Hi all,
I’m looking at making a robotic garbage bin collector. This is still in the concept stage. However the expected timeline for this project is 8 months. The real world application for this project is the collection of 35 full size outdoor general waste bins as well as 35 half sized glass recycle bins, from a community of villas within an enclosed compound. As stated this is still at the planning stage so alternative suggestions are most welcome! Also interested parties or people who know people, that know people with experience in robotics or engineering who might be interested in this project would be great too. Please feel free to P.M me for details. Or I will be posting a detailed project outline within the next week.
Stay Awesome,
Mitch.
just a few questions:
1- are you planning autonomous emptying too, or just relocation?
2- maximum full weight of one of these “240litre” bins might be over 400kg eg, if its full of concrete and lead from a renovation how will u deal with that?
3- would 35 permanently attached wheelbases, and a mechanism that just tows them around be within-spec?
4- what is the width/space of the area in which the bins will be located? Eg, will there be need to limit the width of the vehicle to less than (eg) a standard car/truck width?
5- Will the vehicle ever have to reverse, or would a X sized turning circle work?
6- will the vehicle be operating concurrently with other users?(pedestrians/drivers? )
7- who will be liable for when it drives into a ferari, or over a loved pet, or worse?
Great project!
Question 3 is great. If a wheel mechanism could be attached to the bottom that would present interesting options to avoid the need to lift.
Also +1 on the do they just get moved to somewhere and then the bin truck comes and empties them question.
My last workplace was like that. The wheelie bins from around the school were wheeled to a central point for the bin truck company to come and empty by rolling them up to the truck. There wasn’t a central dumpster, just a pen for bins. If this is the case for your scenario too the bin company might not be adverse to having clip on wheel bases as long as it doesn’t affect their lifting mechanism.
Is this hypothetical or real world?
Hi Buzz,
Thanks for the great questions.
-
For the time being just relocation however there is the intention to eventually develop the ability to empty into a either a 660L bin or a small skip. There is also the intention to have these be cleaned as near to autonomously as possible. (Please note the intention here is to eliminate manual handling not workers).
-
Maximum full weight of these bins on average is 96kg. The bins however are emptied daily so are often much less than this ( usually 20kg or less). The bins are also in a very controlled setting were rubble and cement etc is not an option.
-
This is an option however a completely separate system is desirable although not compulsory. These bins are renewed at least annually if not more frequently. The bins may be modified if required.
-
Desirable form factor is up to 1.6m2 however 1m or less is desirable.
-
Turning on a dime or reversing short distances is desirable however other options may suffice. This is only relevant to the initial loading of the bin onto the BinBot.
-
Very rarely will there be others vehicles or pedestrians however this is in an environment that can be accessed by select members of the public. It is expected that the machine will operating outside regular hours.
-
This vehicle will be operating within a closed privately owned compound. All items within the compound are owned by the same entity. Public liability and other insurances are in place in the event this dose happen. However safety is priority number one on this project.
Hi Nathan,
Thanks for your interest 
The bins are taken to a bin compound and emptied via a rear load compactor.
This is for a real world application.
so you need:
A - some motive equipment capable of moving around maybe 100kg.
I was thinking something on tracks for manouverability,
or mecanum wheels for uber manouverability
something off-the-shelf if possible to minimise development time/cost
if it’s a residential area, then something electric powered has noise advantages, particularly for night operations
B - a control/autonomy system
capable of driving the vehicle under GPS steering with preferably accuracy of a few cm or less, not a few meters.
capable of “aiming” the relevant part of the vehicle at the bin
C - mechanical interaction with the bin/s
for hookup if on a fully wheeled cart
for liftup, if “carried” onboard
for some combination ( eg, locking onto handles, and tilting bin onto its own wheels for tow)
random thoughts/solutions:
A motive system…?
dingo yard digger ( http://www.dingo.com.au/ ) is skid-steer ( turns on a dime), can also have optional “tracks”, includes hydraulic bucket “lift” and hydraulic bucket “tilt” that seems to me like it would be trivial to add an attachment to, for a “bin lift/tilt” mechanism , has a *fully remote controlled" option ( http://www.dingo.com.au/dingo-range/contractor-remote/ ) that could be taken advantage of to get the end solution working much faster.
Downsides: it’s not electric, so noise may be an issue, but adding a custom exhaust to one of tehse puppies is going to be way easy compared to making one yourself.
OR: a modified electric forklift, maybe with mecanum whells like this: https://youtu.be/vAiwLRGsNrE?t=26s the “lift” in the forklift gives options for interacting with the bins , also some forklifts already have “remote control” ( ie drive-by-wire ) , such as MORA forklifts , so that also helps minimise development time/risk ( https://www.youtube.com/watch?v=TO4t-fF1gKA )
other less well formed ideas: maybe a 1/4 scale RC tank, or tractor? ( can u find something like like this off-the-shelf? ) or a quadbike, or a golf cart?
B control / autonomy…?
the “control” part is solved on the dingo and drive-by-wire forklift already, but the “autonomy” part would need something like “ArduRover” ( http://rover.ardupilot.com/ ) here’s two actual videos of ArduRover being used to steer a typical farm tractor down a road: https://www.youtube.com/watch?v=HBQwaaLT56o and being used to steed BOTH a combine harvester AND a tractor by the same operator, at teh same time: https://www.youtube.com/watch?v=Ybxhvlyw-X0&spfreload=10 . You might need an expensive high-resolution GPS if you need more than 10-20cm accuracy, but then, that might be OK.
alternatively you could pay big $$ to a commercial company that specializes in GPS navigation systems for agriculture ( maybe John Deere, etc https://en.wikipedia.org/wiki/Driverless_tractor#Current_leading_manufacturers ) , and come out $100,000 or more worse off, but have it all sorted out for you.
third option you use some form of guidance that’s not GPS based - eg underground buried wire, or painted line on the ground, or [insert madness here]
C mechanical interaction…?
the dingo and forklift both have what I think is the essence already, the can both lift and/or tilt heavy things with their existing capabilities, so it’s then just a case of interacting with the bin itself, either with a “grasper” ( like the rubbish trucks ) which is complicated and expensive , or with a hook that grabs the hinge/handle and/or “lip” ( it sticks out, so is easier to work with , but requires bins be placed the right way around ) .
related:
also: electric bin tippers exist: https://www.youtube.com/watch?v=xH4WM3SSul0
and wheelie bin “towers” ( for behind a car ) : https://www.youtube.com/watch?v=qbPrdo7QU3M&spfreload=10
The easiest solution I can think of for cleaning the bins would be for them to trundle off to a separate cleaning area after they’ve registered as being emptied, then a spray mechanism attaches to the opening, hoses the inside, then inverts and drains the bins.
For GPS correction
Hi Firedrake this is exactly what I was thinking also.
Hi Catprog,
dgps looks very interesting. Are you aware of any commercial packages available for this kind of tech?
regarding GPS… you’ll need to learn about more than DGPS is you want to know about super-accurate GPS. you’ll need to know about these to: RTK, RTCM, SBAS, WAAS, L1 & L2 bands, and possibly even GLONASS. I suggest you google all these terms and learn about them.
Hey Buzz,
Once again hitting the nail on the head!
Tracks unfortunately are not an option however lines or embedded markers is not out of the question. It would be nice having something a bit ‘smarter’ than these systems though.
Electric is definitely the go. Perhaps we could even convert a couple of those electric unicycles from Anderson electric. Hint, hint. Nudge, nudge. Wink, cough* subtle suggestion. Although the dingo and forklift are suited for lifting etc they are to large and require to much maintenance etc. For this scenario. I intend to build a custom frame for this application.
I am not aware of any commercial DGPS but a google search has given me this link
https://www.ja-gps.com.au/DGPS/DGPS-Receivers
http://diydrones.com/m/blogpost?id=705844%3ABlogPost%3A1984437
http://diydrones.com/m/blogpost?id=705844%3ABlogPost%3A2148607
Are both high res gps units, and the second one is: “dual frequency”, which traditionally costs more than $10k!