The Eternal (Bad) Golfer Problem

My cousin needed a science project. She didn't really have an idea on how to come up with an idea, so my uncle and I helped her work through brainstorming. Nearly all research comes from the desire to solve a problem. Granted, basic research isn't strictly goal-oriented, but you can always think of it as the problem of wanting to know something. She's a golfer (and a good one: she's on the high school varsity squad) and when you ask a golfer a problem they want to solve there is a universal one waiting in the ether: Where the hell did my ball go?

We came up with various technical possibilities. We looked at RFID and other remote sensing techniques. I considered various dyes (UV or something conductive perhaps), but it turns out that the United StatesGolf Association has a lot of rules including arcane things like "You can't do nuthin' to that little white thing what you hit with the stick" (I'm paraphrasing). So we rule those out (and, as it happens, such things already exist, if you are willing to drop a couple hundred for something that's banned anyway). That pretty much leaves the visual aspect of things. There are several devices that try to help you find your ball. Ranging from the gadgety to the elegant. There's even an app for that. The app and the gadget take pictures of the visual field and look for conspicuous white objects. That's not a particularly good solution thanks to leaves and garbage and things. You find a lot of things that are definitely not USGA approved. The polarizing lenses are a neat trick (elegant, like I said) which eliminate glare and reduce reflections to help you peruse the field a bit more. The problem with that is (and the picture gadgets as well), they assume you know where to look. I am not a good golfer and I have trouble getting to a place where I should be looking, so that's not particularly helpful.

Thus entered physics. Hitting a golf ball and determing how far it travels and where it lands in 3-dimensional space is essentially a ballistics problem, with very well known variables and data points. It turns out that the people who build golf simulators have pretty much nailed the formulas (nerds and geeks alike should check this page out regardless of how you feel about golf - it's pretty neat). So if the theory is done all that's left is: plug and chug (we miss you Doc Oc). In other words, it's a data collection problem.

We live in a brave new world where data collection is no longer the realm of the Research I university or those deep dark DARPA folk. So just spitballing here's what I think is needed to solve our ballistics problem.

1. You need to know a lot about what happens when you hit the ball. There are several apps for this, including iSwing and most intriguingly Zepp Golf Sense (and maybe eventually PocketPro - this one isn't out yet). These little guys are actual sensors. The Golf Sense is something you attach to your glove and the Pocket Pro slides onto the club itself. These will provide you with key data points: club head velocity. You will probably need to use something like iSwing as well since it works by photographing your swing, which you will probably need to get the angle the ball comes off the club. None of these are sophisticated enough to measure spin on the ball, but that too can be derived from the club strike information.

2. You are going to need some basic atmospheric data. Shaka is a neat device that plugs in to the head phone jack to measure windspeed (it was designed for surfers, actually). But for super accurate ballistics, you will need something like the Brunton ADC which will give you the atmospheric pressure and, crucially, air density. Amazon has a couple niftier ones, but Ambient Weather makes a wireless one that posts data to the web. We can enter data by hand at first, but web capability is the dream.

3. Golf nerds also have one more incredible tool at their disposal, thanks to their eagerness to depart with their hard earned money in pursuit of that perfect game: 3D maps linked to GPS locators. And lo, there was GolfShot, and the Lord called it good. There are thousands of courses in here (they even offer to map out courses they don't already have in two days or less!). And, natually, it has an app component. Tap it to let the GPS in your phone (which you are already using to photograph your swing) know where you are and off you go.

Between these three components, you have the beginnings, the germ, the zygote (well, let's call it a blastocyst) of an idea that will help you find your ball. Your weather station, iPhone (or Android phone, whatever - Windows phone, pah!), and sensors are all ready to go. You line up and take your swing and you lose your ball in the sun or whatever. The sensors all feed to a web app which performs the ballistics calculation and sends the result as a GPS coordinate (naturally, it already knows your present location) to your GolfShot 3D map of the golf course. It throws up a compass, a direction arrow, and off you go. And thanks to GolfShot, you will know instantly whether it nailed a hazard or careened out of bounds, because why waste time looking for a ball in the wet?

Yes, it's clunky, what with the various apparti involved. And it will practically require you to have a caddy or at least golf when no one else is behind you. For now. In a perfect world, the club sensor would tell you about angle and spin without having to photograph it, but that's for version 2.0, right? And it just might not be terribly easy to predict with enough accuracy and precision where you ball ended up. But hey, in physics world, you know the one where friction and air resistance don't matter, this is a slam dunk deterministic problem with an arbitrarily precise calculable answer. That's not happening, but hell, I'd settle for a standard deviation of a yard and two sigma certainty. That's an improvement over this golfer's naked eye.

What do  you think? Is it time to Kickstarter?

Photographic Evidence

I'm not sure if I am proud of what I accomplished in 8 hours or ashamed that it took me the best of that 8 hours to do this to my CV axle:



I don't even want to talk about how many abrasion discs I went through on my (borrowed) Dremel.

The lesson here is: sometimes the hard part about repair isn't the repair itself, but the things that don't work in the way they are supposed to. On my barn find, it's a constant encounter with rust and rigid parts that have a death grip on themselves. Can't wait to see what else needs the Dremel treatment. No seriously. Ok, sort of, not really, seriously . . . not at all.

Even so, I prevailed in the end: the part was changed. It took more time and didn't cost significantly less (initial cost of tools is actually pretty high, over the life of my car ownership, I expect that to pay dividends though) . . . but now I know a lot more about my car and myself and that's what this is all about.