|
Hi,
I need to find a goal given the following conditions
-I only know my 4 surroundings (N,W,S,E): they can be empty or an obstacle
-I know the goal coordinates (x,y)
-I know my coordinates
I know bug-2 (and bug-1) are complete algorithms (they always find the goal).
But it is really bad in some maps with large obstacles.
My questions are two.
1. How come there arent quick and well-known solutions to this? Ive been googling for ages and I just find theoretical papers, full of blabla and no pseudocode.
2. A* wouldnt work for this because of my first restriction, right?
Thanks
|
|
|
|
|
I'm sure A* would work - all it requires is to know if a neighbouring tile is walkable or not, and then a cost for moving there. (Sidenote: you could substitute this for simply assigning a cost to moving across each tile and make impassable obstacles have infinite cost?)
Anyway, here's a good A* link: http://www.policyalmanac.org/games/aStarTutorial.htm
This is 8-directional, but it will work exactly the same in 4-directional.
|
|
|
|
|
Hi
You can represent this as a Graph-problem(some kind of shortest path algorithm)
You can apply these restrictions while moving
1 . Trace your current path, if the new slot is in the current path ignore that to avoid a loop
2 . Mark the positions in which no further movements are possible so that we can ignore that.
3 . If we have no further move retract your move using a stack
Back-tracking may not be an optimal solution, but you can apply some restrictions to improve it
Thanks
|
|
|
|
|
Compression is the key; it was mentioned on the page you linked to. MPEG is JPEG for moving images. What you do is trade in a lot of bandwidth for a gigantic lot of computations, and throw away some of the information (just like JPEG does).
|
|
|
|
|
Right... so that in reality you get bit transfer rates in the range of 1 MBit/s (bad quality, small size high-def you tube video) up to 30-40 MBit/s (high quality streams as found on blu ray discs). Still orders of magnitude below those 44MB/s you've found out. Most availbale HD video content is compressed using the H.264/MPEG4 AVC method (go google)...
|
|
|
|
|
I am trying to develop a couple of applications showing data on a global map (actually one is for plotting a planetarium). The data I have is raw lat/long coordinates.
I'm new to map/GIS systems but would like to develop a system 'from scratch' without using third-party components for the calculations and rendering (It's the best way to learn).
My question is does anyone know a good source of information / a tutorial (book, web-site, etc) as an introduction to this kind of application?, preferably with examples in pseudo-code, c, c++, c# or java.
The projection algorithms seem fairly straightforward (although picking which projection to use seems to be a mine-field), but I have other questions that I would like to be able to answer such as:
- clipping and optimisation
- plotting straight lines (great circles) and how that is done in a particular projection
- checking if a point is within a complex polygon
Thanks in anticipation
Ian
|
|
|
|
|
all i know is that Opengl is best for 3d rendering and sdl for 2d.
if you have all the data and only need the rendering.
I personally use Tao framework(Opengl for .net framework) and SdldotNet(Handles screen rendering,keyboard and mouse input).
Tao framework and SdlDotNet both got tools to integrate 3d/2d rendering on your windows application.
All you need to know about opengl http://fly.cc.fer.hr/~unreal/theredbook/[^]
|
|
|
|
|
Your problem breaks down into two parts
1. Datum
This is the basic shape of the earth that is used to model the earth (or any planet). This can be very simple such as treating the earth as a sphere or more complex as a non-spherical object etc. The parameters that make up the model are the datum's, and there are many different datum's, remember there can date back hundreds of years. The most common is WGS84 as used by positioning systems such as GPS. Most modern charts are based on this datum. But not all!
The model also includes what the measurements are referenced to for example Vertical datum can be referenced to sea level at a specific location perhaps using High water springs or Neaps, or an average such as Mean High Water Springs etc!.
If you have data in one datum but are using another then welcome to a world of pain! There are various algorithms available for converting between then (such as Molendsky), check out the US NOAA web sites but accuracy varies. For some datums there are no known accurate conversions.
2. The projection - linked to the datum but deals with how a spherical object is projected on to a flat surface Mercator, Transverse Mercator (TM), Genomic, Lambert, Polar etc.
So you end up with a real mash of things including grids (UTM), military grids, grids for national co-ordinates (OS in the UK etc).
If you are not too concerned about accuracy (planetarium) i.e. working with small scale data (say an accuracy of a couple of miles). Then datums are going to be less important - and you might get away with treating the planet as a sphere!
If accuracy is required seriously think about using a third part package, or prepare for a lot of learning and re-inventing the wheel work!
|
|
|
|
|
hello.
i would like to get help on this book: ebook Cormen, Leiserson, Rivest - Introduction To Algorithms 2nd Edition Solutions (Instructors Manual)
i would appreciate very much if someone has a link to this or an pdf version?
THanks!
best regards.
|
|
|
|
|
Did you try this[^]?
It's time for a new signature.
|
|
|
|
|
Hi Richard, thanks. but i was referring to an pdf version, an ebook, or smth like that
best regards.
|
|
|
|
|
Did you actually try any of the links?
It's time for a new signature.
|
|
|
|
|
i guess one of the links worked 
best regards.
|
|
|
|
|
Hello,
I have some facts like:
asdf(a,b).
asdf(b,c).
asdf(c,d).
etc.
I'd like to do some function
givemelettersinvolved(L) that instantiates in L a list that contains the constants involved in those facts
for example
asdf(a,b).
asdf(b,c).
givemelettersinvolved(L) would give L = [a, b, c]
How can I make this?
I was trying with
...(L) :- asdf(A,B), member(A,L), member(B,L).
But its not working
|
|
|
|
|
wtf are you talkin about
anyway the answer was to use findall(X, predicate(X), List)
|
|
|
|
|
hi there,
I'm looking for a code snippet for conjugate gradient and preconditioned conjugate gradient iterative solver for system of linear equations in C++ or C#.
|
|
|
|
|
if I were needing that, this[^] is one of the places that I would look at; it holds lots of links.
|
|
|
|
|
|
Anyone knows of any math software that has unlimited precision (i.e., 123^123,456,789 mod 987,654,321)? I'm looking into Mathematica Home Edition, but I'm not sure if it does that. I know Maple could calculate PI to as many digits as you want if you were willing to wait, but I'm not sure if that applied to all functions in general or just built-in ones.
|
|
|
|
|
Most, if not all, of the Mathematica/Maple/Mathcad type packages have an arbitrary precision integer and/or rational number capability. Check their online docco and/or user forums. If you're looking for a language to write your own code, Java has BigInteger built in, and there are free packages for other popular languages. (I have used a couple of the C++ ones - the quality does tend to vary a bit, but the good ones are good.)
Software rusts. Simon Stephenson, ca 1994.
|
|
|
|
|
Thanks, forgot about MathCad which I used to love. Does the BigInteger support ridiculously large numbers (i.e., unlimitedly large)? You know any for C#? I downloaded VS2010 to check it out but that think sucks big time.
|
|
|
|
|
Java BigInteger (and any other reasonable implememntation) will grow until you run out of memory (or patience  ). I don't play C#, but a quick google found C# BigInteger Class[^] right here on CP. Re VS2010: I've seen so much suckage reported here I'm not going anywhere near it!
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994.
|
|
|
|
|
That article describes a private BigInteger attempt on an old .NET version (which did not come with a built-in one); since 4.0 .NET has its own, pretty good, BigInteger class, probably very comparable to the venerable Java one.
|
|
|
|
|
Since .NET 4.0 C# has BigInteger too.
|
|
|
|
|
Thanks, Luc. As I noted above, I don't inhabit those realms!
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
