Now that you have learned about how to make the robot navigate the game field, the next trick is to have the robot actually do something when it gets to its desired location. We will do this with particular attachments that will be mounted to the robot chassis and designed to help the robot carry out particular missions. A winning robot only navigates the game field effectively but it must be able to complete the game missions.
The design of attachments is twofold. First, a design needs to be such that it can actually do the desired task without error, and second, an attachment needs to be versatile enough that it can be either reused for multiple missions or designed in such a way that attaching and removing it from the robot chassis is smooth and easy to do. Attachments are the hooks, claws, collectors, or really just about anything your robot can use to complete a task. In most cases, there isn't a single attachment design that can do all the missions on a game field unless the game missions are fairly simple. In FLL events, there will normally be around ten (give or take a few) tasks that the robot must do to complete the missions, and rarely will a single tool be useful for all of these missions. Because of the fact that these tools might have to be swapped out, we must design tools that can be added and removed from our robot chassis, that is, attachments.
Once attached to our robot chassis, a tool becomes a part of the robot and must follow the same guidelines that the robot itself in regards to shape, size, and parts used. The rules that regulate the size of your robot in FLL apply to the complete size of your robot with the attachments included, not just the robot chassis. Keep the FLL design rules in mind when building attachments for your robot; it's easy to overlook them until it's too late.
Since attachments could need to be removed and added during the competition, it's best to design a solution that can be added and removed in as little time as possible. At FLL events, a team only has 2.5 minutes to complete as many of the missions as possible, and swapping out attachments on the robot in base can be one of the most time consuming things a team will do. So don't make attachments hard to put on or remove; keep them simple. Also practice over and over again the addition and removal of your robot's attachments.
LEGO robot attachments can be broken into two basic categories, either passive or powered. In this chapter, we will talk about passive attachments. These are attachments that do not require a motor to operate; they are simply hooked to the robot chassis and controlled by the navigation of the robot itself. Powered attachments will use a motor to enable to attachment to work; for most robots, their third motor will drive an attachment. Recall that in FLL your robot is allowed to use only three motors in total, and in most cases, two are used to navigate the robot. Chapter 9 will cover the design and uses of powered attachments. Powered attachments can also have a subcategory of pneumatics, where LEGO pneumatics are used to power the attachments movements instead of a LEGO NXT motor; Chapter 10 will cover the design principles of pneumatic attachments.
How do you know when you should use a passive attachment versus a powered attachment? There is no right or wrong answer to this question; it simply comes down to determining the right tool for the job. Most robot teams will use a combination of both passive and powered. One of the advantages of using a passive attachment is that most passive attachments have very simple designs and don't require an extensive amount of engineering; for a young or new team, this can be helpful.
The types of passive attachments are pretty much only limited by what your imagination can dream up, but most of them will fall into one of the following categories:
Of course, some ideas will be completely outside of these general categories and there is nothing wrong with that. Again the only limitations are the game rules and what you and your team can dream up. As long as it works, there is no bad design.
When thinking of design ideas for your attachments, look to real life machines or tools that perform a similar function. For example, if you have a mission that requires you to simply push a field item, think of machines that push, such as a bulldozer or snow plow. Both of these push things but have a different-shaped blade depending on the goal for the task.
Maybe the task is to latch onto something, say a loop on the field, so you could make an attachment that is similar to a fishing hook; a hook can grab an item from one direction but not release when moving in a different direction.
Anytime you can associate a desired action with an existing tool or machine, you've saved yourself a great deal of work. Now, you just have to be able to simulate that tool in LEGO and make it applicable to your robot design. These are good things to bring up during any robot design judging that you may have at your competition; often, design judges will ask where the inspiration for your designs came from and being able to point out such similarities with existing tools or machines is a nice touch. It shows you did your homework when designing your robot and its attachments.
One of the most common and easiest attachments to have on your robot is one that pushes. A pushing attachment can be as simple as a flat LEGO wall or something more complicated like a plow that not only pushes but clears a path. Pushing can also be about delivering something on the game field. Many times a game will have missions that require delivering something to a particular place on the game field and really the easiest way to get it there is to just push it along the mat.
A bumper attachment is just that, a small wall that is hooked to your robot's chassis (see Figure 8-1). It doesn't have to be a big wall could be as simple as a small bumper like a car would have on the front. The idea you're going for is a flat surface on the front of your robot that you can use to run into objects. In the FLL 2010 Body Forward game, there were lots of missions that required the robot to make contact and push the field object. The pushing would result in some action happening, such a door opening or a lever lifting something up.
The attachment doesn't need to be fancy or complicated, it just needed to be able to make proper contact with the field object. You could then use the robot's drive system to push the bumper forward. With a solution such at this the programming becomes important since the drive system is not only moving the robot its causing some interaction with the field object. A delicate touch is what is needed most of the time; rarely do you want to hit something hard for fear that you'll damage the field object or the robot.
Unlike a bumper, where you want to push an object forward, you may have the need to push something out of your way or possibly clear a path. If this is the case, you don't want a flat bumper that meets the object perpendicularly; you need something that will make impact and then push the object clear of the robot's path. Think of a snow plow; the blade on a snow plow is at an angle, and as the snow is pushed, it goes off to the side. The simple motion of the plow truck and the shape of the plow blade make this happen. The very same principle can be applied with a LEGO robot.
Just build your bumper with an angle that will move the object without much force. Depending on the object you are trying to move, the size of the angle and the force needed will differ. Don't be afraid to experiment and try different designs. Be sure to take notes of the different designs you try and include the findings in your technical documentation that you present to any design judges at your event. Being able to document why your team built something the way that you did is always good in the judge's eyes.
Your plow should have a smooth face on it as well. Don't just take a LEGO plate with studs facing forward and expect your target object to move out of the way. You want to minimize the friction by having a nice smooth surface on your plow. If you do use a LEGO plate, be sure to add some LEGO tiles to it so that any studs facing out don't cause objects to get caught or not move as you expected. The plow in Figure 8-2 keeps the smooth side of the TECHNIC beams exposed so that any objects that make contact will slide out of the way.
Maybe instead of trying to make contact with a field object you're trying to deliver some objects on the field. You could try to build a complicated claw or attachment that contains the objects and then opens up to release them, but many times, a simple delivery box will do. In Figure 8-3, the robot is using a four-sided box to deliver the ball; without the box, the ball could roll away from the robot.
In the FLL 2008 Climate Connections game, many objects had to be delivered to particular places on the field. The rules stated that the objects must be making contact with the field mat but there were no rules against containing or corralling the object in a box made from LEGO bricks. In order to keep with the rules about making contact with the game mat, you simply built a LEGO box that had four sides but no bottom.
The great part about this was that you just had to drop the pieces into the box that needed to be delivered and then push the box along the game field to the desired location. No special arms or fancy attachments were needed on the robot, just a way to push the box. You could have had something as easy as a bumper or something with a little more design effort that would hold the delivery box on three sides but allow the box to be release when the robot went in reverse. Just a bumper with three sides that fit around the box would be perfect and could possibility be reusable for other task as well.
Note
Attachments that can be used for more than one task not only save you design time but can also save your team lots of competition time by avoiding the need to change out attachments.
When you build a delivery box think about how the box will travel across the game field. You will want the box to have as little friction with the mat as possible. You could put some tiny wheels on the box, but then you run the risk of having trouble if the robot needs to make turns while making the delivery, since the wheels on the box are not going to steer with your robot (they will simply skid). Adding some skids to the box would be a better idea, anything that is smooth and slides easy on the mat surface. Here you can be creative; try using TECHNIC beams on their sides or some LEGO tiles attached to the bottom of your box. Don't be afraid to use parts you would never have thought would be handy in LEGO robots events. I've seen LEGO minifig snow skis used, and they slid across the game field well. Again, just like with the plow attachment, don't be afraid to try different idea and test them out. Just be sure that you document everything so that you can show design judges how your solutions came about.
Hooking objects and returning them to base is one of the more popular tasks on FLL game fields in recent years. Capturing loops seems to be a common mission that robots have had to perform in past years. Many teams will overthink this kind of task and build overly complicated attachments to retrieve the loops. I understand the attraction to build big and complicated for the cool factor, but this is not normally the winning design you want. Having more moving parts just means there are more things to break and go wrong with your attachment. The key to a consistent robot is keeping it simple. The fewer things that can go wrong, the better off your robot will be in the long run.
To start off simple, a basic hook shaped attachment can do wonders with a little good programming behind it. The strategy will be for the robot to navigate up to the object and place the hook in such a position that when the robot moves away the field object is caught on the robot's hook and can be returned to base without falling off. In Figures 8-4 and 8-5, the robot has a simple hook attachments; once the robot moves into position, pulling the object back to base is done without great effort.
While hooking an object may sound easy, it will require a robot that can navigate very smoothly and consistently. Depending on your hook design and what you are trying to capture, there may not be a lot of room for error. Keep this in mind when deciding on your hooks design; even though the hook itself is simple, actually getting it to work each time might be a bit more of a challenge.
If you look at the design of a fish hook, you will notice the end of the hook has a barb facing the opposite direction. When used on a fish, the big hook actually catches the fish, but the barb keeps the fish from slipping off the hook. This same design can be used with LEGO robot attachments as well by simply building a big hook with some kind of LEGO element on the tip that will keep anything the hook catches from slipping back off. Again, don't over think the design; just add a simple bushing or pin on the end of your hook so that the newly captured object isn't allowed to fall off the hook before you return to base with your prize.
In Figure 8-6, the steps for a robot retrieving a loop using a fish hook attachment are shown.
The robot faces the loop.
Now, it drives past the loop.
It turns carefully toward the loop.
As the robot moves backward, the hook is caught.
If you have ever gone rock climbing or seen rock climbing equipment, you have most likely seen a carabiner. These great hooks have a spring-loaded gate on them. The idea is that you the hook will only go in one direction when capturing an object. The gate on the carabiner only opens in one direction; the robot navigates to the object and forces the carabineer hook onto or over the object, allowing the gate to open when hooking the object, but when the gate closes, it will not open in the opposite direction thus holding the object on the hook.
Figure 8-7 shows a carabiner hook attachment over the loop object with the hook's latch in a closed position. In Figure 8-8, you can see the pressure from the attachment coming down on top of the loop caused the latch on the carabiner to open. Once the loop comes inside the carabineer, the latch closes around the loop and locks it in place, as shown in Figure 8-9.
Building such a hook is rather simple with LEGO bricks; you just build a hook with a gate that is held in place with a LEGO belt (you will have lots of various sizes in your LEGO MINDSTORM kit). Be sure to make the gate big enough for your object to fit through, and again, make sure there is some room for error. Don't make the opening so narrow that your object will only fit if the robot hits it perfectly. This is something you should practice repeatedly until you come up with a design that will perform without error. You may need to adjust the belt you used for your gate spring to increase or decrease the tension as well as work with the overall size of your hook.
While, technically, the fork design is not exactly a hook, it does work well when trying to collect loops on a game field. Passive fork designs are not always effective since many times you may need to lift the loop that you are collecting and that will required a powered attachment, which I'll talk about in Chapter 9. There are times when a passive fork design will work though. If you are creative, you might even be able to design a solution where the fork is not powered but can still create lift as it moves into the loop. Figure 8-10 shows a robot with a fork attachment.
The idea behind a fork attachment is just like a fork you would use at the dinner table. It has a set of prongs on the front and is driven into the object you are trying to capture. You will need to keep the size of the object in mind when designing your fork attachment, since the object does need to fit between the prongs on your fork. The number of prongs is also something you need to keep in mind; the more prongs you have, the more room for error you have as well. However, if you make the fork too big, you could run into navigation issues when trying to return to base with your prize.
You could also build a hybrid of the fish hook and the fork; by adding small barbs on the end of each fork prong, you can keep the object from slipping free without having to lift the fork. Again, this depends on how your game field is set up and what type of object you are trying to collect.
At times, your robot will need to deliver an object or a group of objects that need to be put behind or on top of another object. In these cases, just pushing them across the game field will not be enough you will need to find a way to deliver them in a different way. In most cases, dumping the delivery will work. Think of machines that dump; the obvious example is a dump truck. It has a large bed on the back that lifts to let the contents slide out to a location behind the truck. The same idea can work for your robot as well.
Think of how the bed on a dump truck works: it lifts up to cause the items in the bed to dump out. But we're working with passive attachments, so we will need to find a different way to cause the dumping action to take place. Gravity will be our friend when we try to design such an attachment. The simplest way to build a dump bed is to create a tilting bed that can be locked into the load position and has a trigger that will be pushed out of the way to allow the tilting bed to fall and release the contents.
First, build the tilting bed so that it is large enough to hold all the objects you wish to deliver. Also, make sure that the surface of the tilt bed is smooth enough to allow the contents to slide out without a lot of friction. It should also be large enough so that, when the objects are sliding out of the tilt bed, they do not get jammed up against each other causing them to get stuck in the tilt bed.
Figure 8-11 shows a dump bed. It is important that the pivot point is kept far enough back that when the trigger is released, the dump bed will tilt forward enough to release the contents. The trigger is holding the bed in place, and when the trigger is pushed backward, the dump bed becomes unstable and falls forward, as shown in Figure 8-12.
The tilt bed needs to be mounted high enough on the robot chassis so that when the robot moves into place for the delivery, the contents of the tilt bed will be able to reach their destination. For example, in the FLL 2008 Climate Connections game, one of the missions was to deliver a number of items behind a short wall of LEGO bricks. In order for a tilt bed to work correctly in delivering the items, it needed to be mounted high enough to reach over the wall of LEGO bricks and still have room for the bed to tilt enough for the contents to land in the proper location.
Figure 8.11. A dump attachment carring a load of small trees with a trigger in place preventing the dump bed from releasing
Figure 8.12. Once the trigger is pushed backward, the dump bed falls forward and releases the load of small trees.
The tilting action itself will be done by taking advantage of potential energy that is stored in the tilt bed by locking it in place with a lever or pin.
Note
Potential energy is energy that is stored within a system. It exists when there is a force that tends to pull an object back toward some lower energy position. This force is often called a restoring force. For example, when a rubber band is stretched to the left, it exerts a force to the right so as to return to its original, unstretched position. Similarly, when a mass is lifted up, the force of gravity will act to bring it back down. The action of stretching the rubber band or lifting the mass requires energy to perform. The energy that went into lifting up the mass is stored in its position in the gravitational field, and the energy it took to stretch the rubber band is stored in the rubber.
Now, the energy we used to lift the tilt bed in place is storing energy for tilting when the lever or pin holding it in place is removed. The balance of the tilting bed should be designed in such a way that when the bed is in not being held in place by a lever or pin, it will return quickly to the dump state. To adjust the balance of your tilt bed, you just need to move the position of the fulcrum. If gravity is not enough to tilt the bed as desired, a spring can be added by using a LEGO belt attached to the tilting bed. When the bed is in delivery state, the belt is stretched, and once the trigger is pressed, the energy bound in the belt will release and cause the bed to dump its load.
The trigger for your dump bed will be located in such a way that when the robot arrives at the desired delivery location, the trigger will be pressed. Just like with other passive attachments, be sure to make the trigger big enough that you have plenty of room for error. You don't want to make the trigger so small or difficult to activate that you decrease your chances of completing the mission correctly each time.
Now, for your robot to make the delivery, you load up the items into your tilt bed while the robot is sitting in base. Then, the robot navigates to the desired dumping zone or location. Drive forward until the tilt bed trigger is pushed and the tilt bed releases and dumps its contents in the proper location.
Many LEGO robot challenges will require robots to collect field objects and bring them back to base or deliver them to other locations on the field. The hook attachment we talked about previously is one kind of attachment that can be used for collecting particular objects that have a handle or loop that the hook can latch onto. Sometimes, the objects you are trying to collect might be shaped in such a way that they have nothing for you to hook onto, so you will need some other designs for your passive collectors.
A ball can be of the trickier things to try to collect on a game field. Balls don't have anything we can grab with a hook, and they tend to roll away when we try to push them. Building a one-way box for collecting such items is a good solution. For this attachment, you have a box with three stationary sides and a forth wall that is a flap that only opens in one direction, allowing balls to enter the box but not to leave.
With such an attachment, your robot can navigate around the game field pushing the one-way box attachment into the path of the objects you wish to collect, such as balls. When building the opening side of the box, be sure to build it large enough to be able to capture your objects and still have room for the flap to open again without any of the previously collected items jamming the flap and preventing it from opening.
In Figure 8-13, the robot is pushing a three-sided box with a flap on the front that is kept shut by gravity. It has a stop on it that allows the flap to open only inward (it prevents the flap from opening in the outward direction), thus keeping us from loosing anything that we capture. Figure 8-14 demonstrates how the ball will push past the flap and enter the box as the robot moves forward. Now, the ball is trapped in the attachment and cannot escape even if the robot drives backward, as shown in Figure 8-15.
The flap on your attachment can either be spring loaded with a LEGO belt or just use gravity to stay closed when no objects are entering the box. First, try using the attachment with just gravity closing the flap. If the collected objects tend to escape after being collected, you may need to add a spring to the door to force it closed faster. Be careful not to make the tension on the flap so tight that objects trying to enter the box are pushed away instead of being captured.
Figure 8.15. Once the ball enters the attachment, the door swings closed behind it, locking it inside.
Another variation on the one-way box would be to have just a lip on the front of the box instead of a flap. By adding a plate or axle across the front of the box, you can prevent some items from exiting the collection box as long as the robot continues to roll forward. Be careful though, if the robot moves backward quickly or for a long distance, the objects you have collected in the box can get free.
A sweeper is similar to the one-way box in principle, but instead of having a simple flap on the front of the box, you add a more complicated apparatus. The idea might be similar to a vacuum cleaner; there would be a set of spinning brushes or blades on the front of the box. The spinning motion could be activated by a wheel on the outside of the box that is attached to the drive axle of your sweeper. As the robot navigates the field, the wheel rolls along the mat and transfers the motion to the sweeper bushes or blades.
This kind of attachment can be very handy when trying to collects lots of small items that tend to roll away easily or are hard to collect with hooks or bumpers. Most of the time, sweeper attachments are rather large, so keep the overall size of your robot in mind when you build this type of attachment. You do not want to exceed the size limits. Also, these types of attachments require a good bit of testing and reengineering to get working properly, but they can be a great way to collect multiple of objects of various sizes at one time.
Figure 8-16 shows a robot with a sweeper attachment. As the robot moves forward, the wheel will turn the gears that cause the sweeper arms to rotate and push items into the attachment.
As I noted before, attachments that can be used for many different tasks are best, since they will save time and effort on the game field. So if your game rules require you to do a lot of collecting on the field, a sweeper of some kind might be an ideal solution.
Many times, you may want an attachment that performs a power function, but you don't want to use a motor as the source of the power. Using springs or LEGO belts is a great way to get an attachment to perform an action without using a motor.
LEGO belts are, for the most part, fancy rubber bands. Now think of what happens when you pull a rubber band back on your finger and let go at one end? It shoots across the room, because when you pull the rubber band back and stretch it out, you are storing energy in the rubber band, and when you release one end of the rubber band, you are releasing the energy that was stored. With a LEGO belt, you can use the same principle (but do not shoot them like rubber bands; they will break). Figure 8-16 shows that with just a few parts, a powerful flipper can be created. Figure 8-17 shows the flipper compressed against the field table wall, and Figure 8-18 shows the flipper after the robot has turned and released it.
Figure 8.17. A flipper that can be attached to a robot chassis and then pulled back to be released for triggering field objects
Figure 8.19. Robot releasing the flipper by turning away from the wall, releasing the compressed flipper
With passive attachments, you can build flippers that are bound to a belt and pulled back and locked in place with a trigger. For example, if you have a mission object that needs to be knocked off its base, a flipper attachment would be ideal for doing this. The robot starts in base with the flipper locked and loaded. Then it navigates to the field object. When the trigger is released by bumping into a particular element on the field (or the table wall), the trigger will be released, and the energy from the belt will cause the attachment to flip quickly and strike the object that you are hoping to hit.
The drawback with this type of attachment is that you only get one use of the flipper per trip from base, since the attachment cannot reload itself and must be manually loaded in base by a team member.
Now that you have a collection of attachments, you need a way to hook them onto your robot chassis quickly, and you need to be able to remove them just as quickly. As noted before, adding and removing an attachment is one of the biggest time killers for a team when competing. Think of a race car pit crew; when the car comes for a pit stop, the team must work quickly to change out the tires, add fuel, and even clean the windshield. The same is true for when your robot returns to base and your team has to switch out attachments. The team must be well rehearsed in the changing process, and the attachments need to be design in such a way that allows for easy removal and addition.
Making up a system that allows for the attachments to come on and off the robot is very important. With passive attachments, the interface can be very simple, since you do not have to worry about motor attachments. If you are using both powered and passive attachments on your robot, keep in mind that you may need to design an attachment interface that will accept both types.
Using TECHNIC pins is the most common way to connect attachments to your robot chassis and works well as long as you keep the design simple and easy to access. You don't want to fumble with hard-to-access pins or pin holes when trying to connect your attachment to the robot. Remember, saving time is the main goal, so keep everything easy to access and see. If you can use the same pin layout for all your attachments, thus creating a universal interface as shown in Figure 8-20, you will make the process of adding and removing attachments much more streamlined. And when you add new attachment designs using the same universal interface, the learning curve is reduced everyone on the team. If all your team members are familiar with the interface you've been using to connect your previous attachments, adding a new attachment that uses the same interface will be less confusing for everyone.
Some attachments can be connected to a chassis without having to be hard attached; they can simply be held in place by gravity and some nonsnapping pins. To use this type of interface, you have a set of nonsnapping pins made from something as simple as a TECHNIC axle that just rest in some holes on the robot chassis. Nothing is truly snapped in place; it's just set in place and held by the weight of the attachment itself. If this type of interface works for your attachments, adding and removing attachments can be done very quickly without having to snap or unsnap anything. Again, the goal is speed. Figure 8-20 shows a bumper attachment with non-snapping pins being connected to the robot chassis.
Even thought they don't come in the MINDSTORM kit, LEGO does make magnets and magnet holders. LEGO train sets are a great source for these magnets (the trains use them as couplers between cars). They can be used on your robot as a coupler for connecting attachments as well. The magnets are very strong, and depending on the size of your attachment, you may be able to use the magnets alone. If you require a bit more support, you can use the magnets along with the nonsnapping pin system: the pins guide the attachment into place on your robot chassis and the magnets hold everything tightly in place. Figure 8-21 shows a pair of LEGO train magnets installed.
All of these passive attachment designs are purely suggestions; there is truly no limit to the number of designs that can be created to complete LEGO robotics task. Don't be afraid to experiment and try new designs or mix together design ideas. The keys to a good attachment—both active and passive—are reusability, quick addition and removal, and predictability.