bad packets* and require retransmission or what ever. And video the same way. I wasn't thinking of a "toggle switch" where the communications would just be removed for X ammount of time. I suppose it would also be fair to not generate RF errors and limit the bit rate to something that is low enough to represent the average thruput of the RF link. I wasn't thinking when I forgot about the change in the power requirements this would cause (in the favor of the machines w/o radios.) Oops. But at the same time, haven't you?? I would think that a link to another planet would require more power than a link 1000 feet! And besides, it would probably be more dificult to get such a link operating in a place like Chicago than it would be in Oregon (where this contest is to occur.) Ever use cellular out here? There are also a fair ammount of folks Hamming. (I suppose I am nit-picking now, so I'll stop.) And finally I sugested this as a category that COULD be on its own (the origional posting had a selection of categories) and as such it would not compete directly with the others. My comment that this would "even up the competition" was intended for the competition as a whole. As in everyone should be subjected to the same sort of communications errors and delay problems. -- ! John Winans Advanced Computing Research Facility ! ! winans@mcs.anl.gov Argonne National Laboratory, Illinois ! ! ! !"The large print giveth, and the small print taketh away"-- Tom Waits !
paj@mrcu (Paul Johnson) (02/08/91)
In article <778@sousa.enet.dec.com> smith@sndpit.enet.dec.com (Willie Smith) writes: >>Hmmm If there was SOME standardization in the format used, the sponsor of >>the event could also provide an error generator that could inject errors into >>the communication system [...] >> >>This might even up the competition a little bit as well. > >Not really, as the AX.25 data that runs over ham radio packet links is >error corrected [...] What about a bandwidth limit instead? I should think comms to the moon have two problems: latency (mentioned in the orignal post) and bandwidth due to power limitations at the remote end. Anyone got any figures on watts/kHz of bandwidth for a lunar-terra link? Just my $0.02 worth. Paul. -- Paul Johnson UUCP: <world>!mcvax!ukc!gec-mrc!paj --------------------------------!-------------------------|------------------- GEC-Marconi Research is not | Telex: 995016 GECRES G | Tel: +44 245 73331 responsible for my opinions. | Inet: paj@uk.co.gec-mrc | Fax: +44 245 75244