Articles by tag: innovate

Articles by tag: innovate

    Swerve Drive Experiment

    Swerve Drive Experiment By Abhi

    Task: Consider a Swerve Drive base

    During the entire season of Relic Recovery, we saw many robots both in and outside our region that had a swerve drive. As Iron Reign, we never considered a swerve drive in the past but seeing all the robots, I wanted to see if it was maybe possible. One motivation was that I didn't like how slow mechanums were. Swerves generally use traction wheels and create a faster speed than usually can be found with mechanum. Also, it seemed as if swerve could provide the mobility neccessary that a mechanum drive provided. This is why I wanted to consider the possibility of a swerve drive and why I did more investigation.

    I first came across the PRINT swerve for FTC by team 9773. They had a very detailed explanation of all the parts and assebly tools. After reading into it more, I decided that the system they created wasn't the best. First, the final cost of the drive train was very expensive; we did not have a very high budget despite help from our sponsors. If this drive train didn't work for some reason after playing with it over the summer or if the chassis didn't make sense to use in Rover Ruckus, we would have almost no money for an alternate drive train since we wanted to presearve Kraken. Also, they parts used by 9773 invovled X-rail rather than extrusion rail from REV. This would cause problems in the future as we wold need to redesign the REVolution system for X-rail. In the end, I decided this was not worth it to pursue.

    After further investigation, I found a chassis by team 9048. The swerve they developed looked like a more feasible option. By using REV rail and many of the parts we had, I thought this would be a possible prototype for Iron Reign. Because they didn't have a parts list, we had the find the rough estimate of cost from the REV and Andymark websites. Upon further analysis, we realized that the cost, though cheaper than the chassis of 9773, would still be a considerable chunk of our budget. But I am still motivated to find a way to make this happen.

    Next Steps

    Possibly scavenge for parts in the house and Robodojo to make swerve modules.

    Swerve Drive Prototype

    Swerve Drive Prototype By Abhi and Christian

    Task: Build a Swerve Drive base

    During the discussion about swerve drive, Imperial robotics, our sister team, was also interested in the designs. Since we needed to conserve resources and prototype, I worked with Christian and another member of Imperial to prototype a drive train.

    Due to the limited resources. we decided to use Tetrix parts since we had an abundance of those. We decided to make the swerve such that a servo would turn a swerve module and the motors would be attached directly to the wheels. This system would be mounted to a square base. We decided to go ahead and make the base.

    Immediatly we noticed it was very feeble. The servos were working very hard to turn the heavy module and the motors had trouble staying aligned. Also, programming the train was also a challenge. After experimenting further, the base even broke. This was a moment of realization. Not only was swerve expensive and complicated, we also would need to replace a module really quickly at competition which needed more resources and an immaculate design. With all these considerations, I ultimately decided that swerve wasn't worth it to use as a drive chassis.

    Next Steps

    Wait until Rover Ruckus starts so that we can think of a new chassis.

    Position Tracking

    Position Tracking By Abhi

    Task: Design a way to track the robot's location

    Throughout the Relic Recovery season, we have had many issues with the autonomous being inaccurate simply because the scoring was dependent on perfectly aligning the robot on the balancing stone. This was prone to many issues as evidenced by numerous matches in which our autonomous failed. Thus far, we had relied on the encoders on the mecanum chassis to input distances and such. Though this worked to a significant degree, the bot was still prone to loss from drift and running into the glyph pit. We don't know if glyphs will be reused or not but we definitely needed a better tracking mechanism on the field to be more efficient.

    After some investigation online and discussing with other teams, I thought about a way to make a tracker. For the sake of testing, we built a small chassis with two perpendicular REV rails. Then, with the help of new trainees for Iron Reign, we attached two omni wheels on opposite sides of the chassis, as seen in the image above. To this, we added axle encoders to track the movement of the omni wheels.

    The reason the axles of these omnis was not dependent of any motors was because we wanted to avoid any error from the motors themselves. By making the omni wheels free spinning, no matter what the encoder reads on the robot, the omni wheels will always move whichever direction the robot is moving. Therefore, the omni wheels will generally give a more accurate reading of position.

    To test the concept, we attached the apparatus to ARGOS. With some upgrades to the ARGOS code by using the IMU and omni wheels, we added some basic trigonometry to the code to accurately track the position. The omni setup was relatively accurate and may be used for future projects and robots.

    Next Steps

    Now that we have a prototype to track position without using too many resources, we need to test it on an actual FTC chassis. Depending on whether or not there is terrain in Rover Ruckus, the use of this system will change. Until then, we can still experiment with this and develop a useful multipurpose sensor.

    Rover Ruckus Brainstorming & Initial Thoughts

    Rover Ruckus Brainstorming & Initial Thoughts By Ethan, Charlotte, Kenna, Evan, Abhi, Arjun, Karina, and Justin

    Task: Come up with ideas for the 2018-19 season

    So, today was the first meeting in the Rover Ruckus season! On top of that, we had our first round of new recruits (20!). So, it was an extremely hectic session, but we came up with a lot of new ideas.

    Building

    • A One-way Intake System
    • This suggestion uses a plastic flap to "trap" game elements inside it, similar to the lid of a soda cup. You can put marbles through the straw-hole, but you can't easily get them back out.
    • Crater Bracing
    • In the past, we've had center-of-balance issues with our robot. To counteract this, we plan to attach shaped braces to our robot such that it can hold on to the walls and not tip over.
    • Extendable Arm + Silicone Grip
    • This one is simple - a linear slide arm attached to a motor so that it can pick up game elements and rotate. We fear, however, that many teams will adopt this strategy, so we probably won't do it. One unique part of our design would be the silicone grips, so that the "claws" can firmly grasp the silver and gold.
    • Binder-ring Hanger
    • When we did Res-Q, we dropped our robot more times than we'd like to admit. To prevent that, we're designing an interlocking mechanism that the robot can use to hang. It'll have an indent and a corresponding recess that resists lateral force by nature of the indent, but can be opened easily.
    • Passive Intake
    • Inspired by a few FRC Stronghold intake systems, we designed a passive intake. Attached to a weak spring, it would have the ability to move over game elements before falling back down to capture them. The benefit of this design is that we wouldn't have to use an extra motor for intake, but we risk controlling more than two elements at the same time.
    • Mechanum
    • Mechanum is our Ol' Faithful. We've used it for the past three years, so we're loath to abandon it for this year. It's still a good idea for this year, but strafing isn't as important, and we may need to emphasize speed instead. Plus, we're not exactly sure how to get over the crater walls with Mechanum.
    • Tape Measure
    • In Res-Q, we used a tape-measure system to pull our robot up, and we believe that we could do the same again this year. One issue is that our tape measure system is ridiculously heavy (~5 lbs) and with the new weight limits, this may not be ideal.
    • Mining
    • We're currently thinking of a "mining mechanism" that can score two glyphs at a time extremely quickly in exchange for not being able to climb. It'll involve a conveyor belt and a set of linear slides such that the objects in the crater can automatically be transferred to either the low-scoring zone or the higher one.

    Journal

    This year, we may switch to weekly summaries instead of meeting logs so that our journal is more reasonable for judges to read. In particular, we were inspired by team Nonstandard Deviation, which has an amazing engineering journal that we recommend the readers to check out.

    Programming

    Luckily, this year seems to have a more-easily programmed autonomous. We're working on some autonomous diagrams that we'll release in the next couple weeks. Aside from that, we have such a developed codebase that we don't really need to update it any further.

    Next Steps

    We're going to prototype these ideas in the coming weeks and develop our thoughts more thoroughly.

    Testing Intakes

    Testing Intakes By Ethan, Evan, Aaron, and Freshmen as to be determined

    Task: Design a prototype intake system

    In our first practice, we brainstormed some intake and other robot ideas. To begin testing, we created a simple prototype of a one-way intake system. First, we attached two rubber bands to a length of wide PVC pipe. This worked pretty well, but the bands gave way a little too easily.

    For our next prototype, we attached a piece of cardboard with slits to a cup approximately the size of a cube or block. It operates similarly to a soda cup lid with a straw hole. An object can go in, but the corners of the hole spring back so that it can't escape.

    Next Steps

    We probably won't go with this design - we'd have issues seperating the different kinds of game elements, and it may be too slow to feasibly use. But, its a first step and we'll see what happens.

    Brainstorming Two - Enter the Void

    Brainstorming Two - Enter the Void By Evan, Abhi, and Janavi

    Task: Have a 2nd brainstorming session

    Last week, we had a lot of new recruits show up for the FTC kickoff. In fact, a bit too many. Luckily for us, we either scared them off or they realized that they'd like to move to FRC. So, today's session was a bit more managable, and we were able to break down into some new building tasks.

    Intake System 3 - TSA Bag Scanner

    If any of y'all have ever been on an airplane, you've gone through airport security. This part of our robot is inspired by the bag-scanning machine, more specifically the part at the end with the spinning tubes. The basic design would be like a section of that track that flips over the top of the robot into the crater to intake field elements.

    Intake System 4 - Big Clamp

    This one is self-explanatory. Its a clamp, that when forced over a block or a cube, picks it up. It's not that accurate, but it's a good practice idea.

    Lift 2 - Thruster

    We want to make lifting our robot easy, and we're thinking of a slightly different way to do it. For our new lift idea, we're installing a vertical linear slide that forces the robot upwards so that we can reach the lander.

    Next Steps

    We're working on building these prototypes, and will create blog posts in the future detailing them.

    Chassis Brainstorming

    Chassis Brainstorming By Ethan and Evan

    Task: Brainstorm chassis designs

    At the moment, we've used the same chassis base for three years, a basic mechanum base with large wheels. However, we don't really want to do the same this year. At the time, it was impressive, and not many teams used mechanum wheels, but now, its a little overdone. So, as the true hipsters of FIRST Tech Challenge, we want to move onto something new and fresh.

    Thus, we have BigWheel. We used this as a practice design, but we ended up really liking it. It starts off with two large rubber wheels, approx. eight inches in diameter, mounted at the back and sides of the robot. Then, we have two geared-up motors attached to the motors for extra torque and power. In the front, we have a single omniwheel that allows our robot to turn well.

    Proposed Additions

    First, we need to add an intake system. For this, we're considering a tension-loaded carwash that can spring out over the crater wall. It'll pull elements in and sort them through our intake using our seperator, which we will detail in a later post. Then, the robot will drive over to the lander and lift itself up. Since the main segment of the robot is based off of two wheels, we're attaching a telescoping slide that pushes off of the ground at the opposite end and pivots the front of the robot upwards. Then, the intake will launch upwards, depositing the elements in the launcher.

    Next Steps

    We need to create a proof-of-concept for this idea, and we'd like to create a 3D model before we go further.

    Autonomous Path Planning

    Autonomous Path Planning By Abhi

    Task: Map Autonomous paths

    Ahhhhhhh! Rover Ruckus has been around for a while now and it's time to figure out our autonomous plans! This year's autonomous is a lot more hectic than last year's since there is the detaching from the lander component which can take an unknown amount of time right now. That cuts down the potential speed for the rest of the auto but only time will tell.

    Until then, I can only dream of the potential autonomous paths our robot can take. First, we need to know some basic facts. One, the field is the exact same for both red and blue aliance, meaning I don't need to rewrite the code to act on the other side of the field. Second, we have to account for our alliance partner's autonomous if they have one and need to adapt to their path so we don't crash them. Third, we have to avoid the other alliance's bots to avoid penalties. There are no explicit boundaries this year for auto but if we somehow interrupt the opponent's auto we get heavily penalized. Now, with this in mind, lets look at these paths

    This path plan is the simplest of all autonomi. I assume that our alliance partner has an autonomous and our robot only takes care of half the functions. It starts with a simple detaching from the lander, sampling the proper mineral, deploying the team marker, and parking in the crater. The reason I chose the opposite crater instead of the one on our nearside was that it was shorter distance and less chance to mess with our alliance partner. The issue with this plan is that it may interfere with the opponent's autnomous but if we drive strategically hugging the wall, we shouldn't have issues.

    This path is also a "simple" path but is obviously complicated. The issue is that the team marker depot is not on the same side as the lander, forcing us to drive all the way down and back to park in the crater. I could also change this one to go to the opposite crater but that may interfere with our alliance partner's code.

    This is one of the autonomi that assumes our alliance partners don't have an autonomous and is built for multifunctionality. The time restriction makes this autonomous unlikely but it is stil nice to plan out a path for it.

    This is also one of the autonomi that assumes our alliance partners don't have an autonomous. This is the simpler one of the methods but still has the same restrictions

    Next Steps

    Although its great to think these paths will actually work out in the end, we might need to change them a lot. With potential collisions with alliance partners and opponents, we might need a drop down menu of sorts on the driver station that can let us put together a lot of different pieces so we can pick and choose the auto plan. Maybe we could even draw out the path in init. All this is only at the speculation stage right now.

    BigWheel Chassis

    BigWheel Chassis By Evan

    Task: Work on a possible chassis

    We have our first qualifier coming up real soon, and to get ready we decided to get a robot into gameworthy shape. So far we have a few chassis from our work over the summer, so I decided to start on a neat little competition bot out of the one we called Bigwheel, named for its two big drive wheels in the back. The idea for this robot is that it has a collection system that extends into the crater, and folds up on top of the robot. It reaches in with the collection arm, and grabs the blocks/glyphs, drives backwards and flips vertically using the drive wheels as a point of rotation. Here’s a basic sketch of what that looks like.

    The way this will be achieved is with a spring loaded lever connected to the omni wheel that makes up the holy trinity of wheels. So far I have pieced together the arm that reaches into the pit, which is powered by two NeverRest 60s and geared in a two to one ratio to significantly increase the torque. Between the two arm I plan for a horizontal beater bar to intake blocks and a slide attached to a servo to separate blocks and balls based on their size. The idea is to have a way of sorting based off of the physical shape rather than by digital sensing means. The more that can be done purely off the shape of the elements, the better.

    Next Steps

    So far, not much has gone into materializing the lever since this was made while the rest of the team was at an outreach event. Next week, the team will have to make some serious progress since there will be more hands to build. My hope is that the lever will come about soon, even if in its most infant stage, and that some semblance of a functioning robot can be game tested in the next few weeks, just in time for a scrimmage and potentially an early qualifier.

    Intake Sorter

    Intake Sorter By Abhi

    Task: Design a sorter for the balls and blocks

    We've been brainstorming ways to sort the gold and silver pieces, and here's our first one. It's a little bulky, but it's a start.

    The way this works is that when this piece is mounted and both blocks and balls are run over it, the balls run down the top and don't fall in the collector, but the blocks fall in the holes. We modelled this design in PTC Creo, then printed it in ABS.

    Next Steps

    This is a little too bulky, so we're going to have to find a smaller and simpler way to sort game pieces, but it's a start. In the future, we're going to minimize this and probably move to a smaller sorting mechanism.

    BigWheel+

    BigWheel+ By Evan

    Task: Continue work on BigWheel

    Since the last time we have discussed Bigwheel, the robot has gone through a few major changes. First and foremost, it now has a flipper arm. Since the robot itself is the lift mechanism, we had to put a lot of work into the design of the flipper. Right now it is a 10 inch REV rail attached to two of the largest REV gears you can buy, with a custom 3D printed mount housing an omni wheel pair. Right now it’s turned by two pairs of the smallest REV gears you can buy creating a 3:25 gear ratio. It’s a bit odd, but it increases torque which is what we’re going for, the philosophy being that it’s better to be smooth than to be quick, as a quick robot is weaker and less controllable. On top of the flipper, we’ve added extra supports on the arm mounts, as when we went to the Hendricks scrimmage, we found that the two sides were out of alignment, and one was bending more forward than the other, making the arm bend unevenly to one side and throwing the whole robot out of alignment. The next step is to strengthen the arm itself, as the two sides have a tendency to want to do their own things, mainly the side with the intake motor mounted to it. Since the supports have been put in though, Bigwheel has been functioning much better, and the arm no longer flops to one side. General wire management has also taken place, as previously the wires went every which way and created an infuriating mess that had the potential to get caught in the gears. Personally I’ve noticed that we’re a tad out of touch with gears right now, none of our other robots have sported them to much degree. We’ve been really in touch with sprockets though, since they’ve featured heavily in our designs the past two years. That’s just an observation, and since my first awareness of this, I’ve seen our gear savvy grow.

    Next Steps

    Bigwheel was built on a bit of a shabby base, mostly being made of a piece of polycarb and some aluminum bars, and not giving much in terms of change. We’ve cut here and there, drilled a few holes, unattached and re-attached a couple of things, but in all it’s a very stiff robot, and doesn’t lend itself to fluidity of design. That’s why we plan on making a second version of this base, hopefully with thinner polycarb and more secure sides that have been welded together but can be removed more easily. The exact design is still being modeled, but we have a direction to jump off from, and I believe we can make that leap to a better robot.

    Mini Mechanum Chassis

    Mini Mechanum Chassis By Janavi and Justin

    Task:

    To everyone's suprise, Garchomp didn't pan out. But, practice makes perfect, and that’s what we are going to do: create new chassis until we find one that works!

    When looking for inspiration for a new chassis, we found the mini mechanum wheels that we had been fiddling with last season but eventually scrapped in favor of the larger wheels that can be seen on Kraken.

    In the past, Iron Reign has been infamous for creating robots that barely fit within the sizing cube by a fraction of an inch, so we decided to take a different approach this time. We built a low-laying 6" x 6" robot to counter our older approaches. However, this new design meant that many of the pieces we had created in previous competitions would be unusable due to the changes in size and structure. For example, the way we affixed the wheels to the frame didn't work, as the axles were too big and the wheels were too small to use with our former wheel mounts. As well, these new mini mechanum wheels came with almost frictionless bearings, but the axle that fit within these bearings was hexagonal, unlike the round ones we usually use.

    Justin first designed the axle plate below to solve this, but discovered that it raised the robot off the ground quite a bit, something we wanted to avoid, and it was very flimsy since it was so far away from the frame. We discussed these issues with our coach and brainstormed various methods we could use to attach the axle the frame in a more secure way; he suggested that we use a pillow block design that would have one side of the axel touching the frame. That way we would save space, while also having a lower-lying robot. This design worked out beautifully, leading to the design we are currently using.

    The axles and wheels aren’t the only new thing about our robot: we're using motors with a gear ratio of 20 rather than 40s and 60s. This is another reason that we wanted to create such a minute robot. The gear ratio combined with the size will make this robot a speed demon on the field and allows us to dart between the minerals and the depositing location quickly.

    Next Steps

    In the upcoming weeks we will continue to tinker with this chassis design by adding a linear side and our gathering mechanism, and hopefully, we will be able to demonstrate it at the scrimmage next week.

    Intake Update

    Intake Update By Ethan, Abhi, Justin, and Kenna

    Task: Update the intake for the new robot size

    So, we created the corn-cob intake a few weeks ago. Unfortunately, it was a little too big for the Minichassis, so we had to downsize. Thus comes Intake Two. Continuing our history of using kitchen materials to create robot parts, we attached two silicone oven mitts to a beater bar equipped with the Iron Reign signature REVolution system. Then, we attached a REV Core Hex Motor to the design, then added a 2:1 gear ratio to increase the speed, as the motor wasn't exactly what we wanted.

    Then, we attached our new passive sorting system. Instead of being the old, bulky sorting system, the new system is just three side-by-side bars spaces 68mm apart with tilted wings to move blocks upwards. The 68mm number is important - the size of a gold block. This allows the balls to be struck and fly upwards into the intake while sliding the blocks through the system.

    Next Steps

    We need to attach this to the robot to test intake. The most likely way this'll be done is through a pivot over the walls of the crater from the top of the robot.

    BigWheel Arm

    BigWheel Arm By Evan

    Task: Design an arm for BigWheel

    Bigwheel’s arm is going tied with the lifter arm as the most integral part of the robot. It wouldn’t work without it. Since our scrimmage, we have learned how to make this arm much more efficient, starting with some supports. The arm was made of two disconnected Tetrix rails scavenged from the bottom of our scrap bin, lending itself to weakness and instability, much like a country that recently won its independence. The worst part was how the two sides of the arm would be out of sync with one another, creating a twist in the arm that caused it to drive in an odd path. Since then it has been stabilized with cross beam REV rails that have significantly straightened out the robot. Why we hadn’t done it before is a mystery to me, and something we should have recognized as an issue sooner, but hindsight is 20/20 and since we can’t change the past we can only vow never to make the mistake again. Always support your attachments. The next upgrade on the arm is going to be the box to hold the minerals. Right now it’s made of cardboard off an amazon box, and it kind of sucks. I can say this because I cut it out and made it. The plan is to make it out of polycarb but we only came to this conclusion after a bit of debate. The reason polycarb was not our immediate solution is because it’s unfortunately quite heavy, and instead the first thing we came to think of was thin plywood and duct tape. Thin slices of plywood would be taped together to create a fabric like box that still had form. This idea still lent itself to breakage, and we next went to a design using a thin plastic sheet, the same kind of plastic that is used inside milk cartons. The only issue is that it’s super weak and doesn’t form well, so we would have to build a frame for it, much like the plywood and tape. Finally we came to the polycarb and decided upon that. So far it works as a nice way to hold the blocks as we transport them into the lander.

    Next Steps

    Right now we’re toying around with the idea of an arm that not only flips out but also extends using a gear and tooth track made from Tetrix parts of days gone by. The reason for this is to gain a little extra height that we were lacking before in the robot and a little more flexibility when we grab minerals from the crater. To do this I had to take apart the arm from our first ever FTC robot, and use the toothed track and gear plus the extra long tetrix bars to create the slides. So far the slides are surprisingly smooth and we have high hopes for the future of the arm.

    Torque

    Torque By Karina

    Task: Calculate the torque needed to lift chassis

    After seeing how well the robots that could latch onto the lander performed at the scrimmage, Iron Reign knew that we had to be able to score these points. We originally tried lifting with a linear slide system on MiniMech, but it was not strong or sturdy enough for the small chassis, and would definitely not be a functional system on BigWheel in time for competition. And so we thought why not use this opportunity to *flex* on the other teams with an alternative design? An idea was born.

    We decided we would latch onto the lander using the same arm used for intake, and then pivot the main body of BigWheel up off of the ground about an "elbow joint", much like how humans do bicep curls. To do so, our motors would need to have enough torque to be able to lift the loaded chassis off the ground once the arm hooked onto the latch. First, we measured the mass of BigWheel. Then we found where the center of mass was located. The distance from the pivot point to the center of mass became our lever arm, also known as the radius.

    Calculating torque required knowing the forces acting on BigWheel at its center of mass. In this case, there was only the force due to gravity (F = mg). Before we could plug BigWheel's mass into the equation, we converted to units of kilograms (kg), and then used the value to find the newtons of force that would oppose the upward motion:

    Finally, we plugged the force and radius into the torque equation:

    Next Steps

    The next step is to test which gear train will output this torque value based on the motors used and the gear ratio.

    BigWheel Upgrades

    BigWheel Upgrades By Evan

    Task: Get BigWheel ready for the tournament

    Sunday and Monday were good days to be an intake system for big wheel. Mounts were built to attach both types of intake to the rack and pinion tetrix slide and a new way of mounting the arm to the chassis. The original corn on the cob intake was sized down for the system, and stitched to a REV rail design as opposed to the tetrix and REV rail hybrid it was before. It has yet to be chained up but that should happen soon. The idea is that since it’s attached to the rack and pinion track, it reaches high enough for the robot to put the minerals in the lander. The only issue is that it that we have yet to find an ideal gear ratio for the arm, but we imagine that what we have now will suffice (a small 12 tooth gear to a large 86 tooth gear for that extra torque). We have to work on the actual mount that the arm will attach to, the only issue being space and that there may not be enough of it without either switching to bevel gears or chaining the motors to the small 12 tooth gears. We have yet to decide exactly what we’ll do, but since we may not have the bevel gears we want, it might end up being a chained mechanism. Another addition that is simple enough but quite necessary, is a tail for the robot to be able to stop itself from flipping backwards, something that is a very real danger of the design. It will probably be made of polycarb with aluminum or steel support on either side, just incase the polycarb is not enough to support the push of the robot. Part of this process will involve some code tuning so that the robot stops itself, but the tail is necessary as a preventative measure.

    Next Steps

    There’s still a lot of stuff we will have to do to prepare the robot physically for the competition this saturday, but I believe it will get done.