Bicycle driven rift games — Oculus
New to the forums? Click here to read the "How To" Guide.

Developer? Click here to go to the Developer Forums.

Bicycle driven rift games

davejonesdavejones Posts: 26
edited April 2014 in Showcase
So last weekend we launched the games we've been developing at the Nati Frinj Festival.

Its a suite of 3 games built around the idea of using a bicycle to drive the game. Each of the bikes ha an arduino and several sensors attached to measure the wheel speed, and handlebar orientation etc. And the bike is hooked up to a generator that creates enough electricity to easily power the rift and the computer that drives it so the whole thing can be self contained.

The games include a fairly standard bike race up the nearby mount arapiles, A helicopter flight over the same terrain and a sheep herding game concieved by the local primary school kids for the festival. The idea was to come up with a bunch of 'gamettes' that in one or 2 minutes of gameplay, would give people a sense of what the whole thing was about. It would be easy enough to flesh these out into something that would keep you busy for hours but the purpose of this weekend was to get as many people as possible to experience it. Over the weekend we probably had a couple of thousand people look at the demo and maybe around 600 actually take it for a spin.

Ill post some more video/screen captures etc when Ive had a chance to put them together but, meanwhile, Here is a short piece the local radio station did on it.

http://www.abc.net.au/local/stories/2013/11/07/3885986.htm?site=centralvic

Comments

  • I'm really interested in your project, as I've actually been working on basically exactly the same thing on and off for awhile.

    It's funny, it looks like you set up a turntable beneath the front wheel and I'm assuming you connected it to a potentiometer which feeds into your arduino? That's exactly the setup I was thinking of, anyway :p

    The biggest issue I've run into so far in Unity is that my bicycle just has a simple box collider on it right now, so it will stand upright and doesn't fall through the floor, but once it gets moving it usually just falls over. I am planning to implement steering so I think that will help since the bike won't head straight down the hill, but I was wondering if your implementation might be a little different. Did you use another sort of collider? Any scripting beyond what you need to read from the serial port for motion, etc?

    I'd love to pick your brain a bit, since I'm not quite as far along as you are. I took a short video of the progress awhile back which you can see here at:

    https://www.facebook.com/photo.php?v=10100425885389213&l=8552735022874821963

    but haven't made much headway since like August :/
  • Very neat idea!

    However I cringed when I saw the Rift with the lenses face up in the blazing sun, as apparently it can burn the screen...
    Seeing is believing
  • Quite clever, very similar to paperdude VR but without the $1200 bike attachment.

    Would love to get some source code for this and throw 1 together myself and add a couple of hydras to the handlebars to track them a bit better, any chance of some arduino plans and the unity source files being uploaded?

    Cheers,

    Dean
  • Zoltan, no worries! Whether or not op wants to open-source his, I'm probably going to open source mine once I implement steering with the potentiometer. The whole rig can be setup for like $35 with a commercial cycling computer, but a simple reed switch without a computer costs even less.

    There are projects already available online that can help you connect the dots, but I'm happy to open source mine as well. It will probably be a thread on reddit with as many helpful resources as I can muster up. My hope is that if enough people get their hands on it this it can become a reality for people to start exercising with the rift, without being prohibitively expensive like paperdudeVR. If people like the idea, maybe I could just sell individual "levels"/environments for like $5 a pop or whatever...

    Thinking of doing a website where others can submit their environments for the community too :)

    Lotsa plans!
  • I should of known other people would have this idea as well.

    I'd suggest using the Microsoft Kinect, so any stationary bicycle could be used, and with the Kinect body tracking would also be a possibility.
  • Sorry its been a while since I looked here.
    Id be happy to open source the code but to be honest I'm not sure its worth open sourcing.
    Its really just me in unity cobbling together stuff from the oculus demo with the uniduino (unity arduino plugin). If I had to sum it up in a word...shambles would be that word. The real experimentation was with the various sensors that through the arduino would tell the computer what the bike was doing.

    In terms of the bike. The turntable on the ground was just to help the wheel turn there was a potentiometer measuring the angle but that was fixed to the steering shaft of the bike.

    tracking the bike to the gradient was done just by having a point for the front and back wheel. the distance btween the 2 wheels is fixed and known so the height differences will give you the gradient and I used this to increase the load on the motor. This is the thing I would most like to improve in my set up. I was using a MOFSET chip to increase the load on the cyclist by turning up or down light (or pouring that power into a battery). The chips I was using worked great in my testing with school kids. they would coast down a hill and then be pretty much stopped in their tracks when the bike went uphill..then had to stand up in the pedals to generate the power to get up the hill (just like the real thing and very satisfying from a game makers perspective). However when some actual good cyclists hoped on the rig they just powered hard into the hill for a minute or so and then the chip melted. and they were then free to blaze away up the hill with no resistance. Im sure their are some similar chips that would cope with the higher load but I haven't found them yet. If anyone on here know more about this sort of thing than me Id be delighted if they would share some insight.

    Im still intending to put together a bit of info about the arduino and bike sode of things Ive just been sidetracked with other thing since the festival. Im going to remount the bikes again at the Falls festival in Lorne over new years (in the unlikely event that someone who is reading this might be thinking of going there). Its a little intimidating to think how my bike and oculus combo might cope with 16,000 drug fueled kids.
  • ATomasik wrote:
    I should of known other people would have this idea as well.

    I'd suggest using the Microsoft Kinect, so any stationary bicycle could be used, and with the Kinect body tracking would also be a possibility.

    To be honest I think measuring what the bike is doing is more efficient and more relevant in this sort of scenario.

    If you know how fast they're pedalling(arduino/hall effect sensor), what direction they're steering(arduino/potentiometer) and you which direction they're looking (oculus), you can assume their hands are on the handle bars, feet on the pedals and head about a meter above the seat. What else do you need to know?

    Plus I think the bike itself would confuse the hell out of a kinect sensor.
  • Dave,
    That's amazing, I didn't realize you had implemented graduated resistance at all! That was something I wanted to do near the end, but had only a vague idea about how to do. Sucks to hear the mosfet chips are burning up under load, do you mind my asking how you are providing the actual resistance? I was considering an electromagnetic setup but think fabricating stuff would be too expensive and kinda defeats the purpose. So far the only two trainers I know of that support gradient-dependent resistance are the wahoo kikr and computrainer, both running roughly $1500 I think.

    It looked like you were just using a normal stationary trainer in the video, you mentioned having a motor hooked up to it? What sort of motor were you using? I'm guessing the average cordless drill just wouldn't do the trick ;)

    I wish I was able to come ride the thing myself, it looks awesome :)
  • I started with a basic trainer ($70 not $1500) but swapped out the roller for a new one fitter to a scooter motor ($40).
    the original reason for doing this was so that the rider could generate the electricity required to power the whole thing (which they can do comfortably). I was demonstrating to a bunch of primary students how much work it took to power an LED vs a halogen bulb etc etc when it occurred to me that if I could control the load dynamically then that would create a much more compelling experience for the cyclist. Essentially its just a halogen bulb and a dimmer (the MOSFET)...the brighter the bulb the harder it is to pedal. It would be nice to just have this power going into a battery for later use (totally doable) but at the moment I'm more excited about the user experience.

    There are a few more pics in this blog post here

    http://theartofdave.blogspot.com.au/2013/12/art-pumping-action-bike-tech.html

    Its pretty basic but if you have any specific questions Im happy to answer them. I'm pretty sure that if I just get a high spec MOSFET chip (like spending $5 instead of $2)... then it wont burn out. But that's what they had in the local store and Im about to go and install this thing at a music festival with 16,000 youths who will be too out of it to appreciate the difference so I just going to wait and sort that out in the new year.
  • Heres a quick clip of the game in action at the falls festival.

    Responses ranged from genuine interest in how the whole thing worked right through to this

    https://www.youtube.com/watch?v=bZzk3PPB8mY
  • LOL those guys sound like they were having the time of their lives. Really cool to see people getting so excited about the experience! :)

    And thanks for the killer write-up!! If I ever get the time to move toward a more completed phase I will most definitely share the results of my testing on the general populus here in Baltimore :D
  • nerocomnerocom Posts: 88
    I am really looking forward to seeing this go somewhere (tee hee).

    I am borrowing an indoor bike for the winter (basically this guy: http://www.wayfair.com/Trixter-X-Bike-400-Indoor-Cycling-Bike-XBIKE400-XHT1001.html) and it's alright, but nothing like biking on the trail.

    If anyone ever needs someone to test, I am more than happy to :) I have a late 2013 macbook pro (running both osx and windows 7 x64) with the geforce 750 chip, the bike, and the rift!
  • I am having the damnedest time getting steering to work. Whenever I manage to implement it to an acceptable standard I'll do a writeup on the system with a quick bill of materials and open-source the code. Right now steering kind of works but basically doesn't. Mapping the values from the pot to degrees of turn in Unity is more difficult for my nooby brain to wrap around than I expected it to be :/
  • Awesome, this is exactly the sort of thing my house mate (personal trainer) wanted to do the second he saw the Oculus demo.

    Great to see another Aussie doing tome great work with healthy VR!
  • jcservek You should be able to have it just tell you what analogue value you're getting back from the pot max 5 or 3.3 volts. half that is your straight ahead value...then workout (estimate) how far the handlebars turn from that. thats your max turn.

    When I was running my one at the festival the thing that shocked me was that people rode the bike like a computer game. when they wanted to go left, they slammed the handle bars left as hard as it would go and the same for right. that sort of behaviour would send you over the handlebars of a real bike in an instant but it was pretty much the standard aproach.


    the other thing was that some people...the actual proper riders hand an overwhelming urge to lean into the corners. It would be great if you could measure this and use it in the game but as it was, it was a total liability. I had to try and catch those people or physically hold them on the bike to stop them injuring themselves.
  • So I finally got my controls working to an acceptable standard :) I haven't tested the system with the rift yet since my Unity Pro trial expired on my laptop. I'm going to install a new 30 day trial on my desktop and put the rift camera in this afternoon and ride it for the first time, at which point I'll update you guys on how it goes (and maybe make a short video if I have time?)

    My system isn't nearly as robust as Davejones', but the basic functionality is there and the hardware cost is typically less than $50 (Not counting the rift. And $30 of which being the arduino)
  • jcservek wrote:
    So I finally got my controls working to an acceptable standard :) I haven't tested the system with the rift yet since my Unity Pro trial expired on my laptop. I'm going to install a new 30 day trial on my desktop and put the rift camera in this afternoon and ride it for the first time, at which point I'll update you guys on how it goes (and maybe make a short video if I have time?)

    My system isn't nearly as robust as Davejones', but the basic functionality is there and the hardware cost is typically less than $50 (Not counting the rift. And $30 of which being the arduino)

    Excited to see this :)
  • kalimkalim Posts: 3
    Hi All,

    I have just read through this thread and it seems the appropriate place. Great work might i add. I have checked out a few of your links.

    Myself - I am building a bicycle simulator and I want to make this as inexpensive as possible.

    What is the least it would cost to build? without including software and development which I have covered.


    If correct. I need...



    A bike trainer (I want it to be magnetic or motor brake so it engages with the simulator. Any recommendations? Inexpensive..)



    3 sensors - Handle bar - Cadence - Reer wheel Tyre. (i cannot find handle bar sensors ANYWHERE! Does anybody know where?) or do I have to create a potentiometer myself? surely not?



    What else will I need? I know tacx make full working systems and as it is open hardware/software I can manipulate that but I dont want to pay £4-900. Any solutions?



    Thanks a bunch...



    Daley
  • kalimkalim Posts: 3
    Hi All,

    I have just read through this thread and it seems the appropriate place. Great work might i add. I have checked out a few of your links.

    Myself - I am building a bicycle simulator and I want to make this as inexpensive as possible.

    What is the least it would cost to build? without including software and development which I have covered.


    If correct. I need...



    A bike trainer (I want it to be magnetic or motor brake so it engages with the simulator. Any recommendations? Inexpensive..)



    3 sensors - Handle bar - Cadence - Reer wheel Tyre. (i cannot find handle bar sensors ANYWHERE! Does anybody know where?) or do I have to create a potentiometer myself? surely not?



    What else will I need? I know tacx make full working systems and as it is open hardware/software I can manipulate that but I dont want to pay £4-900. Any solutions?



    Thanks a bunch...



    Daley
  • kalimkalim Posts: 3
    Hi All,

    I have just read through this thread and it seems the appropriate place. Great work might i add. I have checked out a few of your links.

    Myself - I am building a bicycle simulator and I want to make this as inexpensive as possible.

    What is the least it would cost to build? without including software and development which I have covered.


    If correct. I need...



    A bike trainer (I want it to be magnetic or motor brake so it engages with the simulator. Any recommendations? Inexpensive..)



    3 sensors - Handle bar - Cadence - Reer wheel Tyre. (i cannot find handle bar sensors ANYWHERE! Does anybody know where?) or do I have to create a potentiometer myself? surely not?



    What else will I need? I know tacx make full working systems and as it is open hardware/software I can manipulate that but I dont want to pay £4-900. Any solutions?



    Thanks a bunch...



    Daley
  • kalim wrote:
    If correct. I need...

    A bike trainer (I want it to be magnetic or motor brake so it engages with the simulator. Any recommendations? Inexpensive..)

    3 sensors - Handle bar - Cadence - Reer wheel Tyre. (i cannot find handle bar sensors ANYWHERE! Does anybody know where?) or do I have to create a potentiometer myself? surely not?

    What else will I need? I know tacx make full working systems and as it is open hardware/software I can manipulate that but I dont want to pay £4-900. Any solutions?

    Daley

    the potentiometer for the handlebar you can buy for about $2 in an electronic shop. Put a roller on it thats about the same diameter as the bike steering column and press it up against that so you get pretty much a 1 to 1 ration as you turn the handlebars.

    hall effect sensor for the back wheel $3. and a magnet ($0.10?).

    Im using a scooter motor on the back ($30) which I can increase or decrease the load by drawing more current from it. If you werent bothered about this you could just go with the restistance that came with the training stand.
    jcservek wrote:
    My system isn't nearly as robust as Davejones', but the basic functionality is there and the hardware cost is typically less than $50 (Not counting the rift. And $30 of which being the arduino)

    Haa! Robust is about the last word I'd use to describe my set up. Glad to hear you've got yours happening now. Post some photos when you get a chance.
  • jcservekjcservek Posts: 11
    Sorry it took me so long to get back, I've been super busy the past few weeks. I had the chance to demo it for other people for the first time tonight at a show at the Maryland Institute College of Art where I work. I only had the opportunity to take a crappy cell phone video, but you can see it here (and PLEASE pardon my awful voice, the phone was much too close to my face)

    https://www.facebook.com/photo.php?v=10100606677929513&l=7781765925291983830

    As you can see the steering is done via potentiometer, but I put mine inside a block with a turntable on it and attached a normal climbing block to the top, so as the wheel turns the block it twists the pot and thats how I get my values. I didn't do a 1:1 ratio because I find slamming the steering wheel right and left severely throws people off balance, so I made the steering quite sensitive which imo felt much better. Sooner or later I'll get some pics up with more detail and do that writeup I keep talking about if anyone is interested in trying for themselves. (I'm not sure that I would do the steering block the same way again, for what it's worth. I think there is a better way, without putting the potentiometer in the bike itself)

    If any of you guys have any ideas please feel free to share! :p
  • davejonesdavejones Posts: 26
    nice vid.

    I also toned down the steering ratio to try and get people to think before slamming the steering hard left and right. Really surpirsed me that would do this...One time doing it on a real bike and you soon learn about moderation....but somehow now the onus is on us to provide a safe experience.

    did you have any resistance on the back wheel?
  • jcservekjcservek Posts: 11
    Only the resistance supplied by the trainer itself. I was using my existing fluid trainer. I'd like to get a magnetic or wind and hook it up to a motor so I can get feedback like you did, but recently my efforts have been focused on other aspects of the system.

    I probably should have asked before, but are you by any chance using Uniduino? Because I'm not (I just coded the steering and pedaling by hand) but it's starting to seem worth it as I keep moving forward. If you're using it, do you like it?
  • sqhsqh Posts: 21
    I'm working on one too! https://github.com/shalperin/vr-bike
Sign In or Register to comment.