Gorgeous Engineering Inside Wheels Of A Robotic Trail Buddy

Robots are great in general, and [taylor] is currently working on something a bit unusual: a 3D printed explorer robot to autonomously follow outdoor trails, named Rover. Rover is still under development, and [taylor] recently completed the drive system and body designs, all shared via OnShape.

Rover has 3D printed 4.3:1 reduction planetary gearboxes embedded into each wheel, with off the shelf bearings and brushless motors. A Raspberry Pi sits in the driver’s seat, and the goal is to use a version of NVIDA’s TrailNet framework for GPS-free navigation of paths. As a result, [taylor] hopes to end up with a robotic “trail buddy” that can be made with off-the-shelf components and 3D printed parts.

Moving the motors and gearboxes into the wheels themselves makes for a very small main body to the robot, and it’s more than a bit strange to see the wheel spinning opposite to the wheel’s hub. Check out the video showcasing the latest development of the wheels, embedded below.

[taylor] has an image gallery of the current hardware to go with the video above. Robots whose primary purpose is exploration are always exciting to see; some examples include PUFFER the rover and the Autonomous Underwater Glider (winner of the 2017 Hackaday Prize!)

23 thoughts on “Gorgeous Engineering Inside Wheels Of A Robotic Trail Buddy

  1. Does anybody knows why this solution, placing of independent motors on each wheel isn’t used for full sized electric cars? It’s much easier to control motors than build differential gearbox ….

    1. Unsprung weight sucks when you scale it up. Basically it gets unpossible to control the wheel and keep it in contact with the road. Wheels not in contact with road = terrible cornering, terrible braking and skippy/squeaky/inefficient acceleration.

      1. If you scale the design, you could use standard suspension techniques while pulling the motors closer to the center of the car. My guess is cost. I’d bet it’s cheaper for linkage systems rather than another motor.

      2. Don’t forget terrible ride quality! For a vehicle as large as a car, if you want individual wheel motors, it’s best to put them inboard and use an axle to minimize unsprung weight.

        Also the reason for the differential has more to do with gearing than packaging. Most single-speed electric cars are geared down around 10:1, if you changed them to direct drive their acceleration would be extremely slow.

        1. Tell that to the AMG engineers who built a prototype SLK with four motors in it. They implemented the differential in software and could tune the parameters in ways that would be impossible with a mechanical diff, e.g. making the car over/understeer by means of torque vectoring.

          1. Four motors do not necessary mean they have to be in the wheels. You can place them in the frame and use half-axles like with a conventional diff. That reduces unsprung mass.

          2. Current (and previous) Range Rover does that torque vectoring stuff with electronically actuated diffs/brakes/clutches, it’s not unique or new.

            Also, prototype to test some particular thing does not mean it would work or be a good idea for production.

          3. @JohnU: traditional torque vectoring via brakes/clutches/diffs doesn’t allow the things that AMGs prototype vectoring allows.

            Traditional vectoring changes characteristics by slowing something down. Want to make the car oversteer in a right turn? Apply brake on front right wheel. This is also how ESP works. But it makes the car slow down and also isn’t possible for extended periods of time. Technically you can’t even call it torque vectoring, because it doesn’t direct the torque to where it’s supposed to go, it slows the side/wheel where the torque is _not_ supposed to go.

            In the case of the AMG prototype, they actually vary the torque applied to each wheel instead of using friction. This allows to improve the turning characteristics without the drawback of slowing down. I mean, we’re basically talking about a race car here, using brakes to alter the handling wouldn’t be beneficial for the laptimes.

            I remember there was a HaD article about a guy who built the same with a 1:10 RC chassis and four brushless motors. Very nice build

        1. You are wrong. No freeway-speed cars running on imperfect surfaces will ever use hub motors. Below ~40mph top speed is possible or perfectly-maintained surfaces (racetracks.) Every single example (of which there are very few, since it is such a terrible idea) you can find has horribly unacceptable ride on real freeways.

          Toys and slow vehicles it works just fine. Kinda like you don’t find exoskeletons on anything above small insects.

        2. Plot twist, brake rotors are way lighter than hub motors required for the application and CV axles are supported at two points.

          Unsprung weight is bad. Really bad. Take a car and put on some nice light wheels and go over a bumpy handling track. Works really nice. Take the same car and put on some really heavy wheels and go over the same track. You will notice the difference in comfort and handling. Mainly because mass inertia is a bitch.

          1. > “Unsprung weight is bad.”
            Then you go on about handling being better is wheels are lighter… which is so.
            So I think you meant more sprung weight over less sprung weight is bad.

            I think we all mean that the wheel assy is sprung weight, although some logic could put wheels as “supported firmly by the road,” which makes the vehicle chassis and body and engine and fuel, all as sprung weight… which may be more accurate, but not how the term is used, colloquially.

          2. @BillSF9c In automotive unsprung mass means parts that are not supported on springs/dampers.

            Which mostly means the whole assembly below the springs. So: Wheel, brake, hub, bearing, knuckle, lower part of damper/spring assembly and partially the control arms. The latter ones partially belong to both sprung and unsprung mass, because they’re attached to the sprung chassis and the unsprung knuckle.

  2. Really nice design. I’ve been wanting to print a rover of some type ever since the cost of 3D printers started coming down, but I don’t think I’ve got the patience in my unfortunately.

  3. Has the builder seen a hiking trail?

    I mean there’s those short mile or two “accessible” trails with 3ft strip of pounded gravel, even asphalt, but often you’ll find stretches where it’s a 9 inch wide groove, with knee high grass/brush each side. Then there’s fallen trees and branches to step over, stiles, steps and stairs….

    This design looks like it might cope with the better maintained sidewalks, but I could probably find it a spot where it would have trouble with a root, broken concrete or a narrow gap.

    1. Gah, HaD hack mis-spin strikes again. From the builders page it read more like it’s an outdoor rover development platform than the purpose designed for hiking trail use robot the Papp writing makes it out to be.

      1. If you actually watch the video, he specifically talks about the NVIDIA software classifying trail objects, and the fact that it was trained/tested on the trails near his location. The platform can be used for whatever you want, but the OP’s intentions for it clearly included using it as a trail buddy. Just because you can comment, doesn’t mean you should.

        It’s a fine article @DonaldPapp, well done!

  4. I’ve wanted to do a rocker bogie style robot for a couple of years, but always get hung up on the drive system. This looks like a great solution with common off the shelf hardware.

  5. > “Kinda like you don’t find exoskeletons on anything above small insects.”

    Aside from most submarines, some spacecraft, and nowadays, even many cars. I suppose bamboo, being immobile, is too vegetative to enter into the fray.

    I see in many HAD discussions, You may have X or Y and some mix of each, but never most of both, so I yield your point… to the degree that sprung motors are involved… but 4 motors may be had and benefits gained without much other ancillary loss, if shafts or etc are an intermediary as motors need not be sprung to have 4×4.

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.