Is Fuzzy logic applicable for robotics research

Question

I found some papers that use fuzzy logic to solve robotics algorithmic problems. But I have recently been told by a few roboticists that fuzzy logic should not be used in robotics because it has recently been proven by some theoreticians to be a mathematically dubious field. And so any results that you generate using fuzzy logic in robotics will not get published. Is that true?

I find it hard to believe. Is fuzzy logic dubious with respect to applications on real-world problems, say robotics? I have talked to mathematicians and they surely believe that fuzzy logic is a robust and clever field. But then mathematicians do a lot of abstract stuff that is completely irrelevant to the real world and must not be touched. Is fuzzy logic one of those things?

Answer 1

I would like to mention that Fuzzy logic is still an active control system used in many industry applications.

In garbage fired power plants, concrete aggregate firing, hydraulics, and the control of flow of powdered 'fluids' in foundries to name a few. However, I will admit, I've only seen them used in 'one off' difficult to model projects, such as power plants and the such. However I have read papers like you that use it in robotics....

I would say it most certainly has it's place in research, and it is certainly used in many control areas that are near impossible to control in more traditional methods. However much like you'll see Neural Networks being sticker'd and used on/in everything under the sun.

Fuzzy logic suffered the same fate, but fell out of interest because of it...It was thought that you could throw away all the old tools and 'save the world'. But it became a marketing failure and other methods became (stayed?) the norm.

I don't think Fuzzy logic is 'dubious' in any way. I think it goes against the methods of many designers to be exact and well...not so fuzzy, and has a poor history of being used in junk machines, and lots of burnt memories.

It's an interesting field, one you should learn about, maybe even play around with in matlab or your flavour and learn about....if not for anything else than to say you like something else more.

Some reading materials:

NOE TS Fuzzy Modelling of Nonlinear Dynamic Systems with Uncertainties using Symbolic Interval-valued data

Advances in Industrial Control ^this is more about PIDS and MPCs

Fuzzy Logic, Identification and Predictive Control

and just for sources sake, a list (in german anyways) of successful applications of Fuzzy logic (maybe just scroll through the pictures to get an idea if you can't read german). As I said, it has it's place....it's just not very big.

Answer 2

Fuzzy logic is definitely used in many of the control systems including but not limited to robotics. See this paper for an example:

https://pdfs.semanticscholar.org/b9a7/332b03d46b3ee08b9d113e64714e6b668601.pdf

and this:

https://ieeexplore.ieee.org/document/1678143

If we consider fuzzy logic as dubious then we should do the same to probabilities. Both Fuzzy logic and Probabilities range between 0 and 1; fuzzy logic uses degrees of truth as a mathematical model of vagueness, while Probability uses mathematical model of ignorance.

Answer 3

Short answer: Fuzzy logic (FL) isn't applicable for robotics research, The long answer is, that in the 1980s as part of the fifth computer generation fuzzy logic was researched in Japan with the attempt to build intelligent advanced parallel computers, but the Japanese researchers have failed. Fuzzy logic isn't a technical idea but a philosophical description of the world and a marketing term used for selling rice cookers and washing machines.

After a short time, the consumers have noticed, that Fuzzy stabilized camcorders doesn't have any improvements over a normal camcorder and multi-value logic was repressed into obscurity. If a robot contains of a so called fuzzy controller, it's a clear sign, that's a non-academic project which can't be proven if it's working or not.[1] The term fuzzy has it's origin in a clumsy cowboy character in western film produced in black/white recording decades ago. It's a synonym for a failed project which doesn't fulfill the minimum standards in academia.

[1] Elkan, Charles, et al. "The paradoxical success of fuzzy logic." IEEE expert 9.4 (1994): 3-49.

Answer 4

I have not seen any industry-grade application of fuzzy logic in space, flight, automotive control systems. Fuzzy logic came during mid-60s and it gradually faded away due to several reasons:

1. It did not solve any control problem that cannot already be solved by the existing methods at that time. Bad news, no major advantage in terms of extending the application domain!

2. It lacked theorems for proving stability, performance of closed-loop system. Bad news, not being so rigorous!

The second item is of utmost importance for safety-critical systems such as space, aviation, self-driving cars, surgical robotics, etc. You are not allowed to deploy an aircraft flying over cities before certifying its flight control system. When you put a fuzzy logic controller in the loop, how can you prove that the closed-loop system is stable? I mean, in the rigorous sense. FAA won't accept the ad-hoc hand-tuning, working experiment results as a certificate. You have to provide the operating conditions that fuzzy logic controller will not work.

Back to your question, nowadays, there are so many publishers in academia. Even if you do fuzzy logic research in robotics, of course, you can find a journal that will eventually accept your work. The real question is if someone will read your paper or adopt your methods for their own application.

A friendly advice, as a researcher, I would not invest my time to do research in fuzzy logic area. Control theory is already a mature field; and fuzzy logic is arguably NOT one of the fruitful sub-areas. I spent one year working on fuzzy logic during my undergraduate years. Then, I came to USA for my PhD and realized that most of control researchers in USA were not working on fuzzy logic anymore. Nonetheless, never say never! Maybe, you can contribute to the field and make it more widely-used in the future.