Category: 

What is Adaptive Control?

Article Details
  • Written By: Mary McMahon
  • Edited By: Kristen Osborne
  • Last Modified Date: 10 November 2016
  • Copyright Protected:
    2003-2016
    Conjecture Corporation
  • Print this Article
Free Widgets for your Site/Blog
President Richard Nixon had an entire speech prepared in case the Apollo 11 astronauts became stranded on the Moon.  more...

December 8 ,  1965 :  Pope Paul VI promulgated Vatican II into ecumenical law.  more...

Adaptive control is an approach to controlling systems that adjusts over time in response to changing conditions and knowledge acquired by the controller. Rather than remaining static or attempting to cope with minor deviations, adaptive control actively responds to changes in the system to improve control. This approach is needed with dynamic systems in unstable environments, ranging from aircraft to medical robots.

With adaptive control, the controller collects data about the environment the system is operating in and uses this information to make adjustments to how the system is controlled. Examples of adaptive control can be seen in some vehicles with the ability to adjust automatic braking systems for wet and icy conditions. In these cases, the system responds to the conditions to improve accuracy, effectiveness, and efficiency to make driving safer and easier in a wide variety of settings.

Control systems are often required to make decisions and adaptive control facilitates smart decision-making based on the conditions at a given time, rather than an established set of parameters developed by the person who built the system. In dynamic environments, this can be critical. Adaptive control can be used to control networks and other computer systems with changing user loads to help the systems respond to changes before problems develop by doing things like shifting users to different users, throttling down traffic to prevent system crashes, and so forth.

Ad

Designing adaptive control requires a number of skills. In addition to developing a control system, the developer also has to integrate some level of artificial intelligence along with data analysis abilities, allowing the system to gather data and interpret it in meaningful ways. Since changes can happen very rapidly, high processing speed is also required. Systems need to be able to respond to changes in fractions of a second, and ideally to take a proactive rather than retroactive approach to managing changing conditions.

People interested in this field can be found at colleges and universities, as well as in labs at technology companies. Many people have advanced degrees in fields like engineering, electronics, computer science, and robotics, as a high level of skill and education is needed to design functional adaptive control systems. Rates of pay vary, depending on the working environment; people in the private sector tend to make more, but people working for government agencies and in university settings can access the latest research, development, and technology, including classified material developed for military use.

Ad

You might also Like

Recommended

Discuss this Article

nony
Post 4

@MrMoody - They already use adaptive control systems in things like cars too. Everyone has heard of cruise control – just “set it and forget it,” right?

Well now they have this thing called adaptive cruise control. What that does is it checks for a safe driving distance between you and the car in front of you. I think it uses sensors to check the distance while you’re driving.

Think about how many accidents this could prevent and how many lives this could save, especially if you fall asleep at the wheel. I don’t know if this technology is only installed in higher end automobiles or if it’s an option on all cars, but I believe that pretty soon it will become standard on all models.

MrMoody
Post 3

@Charred - The technology has been around a long time. Think about what we mean when we say that the “computer lands the airplane.” What are we saying? A computer is taking the controls and doing the stuff that the pilot normally does.

Certainly by surrendering control to the computer the pilots are taking a risk (they can always regain control if they need it). However, the computer can do more than go through the routine checklist of what you must do to land the plane.

It can respond to things like turbulence and other unexpected changes through a robust control mechanism. In my opinion it can do this faster and more accurately than a pilot can, and with split second timing.

Charred
Post 2

@SkyWhisperer - I agree. A robust adaptive control system would have to employ something like that. Also, it would need the ability to recognize patterns. Take medical robots for example. They are supposed to look at organs and determine what they’re looking at, and more importantly, what goes where for performing surgical operations.

They aren’t dealing with neat snapshots of information. They would have to use pattern recognition and image analysis, I would guess, in order to know what they’re looking at. Artificial intelligence plays a big role here. I wish I understood the subject more frankly.

SkyWhisperer
Post 1

I love the integration of the adaptive controller with artificial intelligence. This is what makes it possible to develop truly “smart” systems in the field of electronics in my opinion.

A good adaptive controller would have to be able to work with imprecise measurements. For example, my digital camcorder has image stabilization technology. When you shake the camera, the circuit adjusts for that camera shake to try to make the frame of image steady.

I don’t know exactly how it does it but I’ve heard that it uses something called “fuzzy logic,” where you’re dealing with approximations and not precise measurements. I’ll bet that adaptive control systems use fuzzy logic too, at least as part of their artificial intelligence algorithms.

Post your comments

Post Anonymously

Login

username
password
forgot password?

Register

username
password
confirm
email