My team is looking into an off-season project and a mechanism we have never built is a turret shooter however we use swerve and there is debate as to how useful a turret is when we have swerve. What are your thoughts is swerve going to phase out turrets? Are they still valuable should we continue to make a turret or would there be other mechanisms that would be a more valuable off-season project?
Here is a list of mechanisms we have already done:
Turret VS No Turret is probably going to just be a matter of opinion.
IMO, keeps it simpler to not have one, even with a tank drive style game. I go back to my favorite team…1678, who has had some Einstein appearance level years without Turrets. Less to control, no service loops to figure out etc.
But I’m sure many would argue its worth it. It probably objectively makes you a little less defense prone and might introduce some enhanced shooting on the move capabilities.
It depends significantly on the game and your design.
The nice thing about a turret is that it’s independent of the orientation of the chassis so in a game like Infinite Recharge you can have accurate shots while being bumped by another team. But for games like Steamworks it likely doesn’t matter because you could hug the wall and no team can bump you out of alignment
In Rapid React, some teams utilized swerve with a turret and some opted out of the turret. With the turret it’s easier to shoot on the move (my favorite example is 1706 in the later half of the season). You can still be effective without the turret though, as 1678 showed. Both were swerve, and both were world-class robots, just differences in design.
Turret could be useful in high levels of defence (see 2019 2020 and some of 2022) and it could be useful for some games like 2023 (see 3015)
It’s all skill and opinion based
But as an off season project it seems like fun and can teach you alot!
In a shooting game a turret allows shoot on the move. Rotation is quick enough you aren’t saving a lot of time using one if you aren’t shooting on the move.
Keep in mind, a 10lb turret is significantly easier and more precise to rotate than a 120lb robot, considering weight, carpet friction, etc. That’s what made shooting on the move so much easier with a turret; I tried it with our non-turret robot and even with a hypertuned PID loop it was still too jerky and inconsistent to work more than 50% of the time.
Well… Technically you can shoot on the move with swerve (we did it with our B bot at Roboteer Rumble), but its not nearly as good.
What we have found is a little more precision/speed with the turret because you are only rotating 10lbs to aim instead of rotating 125lbs to aim. It also helped when getting pinned against the wall or against other robots.
The turret itself is way more bang for your buck than you can get with just a swerve.
You can have a turret locked onto a target while your robot gets bashed around or pinned by other robots. You can’t do this with just a swerve.
At least in my experience with a turrets, they are significantly cheaper than a swerve. So if you are limited in resources, I would definitely suggest the turret.
In terms of robot performance assuming that the robots work 100% as intended the order that I would rank styles of robots as follows in defending order; Swerve+Turret, WCD+Turret, Hybrid*+Turret, Mecanum+Turret, Swerve+Fixed, WCD+Fixed, Hybrid*+Fixed, and lastly Mecanum+Fixed.
As a general rule of thumb, I tend to operate under the assumption of if there can be defense played on you, there WILL be defense played on you. So you can take that for what it’s worth.
A turreted robot will struggle just as much with being hit and shoved around while shooting as a non turreted robot.
This is a trap. Turrets are definitely not easy to optimize and if you do not have the resources to do swerve I would not run a turret in the first place (from a competitive standpoint. If you just want to build something cool then by all means go for it).
A well executed swerve turret is 1, I agree. However swerve+fixed will outperform a WCD+turret 9/10 times as shown by teams in 2022 such as 1619, 1577, 1706, and 1678. Mecanum+turret outperforming a swerve+fixed turret is kind of a joke IMO.
I would never suggest any team without the absolute top level of resources pursue a turret from a spending of resources standpoint. There are so many better ways to spend your resources in an FRCFIRST Robotics Competition season.
i definitely think having a turret on top of swerve helps you get faster cycles even if you’re not shooting on the move - it seemed like teams with turrets would save 2-3 seconds per cycle last year compared to us. if only ours wasn’t a non-functional decoration last year
COTS turrets require (if I’m not mistaken) the team to come up with their own feedback/control system and code. They also require cable management… and a means to seamlessly hand off a game piece from a indexer/hopper to the turret. They then have a moderately maneuverable tank robot with an aim-able launcher.
Worth attempting and putting in the hours in a world without COTS swerve (with highly reliable components and downloadable code).
I don’t think turrets are going away. But doing swerve means the whole robot is more maneuverable, AND you can aim your launcher as well. Double win… for way more money.
Turrets are a big mechanical challenge. I suggest starting with some carboard models and an Aluminum turntable bearing so you can play with cable tendering. You MUST have a good method for cables to succeed on a turret.
Remember, your turret will (almost certainly) not be 360. Narrowing the angle massively simplified the cable tendering.
A tip: cheap Aluminum turntable bearings are awesome, BUT not drilled very accurately. Weighs a bit more, but radically simpler to integrate and get spinning.
Turrets themselves aren’t the hardest to make, but they force you to make suboptimal decisions with the rest of the robot. Turrets are never the competitive choice unless you have the resources to solve the other problems and squeeze all the performance you can out.
Strategically, I definitely believe that turrets are incredibly valuable, and with adequate resources, would succeed loads more than a non-turreted bot, no matter what drive train.
Of course, in games like Rapid React and Infinite Recharge where there was little-to-no terrain, swerve drives thrived. But what I saw in Rapid React that negatively affected teams the most was not having a turret, no matter what drive train they had. Defending against non-turret bots was much simpler than turret bots: simply Pin for 4 seconds then clear 6 feet, repeat. Pinning was virtually useless against most turret bots, as they could still align their shots towards the hub. As long as they are accurate from the distance where they are pinned, then the strategies that defend them effectively becomes complex.
I don’t think swerve will phase out turrets, as the combination of both proved to be very effective two shooter games in a row. Maybe the next shooter game would be different, but I think the answer for that one would still be turret + swerve.
I’m not saying that turreted bots are impervious to defense, but this just isn’t true. A bot without a turret can get hit on a corner and misaligned from the target, and will 100% miss the shot. A turret bot can get hit on the corner and misaligned, but the turret can compensate to remain aimed at the target and still have a chance of making the shot. It all depends on how much work you put into tuning the turret control loop.
Let’s just say the attitude 254 had toward swerve before doing a complete 180 and naming their 2022 robot “Sideways” is the attitude I have toward turrets.
Show me a turret robot and I’ll show you a non-turret robot that performed just as well. This even applies in 2017. The arguments about turrets making software development / PID “easier” are valid arguments. I’m just of the mindset that turrets are not worth their mechanical / electrical / sensor complexity. I get that others may not agree, and I love seeing the diversity of designs every year.
It would take quite the game for our team to make a turret, especially since we’ve got all this swerve experience.
None of the 8 world champions in 2017 had a turret, and only one world finalist had a turret.
I’d argue 2022 was the most turret-friendly game ever, and yet teams like 1678, 2910, 1577, 1619, 6328, 5940, 604 all competed without them as captains/first picks on Einstein just fine
Of all the mechanical/software issues we had in 2022, the turret wasn’t on that list. There’s a bit of a learning curve on the software side, but just using an off-the-shelf ring bearing with a printed or routed sprocket/gear is pretty straightforward. We didn’t use any sensors except the internal encoder on the falcon and it performed fine. I do think when you’re faced with a game that has a single shooting goal like in 2022, even with a swerve drive, it’s probably worth it to invest in a turret.
This really depends on the game. If you can still get jostled around, such as in a game where you are scoring from a lot of unprotected zones, you may want a turret in order to provide minor adjustments to your alignment, such that you can compensate for any defense played against you, even with your swerve modules in an x configuration. We may see the rise of limited range of motion turrets, compared to the full turrets that we have seen in years past since you no longer need the full range of motion 100% of the time, especially if you are only using it for defense compensation. (A limited range turret is most likely mechanically easier to implement than a full range of motion turret)
