Finding Balance, The Importance of Time Tables In Aerial Assist
An eventful kickoff day has passed, and the Ri3D teams are well into their design and build processes. There are many ways to tackle a game like this, but one thing that caught my eye today was a little table on Team AndyMark’s white board. On the top left of the board (shown above) you can see that they have dissected various scoring scenarios based on time and number of cycles that could be completed. In the past, teams I have been involved with have used this method with much success.
With any game, it is important to experiment, experiment, experiment. Team AndyMark used the human method as shown in the video below:
Its important to remember that humans have this little thing called a brain which allows for quick calculations to adjust the position of his or her extremities. This is why there is a major time adjustment on the table to account for the stiff movement and reaction time of competition robots. Even still, this is a good place to start on any given brainstorming day.
TO HURDLE, OR NOT TO HURDLE
A quick look at Team AndyMark’s table shows that the throw and catch element of the game is not as lucrative as it may seem, and it is possible. They estimated that a throw and catch over the truss would tack on a whopping (and apparently optimistic) 10 seconds per cycle. This could bring the number of complete cycles per alliance down to only 4 per match instead of an estimated 7 for not even bothering with the truss at all. The difference is 40 big points. This is a no brainer right?
BUT WHAT ABOUT DEFENSE
The problem with time calculation is that is easy to underestimate the role that simply having six robots on a field plays. Robots playing defense can very easily eat up five, to ten, or even longer depending on how fast or skilled they are. In a game with a wide open field, it is a common and very effective maneuver to “throw a shoulder” into a passing robot on your way back to pick up a ball, get into position, etc.. This action can spin a robot, cause a control hiccup, or even tip over a robot in an instant. The driver of the team taking the hit is forced to react and will often overcorrect or even freeze up at the controls as a result. This once again adds seconds to what should be an easy cycle. Then there is the simple matter of the goaltending robot. Say for instance a robot has a blocker and stops a shot from going through the high goal, now you have to regather yourself, plan a new course, and find a way to get yourself back on track. This takes a huge amount of time, and could wind up costing a team 40 major points in a round if it happens at the most inopportune times.
In the end, no scenario is going to be perfect, but there are factors involved that could help you make a decision. In terms of the throw and catch vs. don’t throw and catch scenario, there is the factor of something I like to call “points in hand”. Points for assists are based on a contingency of somehow passing the ball through one of the goals. If you don’t score the ball, you do not get the points. A successful throw and catch over the truss gets you the security of knowing that you have twenty points in hand regardless of whether the ball passes through the goal or not. Of course, there are other factors that play into it (i.e. difficulty of the maneuver, tradeoffs in robot design, etc…), but when all is said and done, time alone is not always the end all be all of existence for a particular strategy. My advice is to use that time table, but to use it smartly. Think about how the game slows down when you have five other moving obstacles around, and adjust your thought process accordingly.
Good luck on your decision making and don’t forget to keep that fridge stocked with Dew when arguments get tedious in your strategy sessions!