comparative analysis of path finding algorithms
Posted by mwojcik
|
comparative analysis of path finding algorithms July 11, 2015 08:39AM |
Hello guys
I am writing to you because I need your help. I’m doing my master thesis which topics is Comparative analysis of algorithms for path finding in 3D printing. Mainly, in this thesis I compare some algorithms which fill the layer. I don’t care about whole 3D mode, but just one layer. Right now I got several path finding algorithms such as: ZigZag, Spiral, Contour and two fractal algorithms, with Hilbert curve and Peano Curve. After that I try to optimize those paths, which was found by path finding algorithms, by optimization algorithms. I got just two of them: Ant colony optimization and Genetic Algorithm. I also wont to expend the group of optimization algorithms by Neural network and add maybe some fractal algorithm to the group of path finding algorithms. Mainly the zigzag strategy is the best one. (it depends about the shape of the printing area). Rest are really week compare to zigzag strategy. That’s lead me to the questions. First one: Do you know some other path finding algorithms which I can use in my master thesis? Second question is more general. What do you think about this project.
I am writing to you because I need your help. I’m doing my master thesis which topics is Comparative analysis of algorithms for path finding in 3D printing. Mainly, in this thesis I compare some algorithms which fill the layer. I don’t care about whole 3D mode, but just one layer. Right now I got several path finding algorithms such as: ZigZag, Spiral, Contour and two fractal algorithms, with Hilbert curve and Peano Curve. After that I try to optimize those paths, which was found by path finding algorithms, by optimization algorithms. I got just two of them: Ant colony optimization and Genetic Algorithm. I also wont to expend the group of optimization algorithms by Neural network and add maybe some fractal algorithm to the group of path finding algorithms. Mainly the zigzag strategy is the best one. (it depends about the shape of the printing area). Rest are really week compare to zigzag strategy. That’s lead me to the questions. First one: Do you know some other path finding algorithms which I can use in my master thesis? Second question is more general. What do you think about this project.
|
Re: comparative analysis of path finding algorithms July 13, 2015 04:39AM |
Sounds like a lot of math 
What is it you want to optimize?
Strength, speed, looks?
For strength, a 3D-honeycomb could be better than zig-zag? But it yet hasn´t been implemented AFAIK. ( only 2D honeycomb )
I tried to implement skin pattern printing once, but didn´t get far.
-Olaf
What is it you want to optimize?
Strength, speed, looks?
For strength, a 3D-honeycomb could be better than zig-zag? But it yet hasn´t been implemented AFAIK. ( only 2D honeycomb )
I tried to implement skin pattern printing once, but didn´t get far.
-Olaf
|
Re: comparative analysis of path finding algorithms July 13, 2015 09:25AM |
Quote
Slic3r 1.2.9 stable is out!
It includes real-time incremental slicing, 3D preview, toolpaths preview, 3D honeycomb, customizable bed shape, integration with OctoPrint, pressure regulation and much more!
Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse
|
Re: comparative analysis of path finding algorithms July 14, 2015 05:03AM |
Cool, they must have read my mind
|
Re: comparative analysis of path finding algorithms July 28, 2015 10:25AM |
Quote
mwojcik
Hello guys
I am writing to you because I need your help. I’m doing my master thesis which topics is Comparative analysis of algorithms for path finding in 3D printing. Mainly, in this thesis I compare some algorithms which fill the layer. I don’t care about whole 3D mode, but just one layer. Right now I got several path finding algorithms such as: ZigZag, Spiral, Contour and two fractal algorithms, with Hilbert curve and Peano Curve. After that I try to optimize those paths, which was found by path finding algorithms, by optimization algorithms. I got just two of them: Ant colony optimization and Genetic Algorithm. I also wont to expend the group of optimization algorithms by Neural network and add maybe some fractal algorithm to the group of path finding algorithms. Mainly the zigzag strategy is the best one. (it depends about the shape of the printing area). Rest are really week compare to zigzag strategy. That’s lead me to the questions. First one: Do you know some other path finding algorithms which I can use in my master thesis? Second question is more general. What do you think about this project.
Firstly, you have mentioned a couple of algorithms .. are these open source or closed? Because most suggestions are available for opensource. Your project is interesting because any math problem is good enough for me!!
Edited 1 time(s). Last edit at 07/28/2015 10:26AM by cosmorock.
|
Re: comparative analysis of path finding algorithms July 30, 2015 05:09PM |
First of all, thanks for replay. Yes there are a lot of math formulas and wired thinks.
About this opensource, so the idea of the algorithms is well known and mostly it is taken from some research paper (in google scholar you can find a lot of them). So i am not sure if there are some patents for those algorithm, but there could be. For my work, i don't need their code but just the idea of algorithm. I need to learn them and know how they works.
About this. I am trying to optimize the speed of printing. But of course, i will test those algorithm and by the way check some different things like strength, accuracy. Abut this honeycomb.Firstly, thanks for this idea. I heard about this and I didn't check it because i thought that it is not a path planning algorithms. I thought that it is just a shape which is applied to layer to print less material. And the paths, for extruder, are taken from some different algorithm. I don't know it for sure but i will check it.
When i finish those research i could give you my feedback abut them.
About this opensource, so the idea of the algorithms is well known and mostly it is taken from some research paper (in google scholar you can find a lot of them). So i am not sure if there are some patents for those algorithm, but there could be. For my work, i don't need their code but just the idea of algorithm. I need to learn them and know how they works.
Quote
o_lampe
Sounds like a lot of math
What is it you want to optimize?
Strength, speed, looks?
For strength, a 3D-honeycomb could be better than zig-zag? But it yet hasn´t been implemented AFAIK. ( only 2D honeycomb )
I tried to implement skin pattern printing once, but didn´t get far.
-Olaf
About this. I am trying to optimize the speed of printing. But of course, i will test those algorithm and by the way check some different things like strength, accuracy. Abut this honeycomb.Firstly, thanks for this idea. I heard about this and I didn't check it because i thought that it is not a path planning algorithms. I thought that it is just a shape which is applied to layer to print less material. And the paths, for extruder, are taken from some different algorithm. I don't know it for sure but i will check it.
When i finish those research i could give you my feedback abut them.
Sorry, only registered users may post in this forum.