Technology: The naughty kid ruined my college entrance examination, but I turned over to nuclear fus

Although the Kalman filter algorithm has many advantages, this does not mean that this algorithm can really be far ahead of the support vector algorithm.

You want to really put this algorithm into play.

It is necessary to select the appropriate motion model and observation model, and determine the corresponding parameters and covariance matrix, in addition, it is necessary to select the appropriate initial state and initial error, and constantly update the state and error according to the observation data.

All in all, the requirements are not low.

But in the same way, the support vector machine algorithm also has its own advantages.

Lu Yu sighed and looked at his several quarreling subordinates in distress, not knowing what to say for a moment.

"You guys: Actually, that's right.

"However, the limitations are too great."

He bit his lip, wanting to make his point.

But in the end, he chose to remain silent.

In the microsatellite network system, the space used by the micromotor, that is, the vast and mysterious universe full of unknown dangers.

Support vector machine algorithms that can handle higher-dimensional data are obviously a better choice.

However, it's like a blind man touching an elephant.

If you can touch the tail, then you are doomed not to touch the trunk, and vice versa.

People who touch the nose of an elephant think that the elephant is a caterpillar. People who touch the elephant's legs think that the elephant is a pillar. From their point of view, there is no problem with speaking.

But... An elephant is an elephant!

As is the case now

, although both algorithms have their own advantages, they have many limitations between them.

This is also why Lu Yu did not directly express his opinion!

It's hard!

To support the Kalman filter algorithm, you have to find a suitable method to refute the support vector machine algorithm.

The reverse is also true.

Lu Yu's brain also ran at high speed and began to make choices.

"If you use a support vector machine, it requires a lot of training data and computing resources, and there are certain requirements for the quality and distribution of the data, if the training data is insufficient or unbalanced, the support vector machine may overfit or underfit. Appropriate kernel functions and penalty parameters, data normalization and normalization..."

At this thought, he spoke.

"Let's be quiet first."

After thinking for a moment, Lu Yu spoke.

"It's good that people are willing to discuss it."

"But I hope everyone can also realize that Kalman filters and support vector machines have their own advantages, disadvantages and applicable scenarios in micromotor system error compensation."

"In general, if the system is linear or nearly linear, has a clear physical model and parameters, and needs to be predicted and updated in real time, then the Kalman filter is a good choice."

"Conversely, if the system is nonlinear or complex, the physical model parameters are very complex and require high-precision classification or regression."

"Support vector machines are a better option."

"Of course, we can also combine these two algorithms, such as using the Kalman filter as the input or output of the support vector machine, or using the support vector machine as the model or parameter of the Kalman filter to achieve better results."

Lu Yu said all his thoughts.

But the researchers in the quarrel were not satisfied.

"Chief Engineer Lu, aren't you with thin mud?" One researcher complained, half-jokingly.

"No way, there is no high or low and good or bad right or wrong in this matter, what if I don't get along with the mud?"

"Do you just want to keep arguing?!"

Lu Yu spread his hands helplessly.

He was just about to say something, but an inexplicable light suddenly appeared in his eyes, as if he had inspiration!

"Wait, you all don't talk yet, all be quiet."

Lu Yu said directly.

The researchers, who were still arguing just now, suddenly fell into silence, although they had just quarreled fiercely, but they quarreled with each other. It is also to be able to make the research mechanism have a better future!

Now hearing that Lu Yu suddenly shut them all up, everyone can guess that Lu Yu must have some new idea or new idea!

Then at such a point in time, they spoke out to Lu Zonggong's thinking, which is obviously not a good thing for research!

So everyone obediently closed their mouths, and all turned their eyes to Lu Yu again, waiting for Lu Yu to give a new conclusion.

The experiment, which was originally noisy like a vegetable market, became so quiet that even the sound of a needle falling to the ground could be heard.

"It seems that Chief Engineer Lu has new inspiration."

"This kind of inspirational thing is something that we can't ask for, but for a genius like General Engineer Lu, it's simply ... Envy.

"It is estimated that the conclusion will be reached immediately."

Although the scholars shut up, they were all secretly thinking about something in their hearts.

They have experienced this situation more than once, and whenever the chief engineer falls into this state of inspiration, he often gives some refreshing speeches!

And his speech may directly make a huge change in the entire research plan.

For others, it is undoubtedly extremely difficult to make changes directly to the entire research mechanism, but for a genius with great knowledge and innovation like Lu Yu, this is definitely not too surprising!

What's more....

While everyone was thinking, Lu Yu suddenly spoke.

"I have bad news and good news."

"Which one would you like to hear more?"

A smile suddenly appeared on his face, and he looked at the two parties who were still arguing before.

"Good news and bad news?"

"What does that mean?"

When the quiet atmosphere of the institute was just broken, everyone couldn't help but feel a sense of nervousness in their hearts, and even the atmosphere did not dare to gasp easily, anxiously waiting for Lu Yu's speech, but when they really heard what Lu Yu said...

Doubts, confusion and puzzlement appeared on everyone's faces.

What is good news and bad news?

"Chief Engineer Lu, let me listen to the bad news first."

A researcher who had just supported the vector machine algorithm very much said with a wry smile: "I'm relatively timid, first listen to the bad news, and then listen to the good news to relax."

"Okay, then I'll tell you the bad news first."

"The bad news is."

"After thinking about it, I just found that the support vector machine algorithm is not suitable for the supply problem of our micro-satellite network micro-motor system, which is a compatibility obstacle to solve the problem of electromagnetic interference in microsatellite network... It is not a good option.

Lu Yu said with a smile.

"Ah..."

Hearing Lu Yu's answer, the researcher who had just spoken was suddenly a little deflated.

He had just argued with the female researcher for so long to prove that the support vector machine algorithm was a more suitable algorithm for micro-motor systems in microsatellite networks.

In fact, he himself is also very confident in his conclusion, after all, this is not a plan he came up with alone, but a unified conclusion after discussing with several other people for a long time.

And yet....

Lu Yu directly denied this conclusion.

Doesn't this prove that he has just argued with that woman for so long, and what he said is all wrong?

At this thought, the researcher's face also showed a little disappointment and unwillingness.

If others deny his idea like this, then he will definitely be very unwilling, and even blushing and arguing with that person.

But the person who spoke was Chief Engineer Lu.

Not to mention the terrible technical gold content that Chief Engineer Lu showed before, just looking at the fact that Chief Engineer Lu was still in the mud before, he suddenly changed his tone now, obviously thinking of something he didn't realize before.

Refutation... I'm afraid it really doesn't make sense!

"Alas."

"So we're really thinking wrong?"

"But it shouldn't be, I really think the support vector machine algorithm is definitely a good algorithm!" Why can't this be used?

"I can't figure it out, I can only say that we need to precipitate again."

The hearts of several other researchers were also a little depressed, especially the fiercer ones who had just quarreled, and they didn't even dare to look up at the female researcher, after all, once the reason is lost, people will be instigated.

Compared with the instigation of these people, the female researcher who had just quarreled red-faced her face suddenly became angry.

"Hmph!"

"Didn't I tell you a long time ago?"

"The Kalman filter algorithm is a better algorithm, you still don't believe what I said, alas, you should believe it when Chief Engineer Lu speaks!"

The smile on her face also became proud.

Although several other researchers were very unhappy when they saw her villain's victorious appearance, since the plan they proposed was rejected, it was not good to say anything.

Just now, the researcher suddenly thought of something, and there was a curious look in his eyes.

After half a ring, he once again turned his gaze to Lu Yu:

"Chief Engineer Lu, what you just mentioned is good news and bad news."

"Bad news, you just told us."

"What's

the good news?"

As soon as these words came out, several other researchers also reacted, and all focused their curious eyes on Lu Yu's body.

Yes, at the beginning, Lu Yu proposed good news and bad news, and now that the bad news has been said, what will be the remaining good news?

For a moment, everyone's hearts couldn't help but show expectations.

"Hmm... In fact, this so-called good news cannot be called good news in the real sense, but it is good news for you people.

Lu Yu said with a smile: "The bad news is that the support vector machine algorithm is not suitable for our project.

"And the good news... Yes.

"The Kalman filter algorithm is also not the most suitable algorithm for us."

"In other words."

"Both algorithms should actually be eliminated."

Yes?!

All the researchers in the lab were blindfolded at once!

Everyone looked at Lu Yu in disbelief, and even the female researcher who was radiant and had a very proud face just now showed an extremely surprised expression!

She almost didn't even hold the glass in her hand.

Just now, this female researcher was still feeling proud of her idea, how could it be that in the blink of an eye, the plan she proposed was also rejected?!

"No, Chief Engineer Lu."

"If neither algorithm is the optimal solution, what do you think is the optimal solution?"

A researcher finally couldn't help the doubts in his heart and poured it out!

In fact, he is indeed extremely puzzled, Kalman filter algorithm and support vector machine algorithm, are the better algorithms in their respective fields, although both have certain shortcomings, but for solving the current problem, it is undoubtedly two more suitable options!

The reason why they have just argued with each other over this question is mainly to choose which one is the most suitable between the two algorithms.

But who could have thought....

Lu Yu actually denied the idea of both sets of them at the same time!

They really can't think of what algorithm is more suitable for the compatibility problem of micromotor systems than these two algorithms.

The atmosphere is almost the same.

Lu Yu said with a smile: "Have you heard of genetic algorithms? "

Genetic algorithm?"

The researchers' faces all showed a look of inexplicable doubt, looking at each other, and no one knew how to speak.

"It's normal that you haven't heard of it."

"The main application area of this algorithm... Hmm..." Lu

rubbed his temples, thinking about how to explain it to the researchers.

The genetic algorithm is a kind of algorithm used more in his previous life, although the main application field is not in the aerospace field, but the aerospace field will be more or less involved.

And these people have not heard of this algorithm, which is obviously abnormal.

Then there is only one explanation that can be derived, that is, the current world, or the genetic algorithm in this blue star, has not really come out.

It may only exist in some front-line theoretical laboratories, or even in some front-line theoretical laboratories, and there is no sign of birth at all.

It's no wonder that none of these researchers know about genetic algorithms.

Therefore, Lu Yu simply did not explain to them which field this algorithm was used in and how it was researched, but directly told them what a genetic algorithm was.

"The genetic algorithm is a simplified version of the evolutionary implementation of Darwin, which is analogous to nature."

"By mimicking the process of natural selection and reproduction, genetic algorithms can provide high-quality solutions to a variety of problems involving search, optimization, and learning."

"At the same time, they resemble natural evolution and can therefore overcome some of the hurdles encountered by traditional search and optimization algorithms, especially for problems with a large number of parameters and complex mathematical representations."

"Even if you haven't used it, you should have heard of it, right?"