I'm trying to implement a fuzzy logic membership function in C for a hobby robotics project but I'm not quite sure how to start.
I have inputs about objects near a point, such as distance or which directions are clear/obstructed, and I want to map how strongly these inputs belong to sets like very near, near, far, very far. Does anyone have a tip on how to start? Thanks.
Disclaimer: I've never implemented a fuzzy controller (I've only ever used PI or PID in real-life) and control class was 10 years ago.
Here's an presentation demonstrating moving towards a target using distance and angle for inputs and power as the output. FuzzyTech's Example positioning a crane
This just presents the topic and theory i.e. no code.
Best source is probably one of the robotics groups
e.g Seattle Robotic Society fuzzy logic tutorial it is technical ... and long.
if you can access technical journals then search Google scholar for "fuzzy logic" "path planning" robotics
if you're looking for some ideas on how to implement fuzzy logic then perhaps a Application Note from one of the microchip manufactures will get you started e.g Microchip's paper on Airflow control or servo control. I know it's not Arduino but Microchips papers are usually very clearly presented.
And finally an example in c++ its probably more complex than you're looking for. Free fuzzy logic library
Good luck.
I'm not expert with fuzzy logic, but according to my basic understanding, you could start by deciding what distances would constitute near (say 10 cm) far (say 1m), then you use probabilities to fill in the range in between (so 55cm might be 50% near, 50% far). Then you do something similar for your other properties, and combine the probabilities associated with each property with more probabilities.
Do you have a good reference for designing fuzzy controls?
I suppose you could start here. I think they at least describe simple fuzzification and defuzzification routines.
The guys at MakeProto have created an automatic code generator for Fuzzy Systems that outputs C code from Matlab fuzzy systems, or by a hand-defined fuzzy system.
Might be worth taking a look at.
http://makeproto.com/blog/?p=35
Fuzzy inference system can be implemented in both C and C++. Learn How to frame fuzzy logic in c
Related
I'm implementing an accelerometer-based pedometer, and I was wondering if there was any known algorithms to handle that.
You have probably found this:
Enhancing the Performance of Pedometers Using a Single Accelerometer
Anyhow, I am also interested in finding a good algorithm, I am curios what other answers you will get. :)
There is an app called Sensor data that you can uses to gather experimental data so you can then analyze it and try to find an algorithm.
Its going to be quite tricky to find a very good algorithm especially for the iPhone since its accelerometer is quite noisy
There's an interesting paper (with source code) here that may be of help: http://www.analog.com/static/imported-files/application_notes/47076299220991AN_900.pdf.
The charts are interesting. If I were to do this myself I would probably sample the data at a fairly high frequency, convert to frequency domain with a FFT, apply a digital band-pass filter to cut off all frequencies outside the expected minimum/maximum walking speeds (including any DC offset), do a reverse-FFT to reconstruct the now-filtered signal and then run the resulting data through an edge-detector with a Hysteresis function. This is all pure speculation of course but looking at those charts I think it would work, it would be relatively fast to code up and well within the processing power of a mobile phone.
I've been studying hierachial reinforcement learning problems, and while a lot of papers propose interesting ways for learning a policy, they all seem to assume they know in advance a graph structure describing the actions in the domain. For example, The MAXQ Method for Hierarchial Reinforcement Learning by Dietterich describes a complex graph of actions and sub-tasks for a simple Taxi domain, but not how this graph was discovered. How would you learn the hierarchy of this graph, and not just the policy?
In Dietterich's MAXQ, the graph is constructed manually. It's considered to be a task for the system designer, in the same way that coming up with a representation space and reward functions are.
Depending on what you're trying to achieve, you might want to automatically decompose the state space, learn relevant features, or transfer experience from simple tasks to more complex ones.
I'd suggest you just start reading papers that refer to the MAXQ one you linked to. Without knowing what exactly what you want to achieve, I can't be very prescriptive (and I'm not really on top of all the current RL research), but you might find relevant ideas in the work of Luo, Bell & McCollum or the papers by Madden & Howley.
This paper describes one approach that is a good starting point:
N. Mehta, S. Ray, P. Tadepalli, and T. Dietterich. Automatic Discovery and Transfer of MAXQ Hierarchies. In International Conference on Machine Learning, 2008.
http://web.engr.oregonstate.edu/~mehtane/papers/hi-mat.pdf
Say there is this agent out there moving about doing things. You don't know its internal goals (task graph). How do you infer its goals?
In way way, this is impossible. Just as it is impossible for me to know what goal you had mind when you put that box down: maybe you were tired, maybe you saw a killer bee, maybe you had to pee....
You are trying to model an agent's internal goal structure. In order to do that you need some sort of guidance as to what are the set of possible goals and how these are represented by actions. In the research literature this problem has been studied under the terms "plan recognition" and also with the use of POMDP (partially observable markov decision process), but both of these techniques assume you do know something about the other agent's goals.
If you don't know anything about its goals, all you can do is either infer one of the above models (This is what we humans do. I assume others have the same goals I do. I never think, "Oh, he dropped his laptop, he must be ready to lay an egg" cse, he's a human.) or model it as a black box: a simple state-to-actions function then add internal states as needed (hmmmm, someone must have written a paper on this, but I don't know who).
I have a course in my current semester in which I'm required to do a project on application of AI. I have decided to do this on game AI. I have 2 basic ideas: implementing an FPS bot(s) or implementing soccer AI.
I'm quiet a noob at AI right now, I've implemented basic pathfinding algos (A*, etc), and have studied about Finite state machines, some First Order logic, basic Neural Network stuff(Backpropagation ALgo), and am currently doing a course on Genetic Algorithms.
Our main focus is on the bot right now. Our plans include:
Each 'bot' would be implemented using a Finite State Machine (FSM), which would contain the possible states the bot could have; & the rules for the action/state changes that are going to take place when it receives an input.
In bot group movement, each bot would decide whether to strike, ways to strike; based on range, number of bots, existing fights using Neural Networks.
By using genetic algorithms the opponents next move could be anticipated based on repetitive moves.
Although I've programmed a few 2d games till now in my free time (like pacman, tetris, etc), I've never really gone into the 3d area. We will most probably be using a 3d engine.
We want to concentrate most of our energy on the AI part. We would like not to be bothered with unnecessary details about the animation/3d models, etc. For example, if we could find a framework which has functions like Moveright() which just moves the bot to the right, it would be really awesome.
My basic question is : is it too ambitious to go about it in the way we have planned, considering the duration of the project is abour 3 months? Should we go 3d and use a 3d game engine? is it easy to use such engines, if you have no experience with them before? If yes, what kind of engine would be suitable to our project?
I came accross another idea, given in the book AI Game programming by example, where the player would have a top down view of the bots. Would that way be more appropriate?
Thanks .. sorry about the length of the question .. it's just that my problem is a bit too specific.
My basic question is : is it too
ambitious to go about it in the way we
have planned, considering the duration
of the project is abour 3 months?
Yes -- but that's not necessarily a bad thing :)
Should we go 3d and use a 3d game
engine?
No. Mainly because you said:
We want to concentrate most of our
energy on the AI part.
Here's what I'd do, based on my experience (and knowing that, as a student, I often bit off way more than I could chew, too):
Make your simulation function irrespective of a graphical component. Have it publish "updates" to another layer, that consist of player and ball vectors. By doing so you'll be keeping your AI tasks separate from everything else, which means you have fewer bugs to worry about, and you can also unit test your underlying simulation much easier.
Take those "updates" and create your first "visualization" layer -- make it the simplest 2D representation possible. It could just be a stream of text lines: "Player 1 has the ball / Player 1 kicked ball at (30,40) with speed 20kph". That will be hard enough for your first pass since you'll be figuring out how to take data published by the simulation and doing something with it.
Your next visualization might add a 2D grid of ANSI graphics (think rogue-like) to actually show players and the ball moving. Your next one after that might be sprites. And so on. Note how you incrementally increase the complexity of your visualization... don't make your first step go to using a technology (3d graphics engine) you've never used before. (You'll never finish your project in that case.)
As for your questions about which route to take -- FSMs, NNs, GAs, top-down design -- you should rank your interest in them from most to least (along with the rest of your group) and then tackle them, in that order. You might consider doing one style for one team and a different design for the other team. You might want to make your FSM team play against a FSM team that's had an additional tweak done to it, in order to compare and contrast if you think your changes are actually being beneficial (you might be surprised and find out they make the team worse). Actually, that's where unit testing and splitting the simulation from the visualization come in very, very handy -- you should be able to "sim" as many games as you need to to get experimental results without worrying about graphics. You might even do it in batches overnight with scripts.
In general, my advice to you is this: break down your project into the tiniest pieces you can, and tackle them one at a time, so no matter where you're at when time runs out, you'll have something interesting to show off.
You could have a look at guntactyx, that's what I had to use when I did my AI unit at uni.
It takes care of all the display, physics, sound etc... for you, all you have to do is program your team of bots.
The API includes functions to make the bot move left or right, shoot, hear sounds (like gun shots) etc... and it comes with a few sample bots so you don't start from scratch.
Also, it's quite fun to watch your bots battling your friends' bots :)
Anyone have good book / article recommendation for procedural generation of background music? (No vocals, just instruments).
I'm not interested in:
How do I generate the sound of a particular note on a particular instrument
I'm interested in:
How do I generate the melody / score for the music.
Thanks!
EDIT:
Thanks for the reference to Brian Eno. I'm definitely looking into the ambient/user can ignore type of music. I.e. think the background music of a game. It's there to provide some basic mood, but the focus is the game.
Sometime ago I ran into ChucK, which is a programming-language to generate music/sound/audio:
ChucK presents a new time-based, concurrent programming model that's highly precise and expressive (we call this strongly-timed), as well as dynamic control rates, and the ability to add and modify code on-the-fly. In addition, ChucK supports MIDI, OSC, HID device, and multi-channel audio. It's fun and easy to learn, and offers composers, researchers, and performers a powerful programming tool for building and experimenting with complex audio synthesis/analysis programs, and real-time interactive control.
I believe the end result can be converted into MIDI, which can then be converted into a score or sheet notation.
I don't know if this is what you're looking for. Hope this helps!
EDIT
After thinking about this a little longer, I think what you can possibly do (and this sounds a bit crazy) is write code that generates ChucK code. So define a set of rules for your music/score generation and then use that to create valid ChucK code. After you run the ChucK code, you can get a MIDI file which you can then convert into score/sheet-music.
The book "Computer Models of Musical Creativity" by David Cope should help you along with the theoretical side of computer-assisted composition, though you might want some music theory under your belt before you dive in.
If you are interested in procedural music check out the Condition30 site -- condition30.com
This music is all procedural.
If you're interested in an implementation of procedural music based on cellular automata in C#, you could grab the source code from http://proceduralmidi.codeplex.com/. A binary is also available.
For a linguistics course we implemented Part of Speech (POS) tagging using a hidden markov model, where the hidden variables were the parts of speech. We trained the system on some tagged data, and then tested it and compared our results with the gold data.
Would it have been possible to train the HMM without the tagged training set?
In theory you can do that. In that case you would use the Baum-Welch-Algorithm. It is described very well in Rabiner's HMM Tutorial.
However, having applied HMMs to part of speech, the error you get with the standard form will not be so satisfying. It is a form of expectation maximization which only converges to local maxima. Rule based approaches beat HMMs hands down, iirc.
I believe the natural language toolkit NLTK for python has an HMM implementation for that exact purpose.
NLP was a couple years ago, but I believe without tagging the HMM could help determine the symbol emission/state transition probabilities of n-grams (i.e. what are the odds of "world" occurring after "hello"), but not parts-of-speech. It needs the tagged corpus to learn how the POS interrelate.
If I'm way off on this let me know in the comments!