More than 3000 engineers find our updates useful. You can get them at your mail box!
  
Search your paper presentation and project titles:

Department/Area of interest: ( To list the projects / paper presentations)

Mechanical               Scada technology              Communication             Computer science           Alternative energy
Electrical                  Robotics                        Biometrics                     Artificial intelligence             Electronics

Artificial intelligence (AI)- space!

Artificial intelligence (AI)

The term artificial intelligence (AI) is commonly taken to mean the study of thinking and perceiving as general information-processing functions— or the science of machine intelligence (MI).
In the past few decades, computer systems have been programmed to diagnose diseases; prove theorems; analyze electronic circuits; play complex games such as chess, poker and backgammon;
solve differential equations; assemble mechanical equipment using robotic manipulator arms and end effectors (the “hands” at the end of the manipulator arms); pilot uncrewed vehicles across
complex terrestrial terrain, as well as through the vast reaches of interplanetary space; analyze the structure of complex organic molecules; understand human speech patterns; and even write
other computer programs.

All of these computer-accomplished functions require a degree of intelligence similar to mental activities performed by the human brain. Someday, a general theory of intelligence may
emerge from the current efforts of scientists and engineers who are now engaged in the field of artificial intelligence. This general theory would help guide the design and development of even
smarter robot spacecraft and exploratory probes. Artificial intelligence generally includes a number of elements or subdisciplines. Some of the more significant of these elements or subdisciplines
are: planning and problem solving, perception, natural language, expert systems, automation, teleoperation and robotics, distributed data management, and cognition and learning.
All artificial intelligence involves elements of planning and problem solving. The problemsolving function implies a wide range of tasks, including decision making, optimization, dynamicresource
allocation, and many other calculations or logical operations. Perception is the process of obtaining data from one or more sensors and processing or analyzing
these data to assist in making some subsequent decision or taking some subsequent action. The basic problem in perception is to extract froma large amount of (remotely) sensed data some
feature or characteristic that then permits object

identification.
One of the most challenging problems in the evolution of the digital computer has been the communications that must occur between the human operator and the machine. The human
operator would like to use an everyday, or natural, language to gain access to the computer system. The process of communication between machines and people is very complex and frequently requires
sophisticated computer hardware and software. An expert system permits the scientific or technical expertise of a particular human being to be stored in a computer for subsequent use
by other human beings who have not had the equivalent professional or technical experience. Expert systems have been developed for use in such diverse fields as medical diagnosis, mineral
exploration, and mathematical problem solving. To create such an expert system, a team of software specialists will collaborate with a scientific expert to construct a computer-based interactive
dialogue system that is capable, at least to some extent, of making the expert’s professional knowledge and experience available to other individuals. In this case, the computer, or thinking machine,
not only stores the scientific (or professional) expertise of one human being, but also uses its artificial intelligence to permit ready access to this valuable knowledge base by other human users.
Automatic devices are those that operate without direct human control. NASA has used many such automated smart machines to explore alien worlds. For example, the Viking 1 and 2
lander spacecraft placed on the Martian surface in 1976 represent one of the great early triumphs of robotic space exploration. After separation from the Viking orbiter spacecraft, the lander (protected
by an aeroshell) descended into the thin Martian atmosphere at speeds of approximately 9,940 miles per hour (16,000 km per hour). The descending lander was slowed down by aerodynamic
drag until its aeroshell was discarded. Each robot lander spacecraft slowed down further by releasing a parachute and then achieved a gentle landing by automatically firing retrorockets.
Both Viking landers successfully accomplished the entire soft landing sequence automatically, without any direct human intervention or guidance. Teleoperation implies that a human operator
is in remote control of a mechanical system. Control signals can be sent by means of hardwire (if the device under control is nearby) or in a wireless mode via transmitted electromagnetic signals—for
example, laser or radio frequency—(if the robot system is some distance away and operates within ine-of-sight of the transmitter). NASA’s Pathfinder mission to the surface of Mars in 1997 successfully
demonstrated teleoperation of a mini-robot rover at interplanetary distances. The highly successful Mars Pathfinder mission consisted of astationary lander spacecraft and a small surface
rover. NASA named the lander spacecraft the Carl Sagan Memorial Station in honor of the American astronomer Carl Sagan (1934–96), who popularized astronomy and the search for extraterrestrial
life. The mini-rover was called Sojourner, after the American civil rights crusader Sojourner Truth. The six-wheeled mini-robot rover vehicle was actually controlled (or teleoperated) by the Earthbased
flight team at the Jet Propulsion Laboratory

(JPL) in Pasadena, California. The human operators used images of the Martian surface obtained by both the rover and the lander systems. These interplanetary teleoperations required that the
rover be capable of some semi-autonomous operation, since there was a time delay of signals that averaged between 10 and 15 minutes in duration depending upon the relative positions of Earth
and Mars over the course of the mission. This rover had a hazard avoidance system and surface movement was performed very slowly. Starting in 2004, NASA’s Mars Exploration
Rovers, Spirit and Opportunity, provided even more sophisticated and rewarding teleoperation experiences at interplanetary distances, as they rolled across different portions of the Red
Planet.


Of course, in dealing with the great distances in interplanetary exploration, a situation eventually arises in which electromagnetic wave transmission cannot accommodate any
type of effective “real-time control.” When the device to be controlled on an alien world is many light-minutes or even light-hours away, and when actions or discoveries require splitsecond
decisions, teleoperation must yield to increasing levels of autonomous, machineintelligence- dependent robotic operation. Robot devices are computer-controlled
mechanical systems that are capable of manipulating or controlling other machine devices, such as end effectors. Robots may be mobile or fixed in place and either fully automatic or
teleoperated. The more AI a robot has, the less dependent it is upon human supervision. Large quantities of data are frequently involved in the operation of automatic robotic
devices. The field of distributed data management is concerned with ways of organizing cooperation among independent, but mutually interacting, databases. Instead of transmitting
enormous quantities of data back to Earth, an advanced robot explorer will use AI to selectively sort and send only the most interesting data.
In AI, the concept of cognition and learning refers to the development of a level of machine intelligence that can deal with new facts, unexpected events, and even contradictory
information. Today’s smart machines handle new data by means of preprogrammed methods or logical steps. Tomorrow’s smarter machines will need the ability to learn, possibly
even to understand, as they encounter new situations and are forced to change their mode of operation.


Perhaps late in this century, as the field of artificial intelligence sufficiently matures, scientists can send fully automatic robot probes on interstellar voyages. Each interstellar probe
must be capable of independently searching a candidate star system for suitable extrasolar planets that might support extraterrestrial life.

Intense Debate Comments