Geocaching search strategy

October 27, 2006

Dear Geocachers,

We are a group of students with the task to develop an algorithm for geocaching. This algorithm should result in a sequence of caches with a cumulative weight as high as possible. Of course, there is a time constraint, so we look for an algorithm (= search strategy) that yields the highest score in a limited period of time (e.g. 1 day).

So, what would experienced Geocachers like you do? Would you go after the caches with the highest weight, or the caches closest to each other (minimizing travel time between 2 subsequent caches), or those having the shortest search time on the spot? Or do you have other strategies to win such a geocaching competition?

All ideas are welcome!

24.7k · October 27, 2006

Dear Geocachers,

We are a group of students with the task to develop an algorithm for geocaching. This algorithm should result in a sequence of caches with a cumulative weight as high as possible. Of course, there is a time constraint, so we look for an algorithm (= search strategy) that yields the highest score in a limited period of time (e.g. 1 day).

So, what would experienced Geocachers like you do? Would you go after the caches with the highest weight, or the caches closest to each other (minimizing travel time between 2 subsequent caches), or those having the shortest search time on the spot? Or do you have other strategies to win such a geocaching competition?

All ideas are welcome!

The first thing I did was split my team up and find them all at once.

I suspect you need to give us a little more detail. You want to create a geocaching competition? Using what? New caches? Existing caches? For an event? Winning in a day? In a Week? Over time? Cumulative? Etc.

Edited October 27, 2006 by Renegade Knight

October 27, 2006

All ideas are welcome!

If there was a significant cash prize involved, I'd charter a helicopter to lower me onto the coordinates of each cache.

On a more serious note, I always try to find a path that loops through all the caches using the least amount of gas for my vehicle. If fuel is not an issue, I go for the hardest first because I can't find hard caches when I'm exhausted.

October 27, 2006

If I understand correctly, the "winner" would be the cacher with the highest "points" at the end of the day? There is no easy answer for this question.

How are they weighted? I would want to know how many lower ones equal a higher one. This would help determine if the effort for several lower ones is worth it. Example...if 2 lower equal 1 higher, then I am scooping up the lower ones as fast as I can. If 5 to 1, probably could go either way. If 10 to 1, then I am going for the higher ones.

Are they variably weighted? Maybe I would go after 3 lower, 2 medium, and 1 high. Or maybe a bunch of medium ones.

good luck!

3.2k · October 27, 2006

Dear Geocachers,

We are a group of students with the task to develop an algorithm for geocaching. This algorithm should result in a sequence of caches with a cumulative weight as high as possible. Of course, there is a time constraint, so we look for an algorithm (= search strategy) that yields the highest score in a limited period of time (e.g. 1 day).

So, what would experienced Geocachers like you do? Would you go after the caches with the highest weight, or the caches closest to each other (minimizing travel time between 2 subsequent caches), or those having the shortest search time on the spot? Or do you have other strategies to win such a geocaching competition?

All ideas are welcome!

Edited October 27, 2006 by baloo&bd

October 28, 2006

What do you mean by weight? Although there are difficulty/terrain ratings, all geocaching.com finds 'count' as 1 smiley (find).

2.5k · October 28, 2006

I'm guessing you guys watch NUMB3RS. :laughing:

As for me I search for caches along routes I would take for other reasons 80% of the time. I cache near places I go on other business. I also occasionally take a dedicated caching trip, in which case I am looking for a good number of caches in a suitable hunting area, my preference being parks and recreation areas with hikes to find the cache.

Or do you have other strategies to win such a geocaching competition?

Geocaching itself is not a competition. We do have what some may call a score, (our find counts) but there is no organized means of determining who is "winning." There is a lot of variations in how people get their find counts, and there are some cachers with inaccurate counts due to logging caches by, let's put it nicely...."creative" definitions of a "find."

14.3k · October 28, 2006

Dear Geocachers,

We are a group of students with the task to develop an algorithm for geocaching. This algorithm should result in a sequence of caches with a cumulative weight as high as possible. Of course, there is a time constraint, so we look for an algorithm (= search strategy) that yields the highest score in a limited period of time (e.g. 1 day).

So, what would experienced Geocachers like you do? Would you go after the caches with the highest weight, or the caches closest to each other (minimizing travel time between 2 subsequent caches), or those having the shortest search time on the spot? Or do you have other strategies to win such a geocaching competition?

All ideas are welcome!

If I were you guys I would get a good map of all the caches in Palm Springs. Then I would develop a plan/strategy based on your scoring system. Then I would implement the plan in the caching mecca of the west!

Edited October 28, 2006 by TrailGators

October 28, 2006

All right, some more details.

So we have been given a list of about 1500 caches. Each cache of course has two coordinates (latitude and longitude), a weight (i.e. the value of that cache: some caches are worth 1 point, others 2, some even 5, ...) and a length (i.e. the time (in hours) one would need to find the cache after having arrived on the spot with the given coordinates).

Now we play a contest with a number of one-person teams. The person (which travels on average at 60 km/h between 2 caches) that has collected the highest cumulative weight (i.e. the sum of the values of all found caches) within 24 hours, has won. The available time is used to (i) travel from one cache to another and (ii) to look for a cache when having arrived at the given coordinates (= length of a cache). You immediately see that going after the caches with the highest weights is not necessarily the best choice, as it might be better to go after caches with lower weights but also lower distances between them (less travel time)...

We have to come up with a strategy (or an algorithm = a decision criterium (based on the weights, lengths and distances between caches) that will be used each time one has to decide what will be the next cache he/she goes after) that at the end of the day yields a total score as high as possible.

Examples: going after the cache that is nearest to the cache you have just found, choose as next cache the one with the highest ratio 'weight/length', etc.

I hope the problem is more clear now. All strategic insights are welcome (but please try to remain serious :-))!

7.2k · October 28, 2006

All right, some more details.

So we have been given a list of about 1500 caches. Each cache of course has two coordinates (latitude and longitude), a weight (i.e. the value of that cache: some caches are worth 1 point, others 2, some even 5, ...) and a length (i.e. the time (in hours) one would need to find the cache after having arrived on the spot with the given coordinates).

Well, as I see it, there are only two ways to "score" a cache.

One is all caches are equal in value in that they all increment your find count. In this regard there can be no weighting at all, each cache is worth the same number of points.

The other is the entertainment factor of the cache. Let's forget about the subjectivity involved with such an attribute and look at something else. The weighting would have to go from 0 to 5, not 1 to 5. The reason being a 1 value will give the cache some sort of value when it reality the cache may have absolutely no entertainment value at all. By starting the range at zero, then you would be closer to the way some cachers approach caching, completely ignoring zero value caches and concentrate on caches with high entertainment value. Those will continue to produce the entertainment factor down the road as the cacher looks back on their adventures.

A major problem comes into play when you understand one cacher's "point system" may be the inverse of another.

My point system may favor highly difficult physical challenges, a well constructed mental challenge, and would take a long time to complete. The other end of the scale are drive-by caches with no challenge, twist, interesting location, or entertainment value.

Conversely, another cacher's point system may rate the quickie drive-by much higher than the mountain-top all-day hike which they avoid like the plague.

Unfortunately, there's no easy way, if at all possible, to definitively quantify something that is such a moving target.

Edited October 28, 2006 by CoyoteRed

9.3k · October 28, 2006

If I grasp your concept correctly, I think we need more data. Apparently you've already established a point system which you refer to as "weight". How you reach the figure out the "weight" of each cache would be critical in determining a strategy. What constitutes a 1 point cache, a 2 point cache, etc? Does the size of the container play a role, or is it based solely upon the difficulty and terrain ratings? Also, it looks like you might have left out an important factor similar to what you refer to as "length", and that is how long it takes you to get close enough to hunt for the cache. There are some caches that take hours, (days?) to get to, while others would only take 20 minutes or so.

October 28, 2006

I'm confused. Or you're confused. Or both.

I gather from your profile, that you are not a geocacher (yet), and have never found a cache. I also assume that you are referring to a class project assigned to you. So, maybe this well help.

Is this WEIGHT something your instructor has given you? There is no weight factor in geocaching. The numbers 1-5 listed with caches on geocaching.com are for FINDING DIFFUCULTY and TERRAIN DIFFICULTY. Maybe that is what you are referring to.

To see how the Difficulty levels are determined, try this link

http://www.clayjar.com/gcrs/

and press [rate this cache]

Regarding the LENGTH. Is this something the instructor supplies you with. There is no length (time) factor assigned to caches. Maybe you mean terrain difficulty?

Now, with this in mind and with my limited experience in finding only 130 so far, I suggest you would stick to single stage caches

http://www.geocaching.com/about/cache_types.aspx

that are medium size, difficulty 1 or 1.5, with a terrain level of 1 to 2.

Such caches should be quick, easy finds. Difficulty 3 caches sometimes take hours or days to find. (For me).

I also wonder if you will be using the geocaching.com website. If so, there is a map feature that you can use to see where the caches are in relation to one another. If not, how were you supplied with the 1500 cache list? That's puzzling to me. And, what formula will be used to determine the actual score? There are still a lot of questions. It would be helpful if you could supply us with the actual written instructions given by your instructor.

Finally, I have only found about 130 caches so far. There are others who have found THOUSANDS. These addicts geocache every chance they get, often spending entire weekends on the sport. They often find 10 to 20 caches in a single day. I'm sure one of them would love to give you advice.

Good luck.

Edited October 28, 2006 by michigansnorkeler

October 28, 2006

These addicts geocache every chance they get, often spending entire weekends on the sport. They often find 10 to 20 caches in a single day.

You rang? :laughing:

October 28, 2006

All data of caches that I mention (weigth, length, coordinates) are all given by our instructor. Whether a weight is related to entertainment value or the like, is not our concern. We are only interested in gathering a score as high as possible within 24 hours, by following a smart strategy or algorithm.

Try not to interpret the concepts of weigth and length of a cache, but accept them as given. The problem might be slightly different from what geocaching is in real life.

October 28, 2006

Ok, this is the problem as we have been given it:

1. General

With the advent of geo positioning systems (GPSs) during the last 5 years, it has become possible to generate a relatively precise estimate of one’s position, or of one’s car position, or, in fact, of any item that is equipped with a gps. This has created new opportunities in different domains. Apart from the obvious application in driving systems in cars and trucks, there has also been an uprise in games that rely on GPSs. A typical example of such a game is called geocaching, see for instance www.geocaching.com. In such a game the goal is to find a treasure, called a cache, that has been hidden somewhere; the only known aspect of the treasure is its location given by the coordinates of that location. Knowing the location, however, does not imply that the cache is easy to find! Indeed, the inherent imprecise measurement of a GPS (which can determine a position within a radius of about 10 meters), and the smart ability of the one who hid the cache, may result in a challenging search for the cache. There is a community of people that searches for these caches.

2. The contest

In order to celebrate the 5-year existence of the geo-caching community in Belgium, a contest is being organized. You, as an experienced geocacher, want to participate, and in fact you see yourself as a potential winner. The contest is as follows. In Belgium over a 1500 caches are hidden. To each cache a weight is associated, and to each cache a length is associated. Furthermore, for each pair of caches a travel time between the two is known. The goal of the contest is to visit, within 24 hours, a sequence of caches with maximal total weight. Notice for a solution (= a sequence of caches) to be feasible, the sum of the travel times between consecutive caches together with the sum of the lengths of the caches, should not exceed 24 hours.

-----------------

We were also supplied with a list of 1500 caches; every cache in this list has a given weigth, length and set of coordinates.

Our task is to come up with a strategy (= a decision criterium used to decide which cache you will visit next (e.g. the closest cache, the cache with the highest ratio 'weigth/length', ...) to get a score as high as possible.

October 28, 2006

Sounds like a new way of presenting the old travelling salesman problem. Here is the wikipedia take on it.

14.3k · October 29, 2006

Ok, this is the problem as we have been given it:

1. General

With the advent of geo positioning systems (GPSs) during the last 5 years, it has become possible to generate a relatively precise estimate of one's position, or of one's car position, or, in fact, of any item that is equipped with a gps. This has created new opportunities in different domains. Apart from the obvious application in driving systems in cars and trucks, there has also been an uprise in games that rely on GPSs. A typical example of such a game is called geocaching, see for instance www.geocaching.com. In such a game the goal is to find a treasure, called a cache, that has been hidden somewhere; the only known aspect of the treasure is its location given by the coordinates of that location. Knowing the location, however, does not imply that the cache is easy to find! Indeed, the inherent imprecise measurement of a GPS (which can determine a position within a radius of about 10 meters), and the smart ability of the one who hid the cache, may result in a challenging search for the cache. There is a community of people that searches for these caches.

2. The contest

In order to celebrate the 5-year existence of the geo-caching community in Belgium, a contest is being organized. You, as an experienced geocacher, want to participate, and in fact you see yourself as a potential winner. The contest is as follows. In Belgium over a 1500 caches are hidden. To each cache a weight is associated, and to each cache a length is associated. Furthermore, for each pair of caches a travel time between the two is known. The goal of the contest is to visit, within 24 hours, a sequence of caches with maximal total weight. Notice for a solution (= a sequence of caches) to be feasible, the sum of the travel times between consecutive caches together with the sum of the lengths of the caches, should not exceed 24 hours.

-----------------

We were also supplied with a list of 1500 caches; every cache in this list has a given weigth, length and set of coordinates.

Our task is to come up with a strategy (= a decision criterium used to decide which cache you will visit next (e.g. the closest cache, the cache with the highest ratio 'weigth/length', ...) to get a score as high as possible.

Each geocache has many factors that you need to know. For example, each cache has two user assigned ratings that will be critical for your success. The first rating is terrain rating. The scale goes from 1-5 in increments of 0.5. A terrain of 4 is typically a long hike up a tall mountain and will take several hours to find. A terrain of 1 is wheel-chair accessable and is the easiest to get to. The other is difficulty of finding which uses the same scale. Difficulty indicates how difficult it is find the cache. Many caches are hidden with great cammo and can be very tough to find. There are also different sizes and types of geocaches. So you need to create a database of the 1500 caches that show which ones have the highest value assigned by your professor. Then you can link that database with a table that includes all the key factors of each of the 1500 geocaches assigned. Then you can come up with an algorithm to assign an overall difficulty of each cache. After you do all that then you can plot a cache route using length, weight and difficulty.

I would also highly recommend that you do some geocaching for homework to better understand the factors! :blink:

I did some homework myself today!

Edited October 29, 2006 by TrailGators

October 29, 2006

All data of caches that I mention (weigth, length, coordinates) are all given by our instructor. Whether a weight is related to entertainment value or the like, is not our concern. We are only interested in gathering a score as high as possible within 24 hours, by following a smart strategy or algorithm.

Try not to interpret the concepts of weigth and length of a cache, but accept them as given. The problem might be slightly different from what geocaching is in real life.

The attribute you call "weight" is very important. If the weighting is given to level of difficulty, or terrain, then I'd say spend more time searching for the caches with higher difficulty or terrain ratings. If weight is given to something arbitrary like the colour of the cache container, then I'd say don't waste your time on the higher difficulty/terrain caches.

Assuming all caches have an equal weight of one (as they do in real life), then I'd say DON'T look for Multi-caches (many stages) or Mystery caches (requires puzzle solving) as you will spend a lot of time on these. I would suggest you look on the map for clusters/pockets of many Traditional caches, and do all the easy difficulty/terrain ones.

October 29, 2006

so basicly this has nothing to do with real geocaching... it is just a fictional math problem.

I think that is where most of the confusion was coming from.

The question you are asking us actually has nothing to do with geocaching (no more than the ole "two trains traveling in opposite directions" has to do with actual station masters.)

That being said, I would proceed with your math problem by determining the weight/time ratio of each cache so you know which caches are worth it. Then focus on an area that has the highest concentation of high weigh/time ratios that are centrally located.

14.3k · October 29, 2006

All data of caches that I mention (weigth, length, coordinates) are all given by our instructor. Whether a weight is related to entertainment value or the like, is not our concern. We are only interested in gathering a score as high as possible within 24 hours, by following a smart strategy or algorithm.

Try not to interpret the concepts of weigth and length of a cache, but accept them as given. The problem might be slightly different from what geocaching is in real life.

The attribute you call "weight" is very important. If the weighting is given to level of difficulty, or terrain, then I'd say spend more time searching for the caches with higher difficulty or terrain ratings. If weight is given to something arbitrary like the colour of the cache container, then I'd say don't waste your time on the higher difficulty/terrain caches.

Assuming all caches have an equal weight of one (as they do in real life), then I'd say DON'T look for Multi-caches (many stages) or Mystery caches (requires puzzle solving) as you will spend a lot of time on these. I would suggest you look on the map for clusters/pockets of many Traditional caches, and do all the easy difficulty/terrain ones.

High terrain caches take a lot of energy to hike to. Most people can only do a handful of these in a day. If a handful of superhigh weight caches is all you need, then that strategy might work. But it is also possible to find 100 or more easy urban caches in a day. However, without understanding your weighting system there is no way for me to predict which strategy is best. But this is the reason for your assignment....

14.3k · October 29, 2006

so basicly this has nothing to do with real geocaching... it is just a fictional math problem.

Actually he/she said that "the goal of the contest is to visit, within 24 hours, a sequence of caches with maximal total weight." So it does have a lot to do with geocaching! :blink:

3.8k · October 29, 2006

I have been involved in several geocaching events that involve finding caches that have a point value assigned to them and also to orienteering events where stations have points assigned to them. In my recollection the winner of the events always use the same strategy. They plot out the spots with point values showing. They then figure out the shortest route to find all of the high point locations. They travel this route at a pace that will allow finding all the high values locations in the allotted time. They then deviate to lower point locations as time allows without going over the time limit.

So for instance if you have 24 5 point locations you get 1 hour for each location. If the time it takes to travel from location 1 to location 2 is 45 minutes you have 15 minutes to find as many lower value locations as you can. It seems like the value of the lower value locations really doesn't matter so much if you find all of the higher value caches. Also it may be required to throw out one or two "outlayers" if you can't fit them in.

3.7k · October 29, 2006

Webscouter is right on.

Except for the GPS component, the described task resembles orienteering events called a 'score-o' (time limits of 1-3 hours) or a 'rogaine' (typically 12- or 24-hour events). For each, the goal is to find as many control locations as you can in a prescribed period, balancing the varying 'weights' (points) of each control against the time required to get to it and find it. Participants have to factor in things like the terrain, distance, climb, difficulty of the actual hide, etc.

You could indeed create a geocaching 'course' similar to an orienteering score-o course. I think the idea is more novel for geocachers than orienteers, but it's an interesting concept.

You might want to visit one or two of the orienteering sites and get a sense of that activity, as what you describe is potentially a blending of the two.

Edited October 29, 2006 by hydnsek

October 29, 2006

There is a total travel time and length for the 1500 caches.

There is a total weight for the 1500 caches which I assume is the final score.

Plot a route to all caches using lowest Total Travelling time.

List the Travelling time in order 1-1500

Beside the Travel time list length of each cache.

3rd column list weight of cache.

You should have listed the shortest possible route from cache to cache. Adding the TT and length you should now be able to scan the list in blocks of TT and Length that Total time is less than or equal to 24. The block with the highest weight score will give you the starting point of a final view where you can then play.

October 29, 2006

Thank you already for all your contributions!

I certainly have some new ideas now, and discuss them with my group members. In the meantime, new ideas are always welcome!

October 29, 2006

Yup, I surmise that this is simply a math problem.

ACTUALLY searching for these geocaches will not be done, is that correct?

Rather than geocaches, it could be visits to cemetaries. (hmmm...sounds like geocaching). But the instructor is trying to make it more interesting by using geocaching.

You are only interested in time and weight. The most weight in 24 hours.

You might have more luck with this problem contacting a mathematics forum.

But, welcome to geocaching nonetheless! Hope we have peaked your interest! Hope you get hooked, buy a GPS, and visit us in the woods.

And, if anyone asks you what you're doing there, just tell them that you are using multibillion dollar military defense satellites to find and investigate objects (tupperware?) that unknown individuals are hiding throughout the nation, and around the world!

Edited October 29, 2006 by michigansnorkeler

October 29, 2006

That's right: the search will not actually be carried out, and we are only interested in weigths and times.

Nevertheless, I had never heard of geocaching, but it seems to be a very nice hobby!

14.3k · October 29, 2006

Nevertheless, I had never heard of geocaching, but it seems to be a very nice hobby!

You should try it! It is a lot of fun! :blink:

Geocaching search strategy

Recommended Posts

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Join the conversation