As we near the beginning of the next season, I thought it might be good to review the year by going over what people think were the 3 most trends-setting features of 2017, and whether they are good or bad. This could be anything robot related that a team(s) did in the past year.
I thought this could be a great way to find out about some things other teams did while also starting some conversation about the future of robots in FRC.
I’ll start with my 3 choices:
(1) 3D Printed Parts - 125
They definitely aren’t the first to use them on a robot, but I would say they really opened up the possibilities of what can be done with printed parts mechanically, as well as how they can be utilized for prototyping. They definitely were the most competitive robot using advanced 3D printed parts that I’ve seen to date. I’m excited to see how much 3D printing is utilized and impacts designs in the coming years.
(2) Multi-wheel and Posi-Traction Swerve Drive - 2767
They spent years iterating a great swerve drive design, and included advanced features like the posi-traction system they used. I think this could be both a positive and negative trend, as it could raise the bar for what performance competitive teams get out their drive train, while it could also convince more teams to do swerve too quickly due to their relatively simple design that should probably spend more development time on it. I’m also curious to see more how much the posi-traction system truly makes a difference. I don’t think there’s a doubt that it does improve drive accuracy to some degree, but I wonder by how much and if it’s worth implementing. More years and more teams doing it will shed more light on this.
(3) Mecanum Drive - 1986
Obviously mecanum drives have not only been around for a long time, but have been pretty popular overall for a while now. I don’t think the success 1986 had with them will necessarily make them used even more often then they have been, but I think it will mean more teams will continue to use them who might have otherwise switched to a more standard WCD, etc. There are many out there including myself who aren’t huge fans of mecanum, so I’m not sure I’m a huge fan of this “trend”. Many people won’t think through the specific reasons 1986 used and had success with a mecanum drive last season.
I’m curious to hear what others think are the biggest current trends in FRC, and why they are good or bad for the sport.
It seems to me like the biggest trendsetting features can be seen on 3310’s bot. Minicims on drivetrains seem to be grabbing the attention of most teams (1836 and many others will likely run that next year) as well as all omni drivetrains. By no means was 3310 the first or only team to do either of these, but they are a very successful team and a lot of others are taking inspiration from these design features that could be implemented in 2018 and beyond.
Of course, it’s difficult to know what innovations will be trendsetting until there have been a few years to set a trend.
That said, non-CIM drive trains have to be at the top of the list. Enough teams did well enough with them that we can expect increased adoption of non-CIM 200+W motors in drive trains in coming years. While teams have used mini-CIMs in drives in the past, it was usually because of a desire for decreased weight or to reduce peak current draw, where this year we saw a number of 3 mini-CIMs per side drive trains - configurations designed to increase performance.
There was also a remarkable increase in the use of compliant intake wheels, due to a combination of a thoroughly irregular game piece, a rather stiff game piece, and a great increase in availability and reduction of price point. As with mini-CIM and low thermal mass motors in drives, we’ve seen these before (most notably in 2013 Ultimate Ascent), but this year they were common enough that they may prove to be the beginning of a trend. Of course, it depends on the GDC giving us irregular or stiff game pieces; if we get compressible spheres the next few years, this trend may never develop.
For FRC-wide trends, most of the other things appear to be continuations of existing trends (e.g. greater use of omni wheels) or are specific to the tasks of STEAMworks, so for my third choice, I’m going to cite a trend that has been going a while, but which 3946 really joined this year - Versaframe. We did use VF in 2016 Stronghold, but only for our launcher arm. 2017 was the first year where versaframe was our preferred structural material above the (KoP) drive chassis. It formed the support of our gear slide and control boards, stands and supports for our gear doors, and the fairlead for our climber. It really helped us make our bagged robot and our practice robot match each other better than ever before.
Other things which may prove trendsetting for 3946 is the increased use of polycarbonate (we’ve finally figured out at least somewhat how to work it) and corrugated plastic (this stuff is incredibly cheap, quite useful, has a low CoF, and is really easy to work with).
I feel that 492 was a better example of a proper use of a mecanum drive this year compared to 1986. It may of not of been as flashy or well known as 1986, but they managed a a high performing bot that took them to Galileo finals and was one of the fastest and eloquently driven bots not only from PNW but of almost any in 2017 and of any mecanum bot in general. Teams should be looking to how theirs was designed and driven if they want to peruse mecanum imo.
This was the first shooting task where increasing launch velocity wasn’t a viable option for improving shooting percentage (even in 2012, shooting fairly hard with as much backspin as possible was generally the easiest way to improve accuracy). As a result, top fuel teams learned a lot more about other ways to improve accuracy that will benefit everyone in future shooting games. By the end of the season, most of the top fuel teams implemented the following ideas:
Fairly low compression applied in the shooter. Compressing the game piece leads to energy waste and inconsistency because two game pieces (or the same game piece in two orientations) may not compress and recoil similarly. In many cases this was implemented using a foam backing on the hood of a flywheel. I do not recall this being a very popular approach in previous shooting games; previously teams have relied on compressing the gamepiece itself rather than part of their robot.
Imparting impulse to the ball over long distances/time periods. Many teams added “accelerator” wheels to their feeding path to ensure the game pieces entered the shooter much closer to the desired exit velocity than in previous games. Others (1986) used very large shooter wheels with long contact arcs. The effect is the same. This is necessitated by (1); low final-stage compression means more slip unless the ball enters with a similar surface velocity. There is some precedent for this from 2013 frisbee shooters, but in 2017 many teams found that just adding a single accelerator wheel was sufficient to dramatically improve accuracy.
Using Talon SRXs for shooter control (and for everything else on the robot for that matter). I know of exactly one elite fuel team (971) [EDIT: 2…1986!] who did not use the Talon SRX on their shooter. (I’m sure there are more that I don’t know of, but I’m just saying - Talons rock).
All of this post is great advice, just wanted to expand on this. Compliant backing surfaces on shooters seem to be more necessary in games with rigid or somewhat rigid game pieces, like in 2013 or 2017. In other games, teams got away with compressing the game piece, because the game piece could more reliably serve as the compliant member in the system.
I do wonder if in games where foam / compliant pieces are shot, if a compliant backing that’s squishier than the foam would increase consistency, but my teams haven’t had any success trying that approach the one or two times we had tested it (2012, 2016). Might be worth looking into.
We used them to help all the vendors sell more silly wheels. Been meaning to reach out for a sponsorship for all that great PR. We’d be happy to demo any other underselling items in the warehouse and make them cool again. /s
A big change this year was the amount of “dynamic” turrets. In the past you’d see even top tier teams struggling to use their turret in more than a couple of positions. Most of the time they’d have just a couple of known positions to make shots from with a known turret angle and a known speed set point.
This is the first time we’ve really seen teams be able to consistently make shots into a small goal from dynamic positions on the field using vision systems. A lot of this has to do with what Jared Russel mentioned.
The sheer amount of Lexan on robots this year, both on hoppers and manipulators is something I think more teams will see the value of, pick up on, and use. Its cheaper and easier to work with than metal for some teams, and its much more accommodating to impact outside the frame perimeter.
Just wanted to mention how awesome this post is and many others in this thread are, it’s insights like this that help us all learn and build better robots. It seems like ever year I learn something awesome from 254 and other great posters on CD, I’d really love to see CD figure out a way of highlighting the great content like this.
The biggest learning for 610 this year was the use of break-beam sensors to auto release gears once they were scored with signal lights once scored. Adding this made a dramatic difference to our cycle times from our first district to our second. We’ve been known for optimizing cycle times pretty much every year, but I’d say this was the biggest single improvement we’ve made to a robot during the season. Seeing 1241 run with a similar setup during offseason events really showed how big of a difference it can make.
With regards to 775pro drives, the fact is that with the coming of the COTS options for gearboxes there are going to be teams that use them without knowing the proper precautions to take.
775pro drive is still very finnicky since any reasonable amount of stalling is going to burn out your drive motors unless you have voltage ramping/current limiting, which is something that these teams probably won’t do.
If done right, however, 775pro drives can package a lot of power in a very small footprint.
How many teams this year were truly held back by the amount of space their drivetrain took up or their raw pushing power? Look at the single-speed, 4-cim drivetrains used this year by 971 and 1678. I’m convinced that for the vast majority of teams, the limiting factor in their performance is mechanism design and driver skill.