For those that were around pre 2012, think back to the state of the average team’s ability. What was it missing that is common today? Machine vision.
Then 2012 happened. 341 happened. They caught everyone’s attention with their closed loop control for aiming at the high basket. They showed that it can be done, and done well at that.
Fast forward to now. We have ni vision, we have grip, we have team 900 who used the same technology in self driving cars in 2016. While having some form of machine vision is not quite the norm, it is trending to be that way.
So, what do you think the next technical leap frc will experience?
it wasn’t so much a technical leap, but wide, varied, and inexpensive COTS gearboxes and the versa planetary are what has had the biggest impact in growth for teams. Very few teams use vision, EVERY team uses gearboxes.
The next technical growth aspect is going to be something that raises the floor for everyone, not just those with an interest.
I don’t think this is a super accurate version of the events.
To some extent, machine vision has been a thing since 2006 - teams used cameras effectively that year for the high goal.
The main reason camera tracking was not a thing in 2009 was the new control system combined with the moving, non-lit target. Teams certainly did use camera tracking to some extent, but often / usually, manual tracking was faster.
In 2010 and 2011, your robot started in a known location facing a stationary goal, so what was the camera even for?
In 2012, a similar argument applied, but when teams started moving around the camera got beneficial. The rules also opened up to start allowing coprocessors sometime before 2012 which was a big help. 341 was a notable, highly visible example of camera tracking, certainly, and they had an effect, but I think it’s a bit simplifying and disingenuous to suggest that camera tracking wasn’t taken seriously until 341 demonstrated it being used effectively.
In 2013 and 2014, see 2010 and 2011. 2015 doesn’t count.
So 2016 was really the only opportunity since 2012 for camera tracking to be an advantage. You had to cross defenses before shooting which made the physical position of the robot on the field not nearly as certain. This is why you saw a lot more cameras for autonomous that year - as well as flashlights for teleop.
So really, the use of cameras is much more driven by the game than by the coolest example of it from a particular year - and cameras have been an available option to FRC teams in one way or another for many, many years. There certainly was a dramatic shift in capability between the pre-2009 CMUcam and the 2009 NI system, as well as a shift in 2012 the first year coprocessors were allowed when a camera was a potential advantage. But camera use is really a response to the conditions of the game more than anything else.
I would agree with Chris’s analysis of cameras. Camera use really does track with the benefit you get within each game. There’s almost always something you can do with a camera, but many times the benefit just isn’t big enough to justify the time investment.
And as Eric said, suppliers like AndyMark, Vex, WCP, and a few others have gone a very long way over the past 10 years or so. The floor for teams is so much higher now than it used to be because of the ability to purchase solutions to a lot of problems, mostly surrounding gearboxes, mounting solutions, and drive train options. That’s not to discount other solutions (BaneBot’s line of wheels for a while, AM’s compliant wheels, linear servo’s, etc) that make particular challenges easier, those solutions just aren’t as universally applicable, depending on the particular game or strategy design.
I think we’ll see continued growth from our suppliers, as they add more and more out of the box solutions to problems teams often have. It’s almost turning into a faceoff between the GDC and the suppliers - can the GDC stay one step ahead of the solutions suppliers are putting out, so the game remains a challenge?
Already this year, you can build half of a robot from off the shelf systems from AndyMark alone. They have an intake system. They have a winch. They have a ton of different drive train and chassis options. I don’t think I’ve seen anything from a supplier with a solution for a gear manipulator, a shooter, or an indexer, so there are still challenges for teams to solve. That’s a huge advantage that wasn’t even dreamed of by most teams 10 years ago.
Not quite a technical leap, but I think 3D printing is going to continue to grow in the FRC community. It’s such a great resource for teams to have. A relatively cheap and convenient way to manufacture parts. Not all teams have access to machine shops, but for under 100$ they can start printing complex parts in their class rooms.
Next plausible leap is likely fire on the move robots that are making high percentage of shots. This is a good deal trickier than static shots teams are currently taking (and the moving shots teams made in 2014 were far easier than a 2016 or 2017 shot).
There is no technical reason why robots couldn’t currently be doing that, just a lot of software to develop.
We tried that back in 2007 didn’t go well for us either
I believe the real leap in vision processing has come in the form of cheap Android phones that are incredible powerful with great cameras on them, props to 254 for leading the way. The tracking frame-rate and ease of development you can get from building vision in an Android app is way ahead of any other co-processor/on-board vision processing I’ve seen. I’d love to see some common tools develop around this, the communication between the App and the Rio can be tricky right now.
It would be awesome to see every kit include a powerful $200 Android phone with some standard tools to do vision processing with.
How about linear actuation? Of course pneumatics have been used forever, but these electric actuators are starting to pop up. They allow you linear motion without the weight of an entire pneumatics system.
The DART actuators last year and now REV has their hi power servos and AndyMark offers their linear servo.
I’d like to see the VersaPlanetary adapted for linear motion. It would provide a smaller footprint than the DART actuators with sufficient power to do many things on FRC robots. The multiple gearing options would continue to make this versatile as well.
We were capable of doing this decently last year, but because the cost of missing your one and only ball far outweighed the benefit of saving a fraction of a second, we didn’t perfect it and never did it on a competition field. Being able to stop on a dime and begin shooting a hundred milliseconds later is nearly as useful as shooting on the fly, but substantially more robust and tractable (given the timing and delay properties of FRC components).
I’m sure you’ll see many robots that are mechanically capable of this feat this year (e.g. turrets and continuously variable shot ranges), but I will venture to say that this will be nobody’s “Plan A”…the set of teams who I believe have the mechanical and software expertise necessary to pull it off are also competitive and pragmatic enough to choose a 99% solution that requires 10% as much work.
Smaller, inexpensive water jet cutters that can cut 48x48x.25in aluminum to .001 in accuracy at 100 in/min for less than $1000 and a tech grant to get one in KoP
When designing a mechanism it is really nice to be able to say “hmm, I’m going to need a reduction of between 3:1 and 100:1 and I’ll put a Bag motor in there but maybe eventually I’ll swap in a 775Pro if I need a little extra power” and just put the same hole pattern in and be done.
I think a huge leap over the last 5 years was the relaxation of pneumatics rules. Letting teams use air for more things and not limiting them to 4 small tanks has changed how teams design. Ok, maybe I’ve just been talking to Foss too much.
I look at this a bit more philosophically…there is a great deal of technology out there that the average team knows about or has access to with …(fill in your favorite search engine)…but the issue is and has always been how to replicate something on a team budget, max component value of $400, CAW max of $4000, or access to advanced manufacturing methods. I think the next leap forward will depend on whatever industry drives the price down on or whatever mass sponsorship resources start popping up.
I agree that the next most probable advance will be in motion control. We are already seeing integrated industrial linear and rotary servo motion falling below $400 and at power and load ratings useful to FRC teams. We are also seeing industrial and tech sponsors catching FIRST fever, from both philanthropic and smart leveraged marketing reasons, giving more manufacturing resources to local and global teams. Also, more teams are embracing CAD and modelling to accelerate the prototyping and design process.
So, my prediction is that the next wave, though less obvious and harder to measure, will be that the average team will make big leaps in build-to-print or build-to-model robot designs. We will see more off-the-shelf technology being integrated into designs as prices fall, and more CNC, laser or water-jet cut parts on the average robot. With social media and the HUGE repository of robot designs, robot reveals and match play videos, the need for testing specifics will decrease. I already see all of this happening in our district, and suspect that this will continue to grow.
I’m with Andrew and others - the VP was a major advance that changed how easily our mechanisms could be iterated through in 2013 and onward.
Vendor proliferation of hex shaft -based COTS components was the next big leap IMO. It takes most of the design work of power transmission around a shaft, allowing us to focus on mechanism design rather than pouring effort into that particular detail.
In the last 3 years, the mechanical technical leap seems to been many more options for intake wheels when interacting with non-compliant game pieces.
Other than that, more DIY-based CNC’s & 3D printers have helped more than a few teams increase their capabilities.
I’ll echo what several others have already mentioned in that the biggest difference for the most amount of teams is made by available COTS items. The hex shaft system makes a subtle but monumental difference. I don’t even want to think about having to use keyways or interference fits or clamping hubs or tapers or set screws every time I want to transfer torque on a robot. I’d love to see hex shafts adopted by real-world industries, but I’m not holding my breath.
As for the next big technological advancement that will impact teams? I’m gonna go out on a limb here and say mid-scale advanced manufacturing services. Already we’re seeing things like the Wazer or other small scale CNC machines. While I don’t think these are viable for anyone just yet, within 5 to 10 years, affordable CNC manufacturing on an FRC-esque scale could be a thing. Also, mid-scale additive manufacturing is a possibility. A vast amount of teams already have small desktop 3D Printers. There’s lots of research and development being done all over the world with large scale additive manufacturing. ahem](http://www.e-ci.com/baam/) But to the best of my knowledge, there’s barely any work being done to create printers in between the extremes. So I think we’ll start seeing additive manufacturing machines much more on the FRC scale start coming from industry R&D, and then we’ll swoop in to reap the benefits.
No, I really want it to be allowed! Limit the fluid volume and give the field crew some bags of kitty litter. If you don’t put hydraulic items in the KOP then they will likely only be used by teams that go out of their way to make a system that works correctly.
