Alternate Title: Michael's FLL Methods Manifesto
Note: this article was published in early 2017 when the FLL masterclass was first launching. It's fully formed now, but we're still going to update it throughout the 2018 season as we receive student questions.
From Coach Michael: When you sign up for the masterclass, you’ll create a login and password and you’ll then be able to access a whole arsenal of video content. You don’t have to log on at specific times or commit to any definite time frame. In the videos, I’ll be teaching concepts and demonstrating design principles with hands-on examples. I’ll test designs on camera (see below for sample) and I’ll show you how to test your own designs effectively to ensure reliability and save time. New videos will be uploaded regularly as the season progresses and as I learn more about the 2017 challenges (and the specific troubles my students are tending to have with those challenges). We'll start with the topics you should be concerning yourself with at the beginning of the season (robot chassis design, overall strategy formulation, teamwork planning, etc.) and in this way, I'll help you stay on track with your project timeline (I'll also help you develop that project timeline). My aim is to turn this year's FIRST LEGO League robot game into a more structured and educational experience for your child. FLL veterans will explore advanced engineering principles and master FLL robotics "best practices" while new students begin to grasp FLL robotics fundamentals and gain the insight required to skip over many of the classic "rookie mistakes". Making mistakes is central to the FLL process, but without skilled guidance kids tend to make the same mistakes over and over without learning from them. There is no shortcut approach, and I cannot teach kids “how to win” (nor would I). I simply teach them robotics. Having taught Lego robotics in a camp and after-school format for 14 years while coaching and mentoring FLL teams, I can verify that kids who understand Lego robotics at a deeper level tend to learn exponentially more in FLL because they are prepared to make the ‘right’ kind of mistakes and to be scientific in learning from those mistakes. I teach engineering principles, Lego robotics best practices, scientific testing procedures, etc. so that kids can become more serious about FLL and start making informed mistakes rather than haphazard ones. Coaches and mentors aren’t usually able to provide this level of pedagogy, and this is ok because the coach/mentor role is simply to ask questions and offer encouragement. I play the role of the teacher who prepares students to become more effective in their learning process. I’ve enjoyed watching my students blossom into mini inventors and engineers as they gain confidence and learn to tackle confusing problems logically and effectively.Note: I am careful not to discuss specific ideas or strategies for this year's challenge; that would spoil all the fun!
What will you learn, specifically?
Robot Base Design
First, I'll show students how to build a robust and capable robot base (using LEGO Mindstorms EV3) that drives straight and gets to where it needs to go RELIABLY. LEGO parts are inherently unreliable, and it’s difficult to build a robot base that functions excellently unless you have years of experience (or expert advice). If you gloss over this step, you’ll end up wasting countless hours of valuable programming time later in the season trying to figure out why your robot won’t drive straight. I’ll demonstrate excellent robot design by turning on my lab cameras and building a new FLL robot from scratch. I’ll brainstorm, prototype and fine-tune my design right in front of you, explaining my thought process as I go.
Attachment Design / Mechanical Alignment
I’ll show you how to construct ideal mechanical attachments (a.k.a. implements) for your robot so that you can accomplish FLL missions easily and reliably. We’ll design specialized Alignment Implements (bumpers, jigs, wedges, one-way-doors, etc.) to help your robot align against walls and mission models mechanically. With skillful implement design, you won’t have to rely on precise robot movements or accurate launches from base.
Learn how to write, debug and optimize programs in the LEGO Mindstorms EV3 language. I will show you how to use sensors to optimal effect without overcomplicating your strategies. Learn how to use gyro (gyroscopic) sensors, touch sensors, light sensors, color sensors and ultrasonic sensors to great effect (free hint: don't use ultrasonic sensors for FLL... they're too unreliable). I'll also show you how best to format, deliver and explain your programs to the judges.
This is one of the MAJOR differentiators between the teams who win top awards for Robot Performance and the teams who end up severely disappointed in their robot’s outcomes at the tournament. How to be the best robot handler possible. We’ll go over timing and prioritization strategies, and Michael will show you how to plan out and choreograph your movements during the round without losing your ability to pay attention and improvise. Michael will demonstrate his “Zen Robot Handling” method by choreographing his own robot handling movements on camera (using a demo setup) and showing you how he would train himself to optimize speed and consistency.
Strategy, Project Management, and Teamwork
Michael will lay out his overall approach to the robot game, explaining his most successful time management and prioritization strategies. He’ll propose timelines and objectives for you, but you’ll be setting your own priorities and deadlines. You’ll have a lot of difficult decisions to make (as a team), and Michael will help you learn to think through those decisions wisely. He’ll also propose methods for collaborating, splitting up tasks, giving useful homework, and generally optimizing your teamwork with the robot game.
This is the big one. It’s fun to talk about building and programming strategies, but you’ll be spinning your wheels and working in circles unless you can learn to test your designs EFFECTIVELY. I don’t just mean running programs to see whether or not they work… that’s not testing. I’m talking about the scientific method. And I happen to be trained in science, so I take methodology. I’ve conducted psychophysics research and published experiments in the peer-reviewed Journal of Vision, but the vast majority of my knowledge and experience comes from the thousands of hours I spent testing robots in ’05 and ’06 before those world championships. Kids usually fall into one of two traps with testing: either (1) they obsessively test their programs but don’t pay attention to the right details, and thus don’t actually learn much from their hours of testing, or (2) they don’t test their designs under the right conditions, thus their tests are invalid and they make bad decisions. In either case, kids will tend to get frustrated and just “go with their gut” instead. Not only does this result in poorer performance, but it short-circuits the learning process! Once your kids have learned to test like a true scientist/inventor, they will have graduated into the real FIRST LEGO League… the game of discipline and imagination that taught me almost everything I know. As an added bonus, they’ll be able to sleep soundly the night before the competition knowing that they’re ready.
Note: Robot Academy is not officially affiliated with FIRST or with LEGO. You don’t need to join our programs in order to compete in the FIRST LEGO League. FIRST provides new teams with basic resources including an explanation of the 2017 challenge.