<DIV><BR>Subject: People Are Robots, Too. Almost<BR><BR><BR>Spotlight Feature <BR> October 28, 2003<BR><BR>People Are Robots, Too. Almost<BR><BR>Popular culture has long pondered the question, "If it looks like a<BR>human, walks like a human and talks like a human, is it human?" So far<BR>the answer has been no. Robots can't cry, bleed or feel like humans,<BR>and that's part of what makes them different.But what if they could<BR>think like humans?<BR><BR>Biologically inspired robots aren't just an ongoing fascination in<BR>movies and comic books; they are being realized by engineers and<BR>scientists all over the world.!
While much emphasis is placed on<BR>developing physical characteristics for robots, like functioning<BR>human-like faces or artificial muscles, engineers in the Telerobotics<BR>Research and Applications Group at NASA's Jet Propulsion Laboratory,<BR>Pasadena, Calif., are among those working to program robots with forms<BR>of artificial intelligence similar to human thinking processes.<BR><BR>Why Would They Want to Do That?<BR><BR>"The way robots function now, if something goes wrong, humans modify<BR>their programming code and reload everything, then hope it eventually<BR>works," said JPL robotics engineer Barry Werger. "What we hope to do<BR>eventually is get robots to be more independent and learn to adjust<BR>their own programming."<BR><BR>Scientists and engineers take several approaches to control robots.<BR>The two extreme ends of the spectrum are called "deliberative control"<BR>and "reactive control." The former is the traditional, dominant way in<BR>which robots function, !
by painstakingly constructing maps and other<BR>types of models that they use to plan sequences of action with<BR>mathematical precision. The robot performs these sequences like a<BR>blindfolded pirate looking for buried treasure; from point A, move 36<BR>paces north, then 12 paces east, then 4 paces northeast to point X;<BR>thar be the gold. <BR><BR>The downside to this is that if anything interrupts the robot's<BR>progress (for example, if the map is wrong or lacks detail), the robot<BR>must stop, make a new map and a new plan of actions. This re-planning<BR>process can become costly if repeated over time. Also, to ensure the<BR>robot's safety, back-up programs must be in place to abort the plan if<BR>the robot encounters an unforeseen rock or hole that may hinder its<BR>journey.<BR><BR>"Reactive" approaches, on the other hand, get rid of maps and planning<BR>altogether and focus on live observation of the environment. Slow down<BR>if there's a rock ahead. Dig if you see a b! ig X
on the ground.<BR><BR>The JPL Telerobotics Research and Applications Group, led by technical<BR>group supervisor Dr. Homayoun Seraji, focuses on "behavior-based<BR>control," which lies toward the "reactive" end of the spectrum.<BR>Behavior-based control allows robots to follow a plan while staying<BR>aware of the unexpected, changing features of their environment. Turn<BR>right when you see a red rock, go all the way down the hill and dig<BR>right next to the palm tree; thar be the gold. <BR><BR>Behavior-based control allows the robot a great deal of flexibility to<BR>adapt the plan to its environment as it goes, much as a human does.<BR>This presents a number of advantages in space exploration, including<BR>alleviating the communication delay that results from operating<BR>distant rovers from Earth.<BR><BR>How Do They Do It?<BR><BR>Seraji's group at JPL focuses on two of the many approaches to<BR>implementing behavior-based control: fuzzy logic and neural networks.<BR>The main
difference between the two systems is that robots using fuzzy<BR>logic perform with a set knowledge that doesn't improve; whereas,<BR>robots with neural networks start out with no knowledge and learn over<BR>time.<BR><BR>Fuzzy Logic<BR><BR>"Fuzzy logic rules are a way of expressing actions as a human would,<BR>with linguistic instead of mathematical commands; for example, when<BR>one person says to another person, 'It's hot in here,' the other<BR>person knows to either open the window or turn up the air<BR>conditioning. That person wasn't told to open the window, but he or<BR>she knew a rule such as 'when it is hot, do something to stay cool,'"<BR>said Seraji, a leading expert in robotic control systems who was<BR>recently recognized as the most published author in the Journal of<BR>Robotic Systems' 20-year history.<BR><BR>By incorporating fuzzy logic into their engineering technology, robots<BR>can function in a humanistic way and respond to visual or audible<BR>signals, o! r in
the case of the above example, turn on the air<BR>conditioning when it thinks the room is hot.<BR><BR>Neural Networks<BR><BR>Neural networks are tools that allow robots to learn from their<BR>experiences, associate perceptions with actions and adapt to<BR>unforeseen situations or environments.<BR><BR>"The concepts of 'interesting' and 'rocky' are ambiguous in nature,<BR>but can be learned using neural networks," said JPL robotics research<BR>engineer Dr. Ayanna Howard, who specializes in artificial intelligence<BR>and creates intelligent technology for space applications. "We can<BR>train a robot to know that if it encounters rocky surfaces, then the<BR>terrain is hazardous. Or if the rocky surface has interesting<BR>features, then it may have great scientific value."<BR><BR>Neural networks mimic the human brain in that they simulate a large<BR>network of simple elements, similar to brain cells, that learn through<BR>being presented with examples. A robot functioning with such a
system<BR>learns somewhat like a baby or a child does, only at a slower rate.<BR><BR>"We can easily tell a robot that a square is an equilateral object<BR>with four sides, but how do we describe a cat?" Werger said. "With<BR>neural networks, we can show the robot many examples of cats, and it<BR>will later be able to recognize cats in general."<BR><BR>Similarly, a neural network can 'learn' to classify terrain if a<BR>geologist shows it images of many types of terrain and associates a<BR>label with each one. When the network later sees an image of a terrain<BR>it hasn't seen before, it can determine whether the terrain is<BR>hazardous or safe based on its lessons.<BR><BR>Robotics for Today and Tomorrow<BR><BR>With continuous advances in robotic methods like behavior-based<BR>control, future space missions might be able to function without<BR>relying heavily on human commands. On the home front, similar<BR>technology is already used in many practical applications such as<BR>! digital
cameras, computer programs, dishwashers, washing machines and<BR>some car engines. The post office even uses neural networks to read<BR>handwriting and sort mail.<BR><BR>"Does this mean robots in the near future will think like humans? No,"<BR>Werger said. "But by mimicking human techniques, they could become<BR>easier to communicate with, more independent, and ultimately more<BR>efficient."<BR><BR>JPL is a division of the California Institute of Technology in<BR>Pasadena, Calif.<BR><BR>Media Contact: Charli Schuler (818) 393-5467</DIV><BR><BR><P>La vie est belle!<BR><BR>Yosé (<A href="http://www.cordeiro.org">www.cordeiro.org</A>) <IMG src="http://us.i1.yimg.com/us.yimg.com/i/mesg/tsmileys/1.gif"></P>
<P>Caracas, Venezuela, Americas, TerraNostra</P><p><hr SIZE=1>
Do you Yahoo!?<br>
<a href="http://antispam.yahoo.com/whatsnewfree">Protect your identity with Yahoo! Mail AddressGuard</a>