Our Autonomous Vehicle Competition is less than a month away now (June 16, 2012 at SparkFun Headquarters). It's shaping up to be an awesome day! We've created an updated course preview so you'll know what to expect on the day of the race. Check it out!
If you have any more questions, leave them below and we'll try to get to them. Also, if you are planning on attending as a spectator, please register at our EventBrite page. Hope to see you there!
Team Viator here...
+1 for moving the barrels throughout the day. I'd like to further suggest not allowing practice runs with the barrels in their new positions so the vehicles truly have to navigate through the obstacles.
I assume from the video walkthrough that there is no longer a requirement for a 3 second delay at the start? The rules still state: "You press the start button, three seconds later the robot begins navigation under its own control.". Can we get clarification on this?
We're very excited about this event!
Didn't see that. I'll fix that. No, there is no need for a delay.
The online rules posted at http://www.sparkfun.com/products/11069 still say that a three second delay is required.
Wayne
I know, I'm trying to get it changed, but long story short, it's locked out for me.
Is there no longer a time bonus for a complete autonomous takeoff?
yes, there is. full rules are still available here
Would there be some sort of "style award" for an autonomous vehicle such as a red wagon propelled by an oscillating fan? :D
yes, there are bonus awards for most dangerous, crowd favorite, etc. but they are only medals for glory and bragging rights.
Is there a provision to set the direction (clockwise/counter-clockwise) of the race such that the aircraft are launching and landing with a headwind? Thanks!
So.... anyone want to fund a SICK Lidar system for AVC?
So, the barrels will move. Are they autonomous?!?
undetermined. they will be on the same stretch though.
Are you implying there is a possibility that the barrels would move while the cars are driving the course (ie. the cars would have to navigate around moving barrels)? I assume not, since it hasn't been that way in the past, and there is no mention of it in the rules. Just making sure. Thanks.
oh, sorry, meant to reply to the comment above, about distance from barrels to hoop.
barrels will not move during runs.
I asked above, but perhaps you missed it. Is there a minimum distance between the last barrel and the hoop?
no, not officially, but it will be at least the same distance as the barrels, 15'.
another future AVC idea!! moving barrels!!
The rules say that there will be 15' between barrels. Is that still true?
yes, this is correct.
How about minimum distance from a barrel to the hoop?
Living in San Diego, I can't exactly drop by Sparkfun's building to size up the course, so this video is much appreciated. However, I still feel I'm at a little bit of a disadvantage on some details. So, if someone can help me out, here are a few questions:
Any specs on the barrels, such as where I might find one of the same color to use for calibration with a color vision sensor?
Same on the color of the paint used on the hoop?
Any dimensions on the depth of the potholes (hard to tell from the video.)
Also, since it appears the ground vehicles have to start 5 at a time, what are the rules regarding interference from another vehicle? That is, if another bot swerves into mine, or rams it from behind, is that just my tough luck?
Wayne
p.s. my daughter and I had a blast showing our car at Maker Faire and it was great to see Sparkfun there, too.
1) they are bright red, like the video. I don't think there's a specific color, just 'red'.
2) same color, yes.
3) depth of the pot holes? without completely modeling them in 3D, this is going to be impossible to answer. they aren't very deep. just plan on having a robot that can handle an inch or so of a dip. they could get bigger in the next month. just don't build a tiny vehicle.
as for the color sensing part... I'm not sure that would be your best bet. you need to be almost right on top of something to detect the color. and if you are right up against it, you can just say "oh, that's an obstacle" and go around it.
as for starting at the same time... that's true. if someone bumps into you, you're kinda out of luck. last year, it wasn't too much of a problem. object avoidance is key with AVC, so just make your vehicle avoid objects and remain on course and you'll be good.
Ok RobertC., there was one thing that bugged me about the video... You took the time to measure the starting block on the video and there's also a shot of you measuring at least one defect in the pavement; but despite specifically bringing up the subject of barrel height you didn't bother measure it either on or off camera?:p
well, I doubt anyone would be trying to go OVER the barrels, so I didn't consider height. So, if you need to know the height, it's 35" tall. The diameter is probably more important, and that's 24".
Oh, I didn't think the height of the barrels was particularly important. Although now that you mention it obstacles for flying vehicles to go over, around, or through is an interesting idea for future years... However, it's just one of those things that bugs me (not a lot, but still) when someone intentionally brings up an issue (like you did with the barrel height), already has the means of conveniently solving the issue (it was well established that you guys had a tape measure while filming), and yet they don't bother to do anything about said issue. Consider it a personality quirk of mine.
you do realize it's a standard 55 gallon barrel too... ;-)
I didn't measure the height of the curbs for the same reason, it's a standard dimension that can be found anywhere online.
Is there going to be a stream? If not it would be awesome if you'd stream it on twitch.tv or justin.tv are just some suggestions. Pretty easy to setup all you need is a web cam.
Don't people read all the comments anymore?
just seeing if anyone will waste time replying.
Had the same thought. People just are too lazy now-a-days...
You need to fix all those potholes!
it's part of the competition. we could just do a straight line too, but what fun is that?
How deep are the potholes? Keep hearing that they're a terrible nusance but I was never really sure on their depth...
Here is my comment (found on the comments here) about the pothole depth:
3) depth of the pot holes? without completely modeling them in 3D, this is going to be impossible to answer. they aren’t very deep. just plan on having a robot that can handle an inch or so of a dip. they could get bigger in the next month. just don’t build a tiny vehicle.
I guess Team Dodging Potholes is still appropriately named.
Whoa! I had estimated 860' using Google Maps. I had no idea it was so accurate!
Is there going to be a live stream from the event this year??
Yes! Watch for details as we get closer to the event.
whenever I see a comment from Mike and it starts off with something like "yes!" or "sure!" all my attention focusses to his avatar and then I start daydreaming about futurama...Damn adhd!
Awesome.
I went on-site to test my car out this weekend and the pothole on the west side of the building is nothing to dismiss. It was full of water from the rain showers. One splash and we were done for the day. Had to go home and order a new ESC. I hope it gets delevered in time for raceday.
Ouch. :( If not, and you're in a pinch I have an extra Traxxas XL5 brushed ESCs and Titan12T you can borrow.
No worries, got the new mamba monster in yesterday.
Just a random question here from a sparkfun & Traxxas fan, but what did you take the XL-5 & Titan from? I upgraded my stock Bandit to the velineon system and have my original esc and motor lying around. I'm planning on using these for an autonomous vehicle maybe one day to be entered in the AVC one year...
Got the XL5 used, so don't know. Data Bus is based on an ECX Circuit.
How much will we know about how many barrels and what segments of the course they will be on? Looking for obstacles slows things down.
So I'd like to know when I've found all the obstacles in a certain section of the course so we can speed up till the next obstacle search...
All of this is explained in the video. 4 barrels, on the first straight leg of the building.
Did he actually say the barrel are only in the first straight? He said 4 barrels, but did not say they were only in the first section.....
"After the first bend, there will be 4 barrels, and then the hoop.... and then turn 2"
yes, they are all on the first straight-away with the hoop.
Awesome. Sounds like some great solutions for crowd control and timing! Nice job! Data Bus and I are looking forward to it!
I'm curious why your using a VCR. I would have avoided that like the plague but I'm guessing there's a good reason.
It will be two VCRs with jog/shuttle control so we can easily replay, and also have a record of the event. we could do a hard drive recorder, but a VCR is much cheaper and we don't need high resolution, we just need to see when things cross the finish line.
there is significantly more redundancy and such that's not portrayed in the video. We wanted some simple to operate as well as robust.
Ahem...shouldn't you be using this in addition to photo finish? I have been doing contract work for a professional race track and they use transponders for the vehicles. It would be very cool to use some Sparkfun parts for measuring line crossings, no?
The reason we're not using RFID tags is that the reader modules we carry only read tags within a few inches. This is fine for most applications, but it would have taken over a hundred reader modules to completely cover our start/finish line. We have to pay for these things too, and the cost was prohibitive.
We ARE looking into related technologies like active tags and more powerful readers, so we may take advantage of this technology in the future.
Yeah, I can see that. I wonder if you could increase the sensitivity over a wide area by using a long magnet wire as part of the inductor. Of course that may take a lot more power to get the tags to read. Also, if two tags are being read it may mess up the data that comes back. So it may not be practical anyway. It would be a cool showcase in the future though. If I come up with something I will let you know. Just gotta get my subconscious working on it in the background. :)
The system isn't reliable and it can't read two at once, and there's no way to tell when it actually CROSSES the finish line. It would just tell you when it approaches it.
Couldn't you use something like what they use in r/c racing? http://www.rclapcounter.com/ It works quite well for timing and can do several vehicles at once.
we COULD, but we need 30' to cover and we don't necessarily want to modify the vehicles. plus, it might be pricey doing 30' and 50 vehicles. plus, the video system is a nice visual fail-safe...
Don't make something you can't guarantee works. I like it. And plus don't we all like saving money? If not I would like to have some.
exactly. I don't want failure points.
failure points = not winning!
The tiny booth at Maker Faire SF must of drained your funds
Yeah, that booth was about as minimal as it could get. I didn't even stop there because it didn't look like SparkFun had anything to show in the booth. Not even beginner kits to peddle. I guess SparkFun has enough experience with Maker Faire to know what they will get out of it so they can calculate the ROI, but if SparkFun can't pick up new customers at Maker Faire, then I'd have to say "you're doing it wrong."
Sparkfun had wanted to continue at the Maker Faire with their same presence as in years past, with soldering classes, freebies, etc., but the big giant brains at Maker Faire Headquarters increased their rent by 400%. Now I understand rising costs tough economy blah blah blah, but none of that warrants a 400% increase in booth space costs. They figured that any presence was better than no presence at all, thus the much smaller booth. **NOTE - I am not a Sparkfun employee (yet), nor am I authorized to speak on their behalf. These are simply my own observations and points of view, and do not necessarily reflect the views or opinions of Sparkfun.
HeliCopterGuy got the gist of it, but you can read more about our decision here
OK, I had missed that article before. Make needs to be careful about the formula or the event will disintegrate. There was definitely a different feel this year, although it wasn't necessarily bad/worse, just different. I've been to every MF except the first one (had a calendar conflict). There were fewer big installations this year. Noticeably less fabric arts, unless I missed where they hid it. Hugely more robotics. Everybody and their dog is doing 3D printers now. Also, it seemed like the "Kickstarter launch event" -- there are certainly some synergies there. Parking and food vendors seemed pricier than in the past -- but San Mateo Fairgrounds is driving that. And it is more crowded. It's less maker-to-maker than in the original days.
One thing made me feel influential, though -- there were Morgan manual injection molding machines being demonstrated in both the TechShop and GE/TechShop booths. I was the person that lobbied TechShop to get the IM machine and developed the first training class on it -- so I allowed myself a satisfied smile over seeing two of them on site. (Disclaimer: Not associated with Morgan, just a hobby molder.)
Dosen't look a VCR... considering you can watch the video with 1080p...