A consistent topic of conversation amongst FIRST Lego League (FLL) coaches and mentors is how much help should the adults be giving students. It is a fine line, on one hand there is the desire for the students to succeed in the competition and on the other the need for the students to learn for themselves and experience success or failure by their own efforts.
In a similar school competition some years ago I watched as teachers followed the rule of the competition to the letter and did absolutely nothing for the students. The students had a great time but walked away disappointed about how 'bad' they were compared to other schools. In that competition I don't know if the top teams received help from their teachers, if they were more experienced with the competition and knew what to do or simply worked harder and were more motivated. But the final result was that the following year students didn't want to participate because of previous perceived failures.
Each year in FLL we see new schools enter the competition only to disappear the next year. Without being privy to their inner team conversations I can only speculate why they didn't return. Although it usually comes down to either cost to enter, time to prepare or failure to compete at the level they think they should be at.
So back to the issue of teacher help. By the strictest rules of FLL the coaches and mentors shouldn't do any of the work for the team. But what constitutes work, if the coach provides a robot design that the team build it who did the work? Or if the coach tells the team the sequence of code to use and the team actually write it?
In both cases I believe even the coach providing the code or the design is not in the spirit of the competition. The coach needs to be there providing support for the team but giving them something to copy doesn't help the team to learn. Real learning can be achieved by directing the team to resources, or demonstrating skills that the team then need to understand and apply to the FLL challenge.
I've talked previously about building demonstration machines from Yoshihito Isogawa's books. In collaboration with this I might demonstrate some different techniques for attaching to the robot but my goal is not to put all of those things together, those final steps of connecting concepts need to be done by the team.
The same goes with coding, before the competition actually kicks off the lessons in class are normally related to skills they might need in the tournament. Finding and following a line, aligning the robot around the mat or using sensors. When the competition starts I remind the team of those lessons and ask them to think about what they learnt or go back and load that code looking for hints.
When it comes to the competition it's important for the team to be honest about what they did and didn't do. Where a coach has helped the team they should be able to explain what the code or mechanism actually does and why.
Invariably there are teams at the tournament that were helped extensively by the coach. While we don't hear about it rest assured that the judges can spot this and will mark the team accordingly (or even disqualify them). As a coach I cannot imagine anything worse than knowing your team was disqualified because you didn't let the team do the work. Equally there is no pride for the team in collecting an award that they know the didn't do the work for.
Friday, 5 June 2015
Tuesday, 2 June 2015
FLL Casts Website Review
There are a lot of website and videos out there to help teachers and students learn Lego Mindstorms building and programming. So in my opinion anybody charging a subscription fee for their website would want to be offering an extremely high quality service. Enter fllcasts.com.
FLL Casts is a website run by five gentlemen in Bulgaria, who have extensive FLL experience in all aspects including tournament judging. At the time of writing this, the website has 92 video tutorials ranging from robot design and building (see my resources page for an adapted version of their EV3 competition robot), to software and sensor tutorials to discussions on how to complete the FLL challenges from seasons past. There is also links to competition rules and details for FLL seasons dating back to 2007. Videos are often supplemented with PDF materials of the designs or Mindstorms software downloads.
As EV3 is only a few years old much of the content is still based on NXT technology which while beneficial for some, it is fast hitting a point that this content will be redundant (at the Victorian FLL Championships last year more than half the teams used EV3).
At the moment they seem to provide a new video around once a week and from what I can tell they seem to have been in operation for just over 2 years. So the one video a week seems pretty consistent over that time.
The videos are clear and concise, with most lasting between 4-10 minutes. When they are longer they usually split them into at least two parts or 10 parts in the case of the "Big Daddy" robot tutorial. Although there is usually around 60 seconds "preview" at the start of each video for the non subscribers.
Community wise they seem to have a number of subscribers that comment on videos although it seems pretty rare to see many more than 5-6 comments on any one video. To their credit they engage with the commenters and respond to their, questions, ideas, or criticisms. I'd like to see this really expanded perhaps even into a forum for discussion on issues, currently this is done on comments, and their Facebook page.
Personally I've enjoyed the videos I've seen so far and have gained something out of the website. Even showing my students one or two of them to give them ideas. But for $8USD a month ($10+AUD) I can't see myself remaining a subscriber all year. I've watched a number of videos that interest me but at one video a week and no huge community to add content and discussion it's not really long term value for money. Maybe I'll sign back up later in the year for another month and watch the videos from the last six months that I've missed.
Summary
Good website, lots of content, not value for money.
FLL Casts is a website run by five gentlemen in Bulgaria, who have extensive FLL experience in all aspects including tournament judging. At the time of writing this, the website has 92 video tutorials ranging from robot design and building (see my resources page for an adapted version of their EV3 competition robot), to software and sensor tutorials to discussions on how to complete the FLL challenges from seasons past. There is also links to competition rules and details for FLL seasons dating back to 2007. Videos are often supplemented with PDF materials of the designs or Mindstorms software downloads.
As EV3 is only a few years old much of the content is still based on NXT technology which while beneficial for some, it is fast hitting a point that this content will be redundant (at the Victorian FLL Championships last year more than half the teams used EV3).
At the moment they seem to provide a new video around once a week and from what I can tell they seem to have been in operation for just over 2 years. So the one video a week seems pretty consistent over that time.
The videos are clear and concise, with most lasting between 4-10 minutes. When they are longer they usually split them into at least two parts or 10 parts in the case of the "Big Daddy" robot tutorial. Although there is usually around 60 seconds "preview" at the start of each video for the non subscribers.
Community wise they seem to have a number of subscribers that comment on videos although it seems pretty rare to see many more than 5-6 comments on any one video. To their credit they engage with the commenters and respond to their, questions, ideas, or criticisms. I'd like to see this really expanded perhaps even into a forum for discussion on issues, currently this is done on comments, and their Facebook page.
Personally I've enjoyed the videos I've seen so far and have gained something out of the website. Even showing my students one or two of them to give them ideas. But for $8USD a month ($10+AUD) I can't see myself remaining a subscriber all year. I've watched a number of videos that interest me but at one video a week and no huge community to add content and discussion it's not really long term value for money. Maybe I'll sign back up later in the year for another month and watch the videos from the last six months that I've missed.
Summary
Good website, lots of content, not value for money.
Learning from Technic Models
Like most kids I always looked forward to Christmas and my birthday and was even happier when a rectangular boxed present would appear as it usually meant I was about to score a brand new Lego model to build (8854 was a particular favourite). Twenty five years later my answer to family and friends questions about things I would like usually involves more Lego (you can never have to much Lego!).
However, when I look back at those models I don't think I fully got the educational benefit of them at the time. It was always a race to get the model built so I could play with it and I didn't really take the time to look at what the pieces were doing in the model.
Now when I build a model I try and ask myself "what is this piece actually for?" Is it going to be part of the steering, the crane or the tip truck. Is this beam going to provide a brace for something, why does it have only two pins and not six. Sometimes the questions can't be answered until the model is finished, and sometimes I get more from by pulling the already completed model apart. Over the last week I've slowly deconstructed set 9395 Tow Truck and currently have the chassis sitting half built on my desk. This model had a driving ring for a clutch gear which I've often wondered about, by pulling it apart I've seen how the Lego designers used it and have subsequently built a working example of one to share with students.
I know this is not an easy concept for students, but sometimes building and pulling apart a model is a great way to learn. Provided the builder has an open mindset and thinks about the model is doing and not just competing in a race to finish.
However, when I look back at those models I don't think I fully got the educational benefit of them at the time. It was always a race to get the model built so I could play with it and I didn't really take the time to look at what the pieces were doing in the model.
Now when I build a model I try and ask myself "what is this piece actually for?" Is it going to be part of the steering, the crane or the tip truck. Is this beam going to provide a brace for something, why does it have only two pins and not six. Sometimes the questions can't be answered until the model is finished, and sometimes I get more from by pulling the already completed model apart. Over the last week I've slowly deconstructed set 9395 Tow Truck and currently have the chassis sitting half built on my desk. This model had a driving ring for a clutch gear which I've often wondered about, by pulling it apart I've seen how the Lego designers used it and have subsequently built a working example of one to share with students.
 |
| Chassis from 9395 Tow Truck |
 |
| Driving ring and shifter to change between two gears. |
Changing the angle of rotation
What is the most important skill for students to have in their robot building?
I've thought about this a lot recently and everything is so intertwined that to pin point a single skill is proving difficult. Students need to have the ability to make a strong model, using a minimal quantity of but also the appropriate pieces. Students need to have the understanding of some simple machines concepts like gears and leverage, not only how the concepts work but how they can implement them in their creations.
So what's the answer? Firstly I think the more students build the better they become, they see which pieces work together and which ones don't. The better they know the pieces the better they can become working with them and knowing which piece is the most appropriate. To practice this I've recently been asking students to come up with multiple ways of 'changing the angle of rotation'. Meaning, connect two gears so that the drive gear is turning on the x axis or horizontally (blue) and the driven gear is on the y axis/vertically (tan).
I've thought about this a lot recently and everything is so intertwined that to pin point a single skill is proving difficult. Students need to have the ability to make a strong model, using a minimal quantity of but also the appropriate pieces. Students need to have the understanding of some simple machines concepts like gears and leverage, not only how the concepts work but how they can implement them in their creations.
So what's the answer? Firstly I think the more students build the better they become, they see which pieces work together and which ones don't. The better they know the pieces the better they can become working with them and knowing which piece is the most appropriate. To practice this I've recently been asking students to come up with multiple ways of 'changing the angle of rotation'. Meaning, connect two gears so that the drive gear is turning on the x axis or horizontally (blue) and the driven gear is on the y axis/vertically (tan).
I find this is a really tricky task for many students. While they easily get the concept they often struggle actually connecting the two and come up with convoluted systems of axles and bent liftarms to achieve something that partly resembles the brief.
That's where these pieces come in handy, each of these pieces change the direction of the technic beams and provide some clues for students. I'm still finding I need to show them some models either physically or from one of Yoshihito Isogawa's books to help them see how the pieces work but with this help they are starting to improve their knowledge of technic engineering, becoming more confident with the pieces and are increasingly solving more challenges.
Friday, 29 May 2015
Competition Robot Design
This year I've been trying to get more teams doing FIRST Lego League in Victoria and the more new coaches I talk to the more I realise their is a lot of misconception about what a competition robot needs to be. So lets look at some of those myths:
The Robot Must Be 100% Designed by the Students
False - ideally the robot should be 100% the students work, but nobody in FLL expects a rookie team of 9-10 year olds to create a World Festival standard robot. The key is that students give credit where credit is due, it's fine to take an idea of the internet but don't turn up saying we designed this robot. What the judges want to hear is where you got the design from, why you chose it, how does it work, which parts did you design, how did you come up with attachments.
The Robot Needs to Be Complex
False - Many teams enter FLL using the standard EV3 education kit robot design. It doesn't have to be fancy and involve huge attachments. KIS - Keep It Simple, and build on the team knowledge each year.
The Coach Needs to Know How to Build a Robot
False - Nothing gives me more pleasure than my students coming up with something I don't understand. It shows they are learning and not afraid to experiment and apply new skills. Have faith that the students will use their creativity to solve problems. As a coach at this point your role changes from showing to questioning. As the team how the robot works? Can they make it stronger? Can they make it smaller/more compact? Can they use fewer pieces?
Success is the Most Important Thing
False - Lots of coaches (and parents) go into the competition wanting their team to win and sometimes that leads to coaches having a little too much involvement in the robot design. Allow the students to fail at times, it's ok for the robot to not work 100% of the time or for something to fall off mid run. Never underestimate what you're doing to a students confidence level if you tell them their idea isn't any good and they should just use what you designed. Next time the student tries to make something are they going to try themselves or go straight to you for help?
So what to do as a rookie team?
Talk as a team about the approach they want to take. Are they confident building? Do they have time? Many teams won't get much more than 1-2 hours a week to prepare their FLL entry and with a project to do as well sometimes they would prefer to jump straight into the programming.
If they are going to build their own robot you can start before the mat is released in late August, come up with a basic robot design and that way when the challenge is released you can just make minor changes based on the new challenges. For example, does your robot need high clearance like in Nature's Fury, fit into tight spaces (World Class) or climb a ramp or stairs (Senior Solutions).
If they would like to use somebody else's design, 'shop' around, look at different designs and talk about which one they think suits them best. Which one has mechanisms they understand and can talk about, remember, if they don't understand how the robot works they really shouldn't use it. Make sure they understand they need to give credit to the designer and not claim the work as their own.
If they're interested in using somebody else's design check out 'Competition Robot V1' from the Resource Download section of this blog. It can be made from ONE EV3 Education box (with the exception of a second colour sensor). It is adapted from a design on fllcasts.com and is a good place for teams to start. The robot design easily allows for attachments (see future blog posts) and has lots of room for the students to modify the design with new attachments, motors and sensors.
Tuesday, 26 May 2015
Yoshihto Isogawa - My Engineering Lesson Saviour
Like most teachers I don't have a background in engineering furthermore like a lot of teachers I don't really consider myself to have an engineering brain. As a result most of the time I've shied away from running these sorts of lessons and instead chose to focus on a standard EV3 (or NXT) model and taught programming instead.
However, this is not always what suits my students. Kids charge towards "Lego Robotics" classes like it's going to involve building huge robots and can easily get disheartened when the lesson is just about sitting on a computer and dragging colour blocks around. So what's a teacher to do with this conundrum? Let loose the control and let students free build? Set a task hoping the students can work out a solution by themselves? (see previous post about Context Help for reasons this can be unsuccessful). I've tried both of these with little success. I also tried introducing Simple Machine lessons to the younger grades and then hoping those skills develop by the time they get to EV3 (again with little success).
Earlier this year my saviour came along, Yoshihito Isogawa.
Yoshihito has around 40 years experience working with Lego and the brilliance of his books comes from the complete lack of text. Usually the books have a few pages of introduction at the front but from there on it's about the images. Simple icons show if the model is about gearing, levers, angle of rotation etc and each model is photographed from a variety of angles to help the reader understand it.
The models are not super complex robots that do it all, instead they are simple mechanisms, 2-3 gears, a rubber band, a double bend liftarm. It is up to the reader to build the model, understand how the mechanism works and either combine it with another design or adapt it to fit their purpose.
For my students I've built a variety of these models which will act as a library for FIRST Lego League. Using these designs the students can see how to change the angle of rotation using gears (or another Lego element) but from there it is up to them to work out how to use this mechanism to solve a challenge (and attach it to their robot).
Yoshihito has four books available on Amazon of which I recommend the EV3 Ideas book as the best. It uses all modern Lego elements while the Technic Idea books use the older style technic bricks, still great but the pieces are less likely to be found in the school (or personal) Lego stash.
(yes these are affiliate links, if you're going to buy the books use these links and help kick back some money for the work I put into this blog, it won't cost you any extra and helps me out)
Context Help
One of the approaches I take in the classroom when introducing students to EV3 is to ask the students to find out for themselves.
It has mixed success, some students relish this opportunity to really dive into the program and work things out for themselves while others stare blankly at the screen and before asking for help. I believe the approach is beneficial for students as it asks them to think about what they know, and search for resources to help them be it other students, built in guides, youtube or other websites. These students also have a better longterm understanding and are able to problem solve while the screen staring group struggle to solve problems.
Of course I usually give the students some guidance on where to find help (see future post about helpful websites) but recently I discovered a feature of Lego Mindstorms EV3 called 'context help'.
It has mixed success, some students relish this opportunity to really dive into the program and work things out for themselves while others stare blankly at the screen and before asking for help. I believe the approach is beneficial for students as it asks them to think about what they know, and search for resources to help them be it other students, built in guides, youtube or other websites. These students also have a better longterm understanding and are able to problem solve while the screen staring group struggle to solve problems.
Of course I usually give the students some guidance on where to find help (see future post about helpful websites) but recently I discovered a feature of Lego Mindstorms EV3 called 'context help'.
The idea behind context help is that a little help box pops up with a description of whatever the mouse is hovering over. While some students can work out what the little icons mean (eg, hourglass means wait) other learners prefer the text to help them.
Sometimes the help can be slow to load depending on where the mouse is hovering but I find by clicking on the unknown button it usually brings it up a little quicker.
It's a great simple tip that can help students to solve a problem without showing them the answer.
Subscribe to:
Posts (Atom)