Artificial Intelligence can be dated back to 1956, when four men, John McCarthy, Marvin Minsky, Allen Newell, and Herbert Simon met at Dartmouth College for a conference. At this conference, the field of Artificial Intelligence research was born. McCarthy, regarded as the founder of AI, defined AI as “the science and engineering of making intelligent machines.” More specifically, AI utilizes computer software and puts it into machines to do functions that humans can naturally do. These supercomputers using AI can simulate what humans can do; communicate, move, and think. In 1960, the Department of Defense started to heavily fund projects that dealt with AI. Laboratories started to become established around the world. However, in 1974, funding was cut in the U.S. when England ran into criticism for setting up laboratories and funded projects for AI.
By 1980, AI research was revived and began to get the funding it needed to get back off the ground. What also really sparked the comeback of AI was the success Japan was having when it created its Fifth Generation Computer. From there, the United States used its laboratories and great research to create supercomputers that could perform tasks of logistics, data mining, and medical diagnosis. America’s success can be attributed to several factors such as the computational power of computers, focusing on specific problems, linking problems and fields, and a new commitment from researchers. All of this led up to what we have today in Artificial Intelligence: The DARPA Grand Challenge, Deep Blue, and Watson.
International Business Machines, or IBM, is known for their ability to push the limits between technology and human intelligence. An example of this was Deep Blue, the chess playing supercomputer who beat the world’s best chess master. More recently, IBM has come up with a supercomputer that not only can beat humans in a game of Jeopardy!, but will also be able to benefit society in ways never done before. IBM created Watson, the supercomputer capable of playing the game show Jeopardy! amongst past champions. Watson was first started back in 2005 and is based off IBM’s DeepQA technology, which is a type of answering system. In order for Watson to have a chance of even competing with Ken Jennings and Brad Rutter, Watson needed to find the answers from the internet and then very quickly hit the button signifying he knew the answer. In order to make this happen, Watson used software such as Java C++, Apache Hadoop, Apache UIMA, and SUSE Linux Enterprise Server 11. What is fascinating about Watson is that during the game show, he was not connected to the internet, but rather the engineers at IBM loaded and saved internet sources such as encyclopedias and databases into Watson’s 16 Terabytes of RAM, consisting of 200 million pages of content or equivalent to 4 Terabytes. Watson’s mainframe was built from 90 commercially available IBM Power 750 servers and POWER7 processors. Because Watson cannot see or hear, the clues were sent to it as text files the same time Brad and Ken were hearing the clues from Alex Trebek.
During the three day competition on Jeopardy!, Watson seemed to be in control from the very start. He did make a couple noticeable mistakes such as answering “Toronto” to a question asking about “Cities in the United States,” but besides that Watson knew what he was doing. Watson managed to beat his competitors to the punch most of the time in order to answer the questions and he even wagered when he selected the Daily Double. Watson went on to win the contest quite easily with a combined three day score of $77,147 while Jennings and Rutter had $24,000 and $21,600 respectively.
Watson is destined for bigger and better things than just appearing on Jeopardy!. Watson has already begun to work with Columbia University in helping doctors diagnose patients and what treatments are needed for those patients. Also at Columbia University, Watson will address critical issues in medicine. After his stay at Columbia, Watson will then be studied at the University of Maryland to discover how supercomputers such as Watson can be best utilized in the medical field. With the success of Watson, IBM has really created something much more than just a successful game show playing supercomputer. Supercomputers can now start to be built to help society in ways never thought before, and everyone will definitely benefit from such a success.
IBM’s Watson is an impressive piece of machinery, but it had some prevalent predecessors that were equally as historic. As one may very well be aware, IBM was in the business of building computers long before Watson, and the company met with a great deal of success. Among the aforementioned supercomputers, the supercomputer known as “Deep Blue” is one of the most notable examples. Deep Blue was developed by Feng-hsiung Hsu, who was later joined by C.J. Tan, Murray, Campbell, and several other talented scientists. The supercomputer was originally designed with the purpose of playing chess at a grandmaster level, a feat which Deep Blue performed spectacularly. Deep Blue wasn’t the first chess playing computer out there, but it certainly proved itself as the best. Deep Blue could calculate two hundred million moves per second. (That’s fifty billion possible moves in the time period a chess player is allowed to make his moves.) After a few tournaments, trials, and various tests, Deep Blue established itself as the world’s greatest chess playing supercomputer. After earning such a stellar reputation, IBM sought to make history. In 1996, and again in 1997, Deep Blue made history by playing World Chess Champion Garry Kasparov in two six-game series of chess. Despite being bested overall in the first series by Kasparov in 1996, Deep Blue became the first computer ever to beat a World Chess Champion in a game of chess. In the 1996 series, Deep Blue won one game, lost three games, and tied two games. And even though Deep Blue lost, it was a major victory in the world of technology. After working on the supercomputer some more, IBM presented an upgraded and modified Deep Blue a year later for a rematch. In the rematch series in 1997, Deep Blue was able to beat Kasparov by winning two games, losing one game, and tying three games. During the historic series, IBM began to provide live updates online. Many people tuned in to witness history and learn about this amazing technological feat. In fact, so many people kept checking the website for the latest news on the historic event that the website eventually crashed. Perhaps even more interesting is the fact that IBM actually used Deep Blue’s server technology to fix the problem with the website, and soon everyone who cared about the matches was able to get the updates they wanted without any problem or interruption. After the series had ended with Deep Blue’s victory, Kasparov accused IBM and Deep Blue of cheating and demanded a rematch. However, no rematch has been scheduled to this day. When Deep Blue was created, its long term goal was to help the world. Since Deep Blue could simulate so many possible scenarios in such a small period of time, the technology could be applied to real world situations to find results for proposed solutions to advanced problems. Today, Deep Blue’s RS/6000 SP2 technology is used to help with many important tasks, such as cleaning up waste, inventing new car designs, making new medicine to help people, and many other tasks.
The Darpa Grand Challenge is a very significant event because of the advanced technologies that go into the automobiles. It is a prize competition between driverless or automated vehicles that happens every year. It was funded by the Defense Advanced Research Projects Agency, which is a research organization of the United State’s Department of Congress. These technologies that are built into the cars can help the United States as well as other countries in fighting wars. The goal of the Challenge is to create the first fully functional ground vehicles capable of riding off-road courses for a limited amount of time. This is important because it means that less people will be injured or killed on the battlefield.
The contest began in 2004 and countries such as Japan, Germany, and Italy have been involved with it. The cars that are built must complete a 130-mile track that is filled with different terrains and obstacles. The cars must complete this without receiving signals to the robot once the race is started. This means the cars must make “intelligent” decisions in real time based on the decisions of other vehicles. The makers of these cars face difficult challenges because they need to use very advanced and special technologies to finish the racetrack.
First, the team must decide which vehicle they want to use to modify before they do anything else. The car has to have good off-road capabilities, but also, the ability to be upgraded. For example, teams must build a variety of different software and hardware combinations for interpreting sensor data, planning, and execution. Various cars that have competed in the challenge wrote code into the car using C++. This C++ code adds greater typing strength, scoping and other tools useful in object-oriented programming and permits generic programmingvia templates. The language is sometimes criticized because it is complicated and thus difficult to fully master, however, it runs very well with C# which is also used in many of the vehicles.
In 2005, “Stanley,” was the first vehicle to complete the Darpa Grand Challenge. It used the C++ and C# coding, along with 17 dual core servers to make the car functional. Stanley used five roof mounted Sick AG Lidar units to navigate the course. Other vehicles that have competed have used up to 40 dual core servers making the car very expensive and sophisticated. In the 2007 Challenge, the Tartan Racing Team won the race. It used a new and advanced control system and it layered mission planning, motion planning, behavior generation, perception, world modeling, and mechatronics. This was a good advancement because it allowed the vehicles to have a better GPS system.
The Darpa Grand Challenge is a very noteworthy event because it has led to new and advanced technologies being built. The Department of Congress is using this to help with the military because it reduces the chance of death if there aren’t any drivers in a vehicle.
Here is the WordPress Page to Team 12′s DSS Project “The Evolution of Watson.”
1991., & Project, t. i. (n.d.). History of artificial intelligence – Wikipedia, the free encyclopedia. Wikipedia, the free encyclopedia. Retrieved April 19, 2011, from http://en.wikipedia.org/wiki/History_of_artificial_intelligen
Artificial intelligence – Wikipedia, the free encyclopedia. (n.d.). Wikipedia, the free encyclopedia. Retrieved April 19, 2011, from http://en.wikipedia.org/wiki/Artificial_intelligence
C++ – Wikipedia, the free encyclopedia. (n.d.). Wikipedia, the free encyclopedia. Retrieved April 19, 2011, from http://en.wikipedia.org/wiki/C%2B%2B
DARPA Grand Challenge – Wikipedia, the free encyclopedia. (n.d.). Wikipedia, the free encyclopedia. Retrieved April 19, 2011, from http://en.wikipedia.org/wiki/DARPA_Grand_Challenge
Deep Blue (chess computer) – Wikipedia, the free encyclopedia. (n.d.). Wikipedia, the free encyclopedia. Retrieved April 19, 2011, from http://en.wikipedia.org/wiki/Deep_Blue_%28chess_com
IBM – Watson. (n.d.). IBM – United States. Retrieved April 19, 2011, from http://www-03.ibm.com/innovation/us/watson/index.html
IBM Archives: Deep Blue. (n.d.). IBM – United States. Retrieved April 19, 2011, from http://www-03.ibm.com/ibm/history/exhibits/vintage/vintage_4506VV1001.html
IBM Research | Deep Blue | Overview. (n.d.). IBM Research. Retrieved April 19, 2011, from http://www.research.ibm.com/deepblue/
Watson (computer). (2011, April 17). Wikipedia, the free encyclopedia. Retrieved April 19, 2011, from en.wikipedia.org/wiki/Watson_%28computer%29