Any mech pilot will tell you that without a working (mechanical) gyro, you’re dead in the water. Or at least that’s what they will say, in the future, when driving ours. Stabilisation is the primary barrier to getting anything weighing several tons off the ground more than a few feet. This post reviews the various gyros options and concepts in detail. We’ll leave the math for another day though, so read on without fear. First, some inspiration from a real world implementation.
Here is a short video showing Lit Motors and the role the gyro plays in their self balancing motorcycle, the C-1. It gives you the general idea of how a gyro can perform in a small package:
Obviously the wow factor is that the vehicle doesn’t tip over. How does it do it? They use a combination of two gyroscopes, working in tandem, to direct the torque built up in their super fast spinning flywheels. Technically this arrangement is called a Control Moment Gyroscope (CMG) – some other areas of application use even more working together to provide control over more degrees of freedom or to provide even more torque.
If the concept of gyroscopic angular momentum is new to you, check out this old and intriguing demonstration to see the forces we’re talking about in action:
By adjusting the orientation of two or more gyros together they can do amazing things – both these videos only show the basics. For instance, the C-1 would be able to have the bike lean into a turn and keep itself “locked” in the position until coming out of the turn, using only a few tiny mechanical adjustments to the orientation of the gyros. On ice, this arrangement could theoretically spin the vehicle in circles – all while staying upright. Throw it into a hovercraft and you’ll have a very interesting control platform – similar to how they use it in the space station. Some small-scale battle bots even use a similar approach to flip their robots over – not necessarily a good idea for the C-1 but you get the point.
How does this apply to a 4m high mech?
Scale up this gyro solution and it could provide the needed forces to help keep a mech balanced. Once that is achieved, then it can take on the roll of handling the movement of the centre of gravity during a walking cycle. Observe yourself walking. You can’t simply lift one leg and move it forward. Instead you have to shift your weight to the other leg first. It’s very subtle but critical. Now observer yourself running, you’ll see that there is less shifting from leg to leg as your spending more time “falling” forward and your legs catch your weight. We want to handle both scenarios using a CMG and crucial timing for our actuators.
You can see how the servos actuated robot platforms do this sort of shift in balance… it ain’t pretty, nor fluid, nor fast as their little servos move the hips around while lining up for taking a step (there are some awesome exceptions however). This is partly due to the mechanical movements of various parts under increased stress – if humans walked like that we’d look pretty funny. The CMG does truly operate more like how our human balancing system. We are constantly shifting weight in small degrees and we use torque-like forces that are available because of our amazing musculature and nervous systems. A CMG uses electronics to shift the axis of rotation. The timing is crucial but gives an unparalleled ability of control.
The forces at play are not insignificant either. By spinning CMGs up beyond 6,000 RPM, they put out serious amounts of force even with a small diameter of, say 12 inches. Enough that in the C-1 two of these can control the balance of an 800 lb vehicle while being slammed with hundreds of pounds of force from the side. Naturally this is different from a tall mech, but the forces at play are similar. (We’ll get to the math in another post, but by scaling up it does appear quite possible to handle the weight of a small mech.)
Barriers to using this CMG approach become obvious after researching the state of current developments. There are few companies creating gyros, even fewer creating ones that are larger than a toy and even fewer that are doing anything like a CMG arrangement. A couple of aerospace manufacturers do this but it is highly customised business and for applications that you and I will likely never see.
So the main problem is sourcing for components to build a CMG arrangement. Thankfully, as a CMG is critical to Lit Motors and the C-1 development, they are now leading the pack in terms of (publicly accessible) R&D on the subject – especially considering that their product is aiming at a consumer market! So we’re hopeful that as they build their expertise in this area even stronger, they will be able to provide solutions to other application areas needing a similar platform. You can read more about their patent here to see that monitoring arrays of sensors, and applying them in real-time to CMG adjustments, is absolutely core to their needs.
More to read:
- Patent granted for Lit Motors C-1 gimbaled flywheel stabilization system (Autoblog.com)
- Not a new idea.. see old Popular Mechanics article
- The Mitsubishi Anti Rolling Gyro stabilizer Rules the Waves – and another sea motion commercial implementation (Seakeeper)
- Not just toys – gyroscopes.com – also have scientific demonstration units including CMG arrangement
- If you want to search for more, be aware that mechanical gyros are units with physically spinning discs or flywheels, whereas a lot of what is called gyroscope today are also known as MEMS gyros – these are simply a chip that detects changes in orientation. Often MEMS gyros are used to provide sensor feedback to a system that can then adjust the some servo or motor to achieve the desired effect – i.e. balance on a Segway, etc. There’s probably one in your cell phone…