What are the Key Ingredients for a Competitive Combat Robot?

What if I told you there exists a sport wherein a remote-controlled robot attacks a flame throwing drone with a rake, and ends up shredding it to pieces with a high-powered spinning blade?

The only correct response is ‘What a time to be alive!’ 😊

Obviously, the sport I’m talking about is combat robotics a la the TV Series BattleBots! A weekly show featuring a couple dozen teams of fellow engineering nerds competing to build the most dangerous machine of all. (Available on Discovery channel, but also on Amazon here for my fellow cord cutters.)battlebots_2018_lineupMy whole life I’ve enjoyed watching robots fight on TV. As a kid I’d be glued to the screen*, loving the creative robot designs and trying to imagine how I would do things differently if the robot were my own. I do the same thing now, except now I actually could build one.

Not to undermine the amount of effort those people put into their work. Heavyweight robots cost thousands of dollars and require hundreds of man-hours & technical expertise to bring to life…only to be destroyed in seconds.

On the other hand 1 lb and 3 lb class combat robots are a manageable size and make competing more accessible. As luck would have it moving to the west coast has brought me closer to fellow enthusiasts and I have recently discovered small battle events not far away!**

With that possibility in mind I’ve found it fun to think about what aspects make for a top notch battle bot. Aside from simply buying better quality components and maxing out the weight limit, I’m curious to know what design features produce the most effective machines.

So I’ve taken a closer look to identify the elements at play, recurring design themes, and tried to answer what are the key aspects of a top tier combat robot?

 To help answer this question, I watched a bunch of youtube videos and tried to find actual data if possible. Real data was much easier to find than I expected.

Google: Battlebots + wiki and get The BattleBots Wikia which presents a detailed history of every battle and every bot in every year of the competition. On top of that gold mine I stumbled across the BattleBot fantasy league website.  

I_found_gold

The fantasy league was interesting because it sorted all the robots in the 2018 lineup into tiers of expected performance based on past performance. Fantasy leagues are for gambling after all, so this was an excellent way to look at an expert opinion.  According to their data, these four robots below are the top contenders.

Battlebots_Tier_one

Every Interviewer: “What’s your strategy in the upcoming battle?”

Every Robot driver: “Well I’m going to target his vulnerable areas with my weapon, and make sure he doesn’t do the same to me.”

claymore
Simple in theory

Four Key Elements of Successful BattleBot Teams:

  1. The Geometry of Destruction- The overall shape of the robot matters.

All other things being equal, the physical geometry of the robot is a major determining factor of the chance that they will strike or be struck by a critical hit. Robots are difficult to drive so an important question to think about is:

If the robots were to move about the ring randomly, what is the chance of my robot weapon striking your weak point? Or striking the arena and damaging itself?

I color-coded the tier one bots to show what I mean:  Green = Weapon, Yellow = Armor, Red = Vulnerability.  When viewed through that lens it is easier to see how difficult it is for the robot driver thread the needle without getting pricked.

Battlebots_Tier_one_Colored

Tombstone’s advantage for instance, is that it is nearly impossible to approach without getting your ankles chopped off. But why leave the back exposed? It seems to be a calculated trade-off  to make the robot easier to drive which is an advantage because the arena and its hazards are a factor too.

A higher blade/body ratio than what Tombstone has results in a full body spinner, which is difficult to steer, likely to hurt itself on a wall, and useless in a corner.

Its tempting but I’ll hold back my criticism & not share color coded examples of robots with absurd geometries that are destined to be destroyed. I’m satisfied that they are there to be destroyed for my entertainment.

Anyhow, with the geometry factor in mind it then becomes clear that a major aspect of fights boils down to a game of:

  1. Rock- Paper-Scissors-Flamethrower: Some weapon types have natural advantages against others, while some weapons are wildcards. Advantages help but don’t guarantee victory.

The robots and their weapons can take whatever form the designer wants with a few obvious limitations. Also, since BattleBots wants to stay in business all bots must have at least one active weapon because passive-wedge-only robots are boring to watch!

Given that design freedom many kinds of weapons have been used, but it seems that the serious contenders focus on using one of 3 weapon types really well. The remaining weapon types are more difficult to reliably win with so I classify them as novelties.

These are the top categories as I see them, color coded with the same criteria as before:

rock_paper_Scissors-robot

All other things being the same:

  • The horizontal spinner*** is strong against the vertical back spinner because its attack area is so much larger. It is physically difficult for the vertical spinner to land a body blow without getting smacked in the process.
  • On the other hand, when the horizontal spinner smacks the armored slanted surface of a wedge bot the resulting force vectors contain a vertical component and the horizontal spinner is thrown off balance into the air, potentially striking the arena walls or floor afterward.
  • The wedge bot, so good at deflecting the horizontal blow presents a large juicy target for the vertical back spinner, which is really good at safely transferring its destructive energy into the wedge tossing it into the air.

The remaining weapon types all require very specific conditions for them to have effective damage potential. While there are examples of these bots winning, these weapon designs are inherently difficult to successfully employ. These are the novelty weapons in decreasing order of effectiveness:

  • Clamp/Grapple arm- Crush opponent or man-handle them into the arena hazards, BUT the opponent has to physically fit in your grasp and be movable.
  • Swinging Hammer/Axe- Pneumatically swing a concentrated blow in a small area, BUT accuracy is difficult even if the opponent was holding still.
  • Cutting blade (forward spinning)- Makes cool sparks and can shred tires, BUT the target must be held still long enough to dig through armor and do real damage.
  • Flamethrower- Looks cool BUT a properly shielded robot will take a long time to catch.
  • Little Buddies- Fun as though they are, tiny assistive bots or drones have yet to be the deciding factor in a match to my knowledge. Better to keep the main bot big.
Complete-Control-vs.-Bombshell-BattleBots
Apparently Grapple + Fire > Swinging Axe. That it actually worked was an awesome surprise

The rules allow you to switch weapons to prepare for a fight so clever teams bring a couple different weapon modules to adjust their bots as needed. But the fact that novelty weapons are difficult to use effectively is really important because:

  1. Reliability, Testing, & Control– Battle bots are difficult to practice with.

Heavyweight BattleBots are so big and dangerous to operate that some teams enter the ring with literally almost no experience driving their own robots!

When a robot has not been adequately tested it shows immediately. These are the robots that spill their guts after one modest smack, or have malfunctioning weapons out of the gate.

I think teams should ask themselves “could my bot destroy a washing machine without killing itself in the process? Could I overhand-throw a cinderblock onto the robot and expect it to function afterward? Can my robot self-right or operate upside down?” And lastly but most important, “Is my soft & squishy LI-PO battery pack adequately secured & shielded?”

white_Smoke
 The signature white smoke of punctured lithium battery pouch. 
  1. The Ironic Team Costume Trend: A silly observation, and yet…
  • Teams that put a ton of effort into making themed costumes do not usually fare well.
  • Teams that put no effort whatsoever into making costumes also perform poorly.

The folks to watch out for show up with a decent custom team shirt/jersey that is subtle enough to wear while grocery shopping without getting crazy looks.

Battle-Royale-With-CheeseBot-S2018-1140x760
I’m sure they are a lot of fun at a barbecue, but successful combat roboticists they are not. Also the thought of leaving a battle event in full garb after a major loss has me weak 🙂

In conclusion I leave you with a top 10 bot fights video I found that happens to support my theory.  But all that said, variety in robot design is part of the fun and it’s clear that driving skill and luck are significant factors in the fights too. The match ups this year have not been easy to predict, that’s what makes it fun to watch!

Notes:

*Using these robots to spark curiosity in children & draw them towards scientific interests is apparently a motivator for many contestants.  All the more reason to love and support the show!!

**Major west coast robotics events:  RoboGames in California and Western Allied Robotics in Seattle.

***It’s interesting to note the form factor similarities between the recurring champs of two different shows:  Tombstone of the American BattleBots, and Carbide of British RobotWars.   Also interesting that the weapon drive motor tombstone uses is of the same motor family (albeit 6p vs 12hp) that I used in my electric motorcycle conversion…jeez!

Advertisements

2 comments

What do you think?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s