1690 Orbit demonstrates a masterclass in engineering synergy, proving that modularity and software-driven precision are the ultimate competitive advantages. Their approach turns complex mechanical challenges into a streamlined, high-performance reality.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
1690 Orbit | Behind the Bumpers | Galileo WinnersAdded:
Hey everybody, it's Tyler here at the first championships. We're checking in with the incredible 1690. It's Orbit World finalist last year and I just watched them play a match uh here uh in day two of the first championship. They looked incredible at that as well too.
Their BPS is so cool and of course we'll be focusing on how they're getting there and uh that impact inspiration they've been able to show the first community has been great throughout the season so far. You'll see all the different aspects of this robot. Some great software stuff. Orbit always brings the game when it comes to incredible software solutions. And we're going to hear a bit more about how their scouting program works as well, too. So, let's learn more about them coming up here on Behind the Bumpers.
Jonathan, let's dive into this robot here. Talk to me a little bit about your drive base and we'll work our way up through in the intake.
>> Okay. So, first of all, at the beginning of the season, we thought that this game was going to be super aggressive and that defense was going to be a huge have a huge impact on this game. So our strategic design for Kepler surrounded around a few main pillars which were to first of all shoot on the move have a very large capacity hopper be extremely durable and all this while being a transbot. So starting with our drive base um it's 29 1/2x 24 1/2 in with the wider side being our intake and and the bumper uh padding is connected directly to the hopper walls over here just like this so we can take them off and view the robot a lot easier. And the wall is this two-piece thing. We have this thin polycarb plate and this net to maximize fuel capacity because of its flexibility, which brings us to around 60, I think. Now, the most interesting thing about a drive base was actually the trans transition that we made mid-season. We originally designed manufactured a new custom swerve module during the offseason based on the MK5s incorporating our previous design custom wheel and wheel studs. But um after the robot was built while uh training with our defense bot, we saw how easily the robot could be pushed. Now we thought that this would happen ahead of the season, but after playing with some defense, we saw how crit credible it was going to be for this game. So very quickly, we made the decision to swap to the MK5s uh because of their very large, very wide, sorry, spike treads that jag themselves into the carpet compared to our thin nitro treads. Luckily, SDS were as quick as us and quickly sent us uh as a sponsor new modules and we swapped them out and it was pretty easy except for one major thing that we had to do.
We actually um integrated top plates of the modules which also gave us the opportunity to make them even more robust and strong for this game's impacts and jumps.
Um and they they've been spectacular since then. So, so far here at Championships, you've been able to play a few matches with them and you have been happy with them overall.
>> Yeah, completely.
>> Awesome. I'm glad to hear that. Let's talk more about the uh intake of this robot. What goes into it?
>> Yeah. So, let me open it up.
Now, when planning this intake, we had a few main key factors in mind, which were to make it extremely durable, obviously.
Um, to make it collapsible on impact, to make it easily swappable because in in case it breaks, because it takes time to break, and all of this while being as wide and efficient as possible. So, it features these two silicone coated rollers powered by an X60 over here. And at the bottom, we have this static kicker roller. Now, the kicker roller used to be a ramp, a dynamic ramp that open and closed along with the intake, but we found it to be a problem when opening our intake above fuel. So, we just made it change it so it opens once at the beginning of the match and then stays there. And since we don't have any other systems that extend beyond our frame perimeter, it wasn't a problem with the rules. Also, in contrast to many other teams in the world, we don't have a slap down intake because they don't collapse very well on impact. And like I said, there was one of our points for while planning the intake. Instead, we have this rack and beer mechanism.
You can view it over here a lot better.
We designed manufactured both these racks uh the polycarb racks and the aluminum and brass gears underneath at our shop. And so far, it's been working great. Now, for robustness, we have this full 25x 50 mm extrusion at the front to protect it against impacts. And also for each side we have this sort of sandwich made of two 6 mm polycarb plates and another set of two aluminum 1 and 12 mm aluminum plates. So this is for durability and flexibility and it's so far it's pretty strong. And lastly to make it easily swappable, this whole U-shape can actually just be lifted up after removing three screws on each side over here. And it's also been pretty easy to swap them and it's been working great. Before we get to our next segment, we'd like to thank the following. FRCT's managed store program is an easy and profitable way to fundra for your team or offseason event. FRCTS will handle everything from production, packaging, bulk or drop shipping, payment collection, and accounting.
Teams can set their own prices and FRCTS will take care of the rest. Fill out a quote request or get more information when you schedule a call back at frT.com.
here uh uh competed the first championships we know of for orbit that these were kind of your first official matches you had an opportunity to play with. What were maybe some of the biggest surprises playing some of your first matches here?
>> So we we were very paranoid uh coming to this competition. We knew that things were going to break for the first time and second time and like we we were expecting the unexpected. So, like especially for the intake, we brought a million spares. And even though we did try to get as many spares as we could, we we still found some trouble with other systems and spares for them. But we try to make the best of everything and get some help from pits around us and maybe try to make parts on our own at over here. Um, but yeah, we do the best we can even though it's hard competing for the first time.
>> Yeah, definitely a difficult view, but as of today when we're filming this, she had a great first match. So we hope that continues uh with you too. So >> yeah uh let's pass over to Yel talk more about uh some of the uh ball path journeys you go through here with your uh indexing system as we work uh kind of our way into the turret as well. So talk to me more about it.
>> Yeah. So this year our indexer is made out of three main components. Uh standing with a roller floor. We we have a a roller floor of 11 active carbon fiber silicon coated rollers uh powered by this Kraken X60 motor here uh that has a 2:1 speed reduction transmission uh for all all of the rollers. Uh for the roller to roller power transfer, we chose a 3 mm pitch GT2 belts instead of the traditional 5mm HDD for its higher efficiency in such small radi. Um two very important factors uh that we think has been making us h have this high BPS is uh firstly between each roller and hauler roller in this area here we have a very slight speed reduction transmission uh uh in order to minimize the the backed up uh jams that we have from the four balls with here into the three and a half balls with we have here. Uh the second feature is that these two rollers uh up and down are not silicon coated. They're actually slick uh to help uh reduce friction when the balls when the balls are traveling towards the shooter. Uh the second aspect of our indexer is our two vertical belts uh here and here. Uh these two belts are 5 mm pitch HTD and not silicon coated. Uh actually we we wanted to make them silicon coated like the WCP ones but we weren't been able to get ones as fast as we needed to uh so it's that just 30 mm wide HD belts powered by two independent Kraken X44 motors uh to give them a surface linear speed of 8 m/s and underneath our belts we have a 1 mm thick PT fishet to keep a consistent press of around 10 mm on the fuels whenever they're traveling here.
Uh next to our conveyor belts, we have this vertical kicker roller. Uh which is actually our purpose piece of our 2024 robots doppler shooter roller. Uh this roller is powered by this Kraken X44 motor here. Uh and it's crucial to mitigate jams in our indexer. The final part of our indexer is our active seash shooter ramp. Our active C shooter ramp uses three rows of two and a/4 in flexible wheels and two rows of 2-in stealth wheels. Uh that all powered by one Kraken X60 motor here to give us a a B speed of 8 m/ second and with actually zero spin coming to the turret because of this uh roller here opposing to all of the roll.
>> When when your reveal video first came out, I think the entire first community was absolutely blown away uh by Cena.
Did your team buddy have any difficulties as you were building your robot of getting the indexing working correctly?
>> Uh yeah, the first time we powered our robot, we didn't actually knew how to spin the vertical the vertical kicker roller and actually the speed of everything. After our pebbot, our practice bot was built. We had I think three or four iterations of speed ratios, HD belts, GT2 belts reduction uh uh everything because at the start it was really not working well and really disappointing but uh after a few iterations we couldn't get it up to as it is today.
>> I know watching uh your shooter I watched your last match here and it's just so impressive just to watch that stream come through. It's so fluid of a process but I know you know the amount of work that it takes to get there is just massive for that. So walk us through like some of the key aspects that have been so successful for orbit.
>> Yeah, thank you. So our main objective while designing the tour this year was making the overall diameter as small as we can so we could push it more into the corner and save more space for index bolts and the opera for a higher BPS and a larger fuel capacity. And so starting with our turret, uh it's driven by an X60 below uh through this gearbox which we made sure to keep low for a better CG. And it goes through those selfmanufacturers uh gears on top. And it all sits on an 8 in bearing. A cool detail about it is that uh the ball path is not is not centered on the bearing.
It's actually to offset it towards the back. So we could take the circle that's going into the outermost edges and push it more into the midpoint of the bearing. And that enabled us to uh push it more to the corner and save more space. Uh so for actual shooter uh we found in our prototypes that slight to no spin of the ball gave us the most uh consistent stream with the lowest energy loss. So, we could just save weight and space by connecting uh the back rollers to the main shooter wheel via those belts and gears. And that was made really compact and simple by our um custom manufacturer wheel here. And we also uh took the motors and put them on an SLS winch that holds them closer to the wheel so we could uh minimize our outward extension and enable us to lower the diameter which was really our main objective. And in the same idea, we also put the motors that drives the hood uh in here on the side of the plate where it does not extend the diameter and goes through those bevel gears which drive those hair muzzles on the back. So you can see and the uh one cool detail also is that uh the lower the lower roller is also sitting on the same shaft and that uh saves space and uh weight. Uh so another uh thing we're very proud of is our wiring uh for the motors and the autoam which we use an custom uh 3D printed energy chain which we actually called because it means fun in the roof cuz it's super fun to play with and uh yeah it's a cool thing we have and so uh we made sure to keep it really as compact and as narrow as possible. Um, and we a cool data about it is that we use 14 gauge wires for the motors instead of a standard 12 or 10. And that enabled us to really make the links super small, super compact. And we also raised the wall shooter up so it has a sort of a neck. And then when it has this uh smaller diameter, the chain wraps around in a small diameter and that uh enables it more space to move and slide. And that enables uh higher degrees of rotation which we eventually made it to 640 which is more than we think most teams do. And you can see we can really do 360 plus uh for each uh side. And uh yeah so I think we're going to do a quick demo if we >> let's take a look.
>> Yeah.
I think what surprised me on this as we're doing is just how quiet this whole whole thing is. Some of the you talk to are just like incredibly loud on it. Uh overall it's just so smooth to watch this uh go on there. Were there any surprises to you in terms of like how well your how well your charge been working or anything like that? Like was it working right away when you first uh got all this together or did you have any difficulties to try to get through?
Uh actually for a BBOT, which was the first robot we made to make sure that everything was well, we had a few problems uh with some uh 3D prints that weren't uh quite as smooth or or as uh much as we wanted to to be. Um so we in our second iteration really made sure that all of the manufacturing was super smooth. Uh we used a spurs that we have for the um for the shooter wheel. So it manufactured the super super smooth and will and we and we really make sure to keep everything as compact as simple and yeah >> speak about compact and simple uh you all you guys do you have the climber on your robot. So I'd love to just hear a little bit more about that and uh kind of like are you planning implementing using it uh here at Champs or have you done it so far? Uh yeah. Uh a fun fact is that we actually started the season with an L3 climb that proved to be to be too heavy and too complex to actually work on a real game. So mid-season after the bot was built, uh we designed, manufactured and assembled on the robot our L1 climbber. Uh this L1 climber is powered by a 2 mm pitch threaded rod that that's actually hidden underneath here. See if you can help me open the climb. Uh yeah. So that's cool because it made us h be able to to not have any speed reduction transmission on the motor at all. Only the threaded rod that has all the red all the reduction to make us climb uh very well. Uh another word about about compactability and efficiency. Our roller hood lifts up uh like a car motor hood to reveal up all of our electronics and our intake adapting shaft which opens the intake.
Before we get to our next segment, we'd like to thank the following. Thank Andyark as your one-stop shop for all your team needs. Antimark's Launcher in a Box is a configurable solution that so many teams are finding success with for the rebuilt season. Also available at annark.com, the AMPD Antimark power distribution board that provides 24 40 amp independently fuse output channels for safe, reliable power distribution.
That's annark.com.
>> Very impressive from the mechanical side, but we know here on orbit there's so much to go in, you know, in terms of software for things. uh you know I think we spent an hour just talking about software quite easily but what are some of the key highlights that you want to cover that have made or so successful?
>> Uh yeah totally. Um so pretty early on in the season we realized that because of the amount of uh April tags on the hub um we want a an a camera on the turret. So what we went with was this article over here connected to an orange pie running photo vision in the back over here. And for uh blind spots where the TH is wrapping around basically uh or uh the camera disconnects for some reason which we had a few uh times that happened to us. We have two Arat for in the back over here.
>> So um for the shoot on the move which is I think what we want to talk about um we basically this year made a simulation uh similar to our 2022 robot. Um the simulation uh we made a few improvements to it. One of the improvements we made was uh this uh GUI uh which allows us to make uh trajectories much much easier because there's a few aspects like delivery this year which went in our 2022 robot. Um this simulation works by basically uh simulating the trajectory using a few forces like gravity, air resistance and uh magnus spec to simulate the ball trajectories for a few distances at robot velocities. Uh using different cost functions for passing or scoring um we uh find the best trajectory for each distance velocity and from that we generate a few polinomials that the robot itself uh can compute. So uh the big idea behind this is to do the heavy lifting off robot and to save uh computing power uh on the robot ear rubber ear. Um so yeah uh another thing we did was we this year factored in the robots acceleration uh to our simulation which improves uh shoot on the move drastically against defense which is a big aspect this year.
Um so that's one of the things we did.
would you say? So, one of the things I would say is when your reveal video came out, I think it really kind of set the bar that showed the move was going to be just absolutely huge in this game. Uh, would you say one of the key optimizations getting ready for roles was trying to figure out how to play against uh heavy defense on something like that?
>> Yeah. Yeah, totally. Um, this year we saw earlier on again when we played against our defense bot um that defense was going to be a big aspect uh this year. Uh, so we did a few things for that. Um, one of them, one of the things we did was, uh, tilt detection and bump detection. So, we can shoot while tilting on the bump if we need to. Uh, and it helps us also shoot early in autonomous uh, and uh, better geometry against defense when you detect the til uh, tilting. Um, another thing we did was defense mode. So, if we if you want uh against defense, we put the swerve in an X configuration uh, so it's harder to push around. It kind of just plants you down in the field.
>> Plants down in the field totally. Uh it cancels you on the move. Um but allows you to empty the hopper much faster and accurately against defense.
>> One of the things I want to ask is in previous years when we've talked to your team, there's been a lot of uh areas where the robot has essentially been autonomous during driver control and that sort of thing like more that there set points that sort of thing. Are you doing anything like that uh for this game or is it very manually driven in order to complete your cycles? So this year what we went with was uh a uh uh with our custom uh path planner.
Um so less uh I guess autom automations uh this year uh than previous than last year uh for example. Um but you can see here um our autonomous is pretty easy to make using bizier curves and kinematics uh in our custom uh path application.
>> Anything else software wise you want to highlight on this awesome robot? Uh yeah, in Oton uh another thing we added was a collision detection. So we can shorten our path if we uh take a collision in the initial rush to the neutral zone. Um which helps uh not fighting against another robot for fuel and some more fuel to our side uh if need be. Overall very cool to to just dive more into this. Like I said uh every time we see it's such an education process uh that you guys are able to help provide to this community. I think one of the things that we haven't talked about with your team is how you handle scouting. So, let's fast over to me to talk more about how your scouting operation works and uh let's dive a little bit more into how it impacts orbit.
>> Okay. So, I'll start a bit of background. So, why we created this system after the 2024 season? Uh we saw that scouters missed a lot of matches maybe missed things in the matches since it's in real time itself and overall scouters uh aren't very accurate and we have a hard time to see how to solve this we created a on demand system where we do the scouting after uh we have captured the the a video of the match.
Erh uh this works on a tablet that we have uh created a program for which is connected to a server and this is a demonstration of the app itself. So what they have they go onto the system they have the video of the match itself. Uh let's say we are doing scouting on them on uh on 1678. We we will see the video itself and when they start shooting we'll press start shooting capture the first time the times of the we'll choose the where they're on the pitch on the field and when they stop shooting uh we'll stop we'll press stop. What this system lets us is go back and forth uh go f go in fast motion slow motion to capture everything and have more accurate things. In addition, we have another process or scouting system which is our shooting scouters which go over each of each clip of shooting and how you will see here here's the clip where I press where I press the robot you'll see only the shooting things and after this they will choose out of our our five main uh ways to count balls AI static AI dynamic magazine based, BPS, and counting manually. Uh, I'll go over one of the AI modes out of the two modes. AI static is a AI we created that runs on a server where we choose a zone and every ball that goes through it, it counts it. We also have detections that it doesn't it doesn't detect balls on the ground, a bunch of stuff like that. And we have another AI which we only use the robot that we press and it uh tracks the robot and counts the balls. I'll show you a demonstration of the output of the AI. You'll see we selected the robot. It tracks it and every ball that goes up detects it and counts it in the counter. You'll see we it every red ball is is us saying that we are not detecting the ball since it's on the ground and the speed of the ball isn't as shooting since the balls that go up have a speed up. So that's how we count the balls itself.
Uh as you see here in addition we don't actually count the amount of scored balls. We only count the amount of shot balls since we we created an analysis equation where we know that if a robot shoots more, it's probably uh scores more. So, we divide the total score from the blue alliance for the alliance and normalize each robot uh scored balls based on how much they shot. And that's how we generally >> You guys just made some of the coolest instant replay software ever. I think with like I think about like whenever we do a sports test like this would be incredible to have uh for something like that too. Incredibly impressive. I love the ability to make key frames as well uh to be able to kind of just go back and reference as well. Really cool I mean really appreciate the time tell us about that and of course uh for all orbit. Thank you so much for uh showcasing this incredible uh set of machines. Obviously an extremely complete team and we can't wait to see here at the first championships how you do as well. But always great to check with you. Thanks for being such a great inspiration to the first community and best of luck the rest of the way. Thanks for watching. Don't forget to like, subscribe, and click the bell to stay uptodate on future fun videos. Think Antimark as your one-stop shop for all your team needs. Annie's Launcher in a Box is a configurable solution that so many teams are finding success with for the rebuilt season. Also available at annark.com, the AMPD Antimark power distribution board that provides 240 amp independently fuse output channels for safe, reliable power distribution.
That's annark.com.
FRCT's managed store program is an easy and profitable way to fundra for your team or offseason event. F FRCTS will handle everything from production, packaging, bulk or drop shipping, payment collection, and accounting.
Teams can set their own prices and FRCTS will take care of the rest. Fill out a quote request or get more information when you schedule a call back at frT.com.
Related Videos
Beyond Robotics | European Rover Challenge 2026
beyondrobotics
189 views•2026-06-01
Beatbot Sora70: JetPulse Technology and AI obstacle avoidance and navigation!
DroidModderX
26K views•2026-06-02
Tesla FSD 14.3.3 Hits Phoenix Streets - FIRST LOOK
anthonystesla
114 views•2026-05-29
Elon Musk Just Revealed Fremont Line for Optimus Gen 3 Mass Production
TheAINexusOfficial
180 views•2026-05-30
人機一体「零式人機 ver.2」 子ども企画【おもしろ発見!モビリティー】 #乗り物 #automobile #robot #shorts
KyodoNews
1K views•2026-05-28
China’s New Luna AI Robot Looks Shockingly Human...
NextGenHumanoids
850 views•2026-05-28
Reachy Mini: the $300 open source robot you can actually hack — Andres Marafioti, Hugging Face
aiDotEngineer
662 views•2026-05-29
柔軟指×AI画像処理食品の仕分け作業システム!#柔軟指 #ロボット #自動化 #製造業をもっと盛り上げたい
KiQ_Robotics_Corp.
113 views•2026-05-28











