The artificial intelligence robot is an amazing creature. Created by Man, however, it is endowed with an intelligence that sometimes seems to be close to that of the human being. To better understand this incredible technology, however, it is necessary to understand how artificial intelligence robots work, from their physical envelope to the computer that serves as their brain.

 Human beings consist of 5 basic components: a body structure, a muscular system to animate this structure, a sensory system to receive information about this body and its environment, a source of energy to activate muscles and muscles. meaning, and a brain system that processes sensory information and tells the muscles what to do. We also have intangible attributes, such as intelligence and morality, but the attributes listed above are the main physical components of the human being.

 A robot consists of the same components. A typical robot is an assembly of a physical structure, an engine, a sensory system, a power source and a computer acting as a brain to control these different elements. Concretely, robots are versions created by the human of an animal. They are machines that reproduce the behavior of a human or an animal.

 One of the pioneers of industrial robotics, Joseph Engelberger, has noted that robots can not be defined, but it is possible to know when one sees them. If we review all the different machines that men consider as robots, it is indeed impossible to propose a concrete definition.

 Famous robots rooted in popular culture include Star Wars R2D2, Sony's AIBO robot dog, Honda's ASIMO humanoid robot, industrial robots on factory assembly lines, Star's robot Data Trek, Comedy Central BattleBots, army bomber robots, NASA rovers, Space Odyssey HAL, Robomower standalone mowers, LEGO MindStorms robotics kit.

 All these machines and artificial intelligences are considered robots and yet they are all different. For most roboticists, however, the definition is more precise: a robot has a reprogrammable brain that can move a body.

 If one opts for this definition, robots are distinguished from other mobile machines like cars by the presence of a computer. Many recent cars ship a computer, but this computer only serves to make minor adjustments. Robots are also distinguished from ordinary computers by their physical nature. Robot and artificial intelligence are intimately linked.

 Most robots have several features in common. First, all robots have a removable body. Some are content with motorized wheels, others have dozens of removable segments. Like the bones of a human body, these individual segments are connected by joints. To rotate their wheels and rotate their segments, the robots are equipped with an actuator. This actuator can be an electric motor or a solenoid, or a hydraulic or pneumatic system.

 To power these actuators, a robot needs a source of energy. Most robots are equipped with a battery, or a cable to connect to a wall outlet. Hydraulic robots also need a pump to press the hydraulic fluid, while pneumatic robots require an air compressor or a compressed air tank.

 The actuators are connected to an electrical circuit. This circuit powers the electric motors and the solenoids. It also activates the hydraulic system by manipulating the electric valves. The valves determine the route of the pressurized fluid within the machine. For example, to move a hydraulic leg, the robot controller must open the valve connecting the fluid pump to the piston cylinder attached to the leg. The pressurized fluid expands the piston and moves the leg forward. Typically, to move their segments in two directions, robots use pistons that can grow in two ways.

 The robot's computer controls everything attached to the circuit. To move the robot, the computer activates all necessary motors and valves. Most robots are reprogrammable, which means that you just have to write a new computer program to change its behavior.

 All robots have sensory systems, and some have the ability to see, hear, smell or taste. The most common sense among robots is that of movement, ie the robot's ability to monitor its own movements. As a rule, wheels are attached to the robot seals. An LED on one side of the wheel emits a beam of light through the slots to a light sensor on the other side of the wheel. When a robot moves a specific seal, the wheel turns. The slots interrupt the light beam as the wheel rotates, and the brightness sensors read patterns of flashing light and transmit the data to the computer. The computer can thus accurately determine the extent of joint displacements based on this pattern. This is the same system used by computer mice. These are the basics of robotic operation. Roboticians can combine these elements in an unlimited way to create more complex robots.

 The robotic arm

 The term robot comes from the Czech word robota, usually translated as "forced labor". Indeed, this term sums up the fate of a majority of robots. Most machines in this world are designed to perform repetitive, painful or dangerous tasks for humans.

 The most common manufacturing robot is the robotic arm. A typical robotic arm is made up of seven metal segments connected by six joints. The computer precisely controls the robot by rotating the motors connected to each joint, and the arm repeats the same movements over and over again. Motion sensors allow it to make sure it does not move too much.

 An industrial robot with six seams looks a lot like a human arm. It has the equivalent of a shoulder, an elbow and a wrist. Typically, the arm is connected to a stationary base structure rather than to a removable body. This type of robot has six degrees of freedom, which means it can rotate in six different ways. In comparison, a human arm has seven degrees of freedom.

 The role of the arm is to move the hand from one place to another. Similarly, the role of the robotic arm is to move an effector from one place to another. It is possible to equip a robotic arm with all kinds of effectors, designed for a specific use. A commonly used effector is similar to a simplified version of the human hand, capable of grasping and carrying different objects. Robotic hands often have pressure sensors that indicate to the computer the force exerted by the robot. Other effectors include paint sprays, torches, or drills.

 Industrial robots are designed to repeat the same task over and over again in a controlled environment. To teach a robot how to do its job, the programmer guides the arm using a controller. The robot memorizes the exact sequence of movement movements and repeats it infinitely.

 Most industrial robots work on assembly lines, and assemble cars. Robots can do this job more efficiently than humans because of their accuracy. They can systematically drill in the same place, and always maintain the same strength, no matter how many hours they work, unlike the human beings who get tired. Manufacturing robots are also very important in the history of computing, because it takes extreme precision to assemble microprocessors.

 Mobile robots

 Robotic arms are relatively easy to build and program because they operate only within a small perimeter. A robot with mobility is much more complex. We must first develop a system of locomotion. If the robot has to move on flat ground, the wheels or tires are usually the best option. However, many roboticists opt for legs that are more adaptable and closer to the human locomotion system.

 Typically, hydraulic or pneumatic pistons move the robot legs forward and backward. The pistons are attached to the different segments of the legs as the muscles are attached to the different bones of the human body. Just as a baby's brain must understand which combination of muscles allows it to walk without falling, the robotician must determine the combination of pistons movements that allow walking, and program it into the robot's computer. Many mobile robots have an integrated balance system, telling the computer when it is necessary to correct its movements.

 Biped locomotion is inherently unstable, which makes it very difficult to implement to robots. To create more stable walking robots, designers are generally inspired by the animal world, and more specifically by insects. Six-legged insects are well balanced and adapt well to a wide variety of terrain.

 Some mobile robots are controlled using a remote control. A human tells them what to do and when and when to do it. The remote control can communicate with the robot through a wire, or using radio or infrared signals. Remote-controlled robots are very useful for exploring dangerous or inaccessible environments such as underwater depths or inside a volcano.

 Autonomous robots

 Autonomous robots can act independently. They are programmed to respond in a certain way to an external stimulus. One can take the example of a robot equipped with an obstacle sensor. Once activated, this robot follows a straight line. When it hits an obstacle, the impact activates its sensor. His program tells him to turn back, turn right and continue moving forward again. Thus, the robot changes direction each time it encounters an obstacle.

 Advanced robots rely on more elaborate versions of this idea. Roboticians are creating new programs and new sensor systems that are smarter and more sophisticated. Today, robots can effectively navigate in a wide variety of environments.

 The simplest mobile robots use infrared sensors and ultrasonic sensors to detect obstacles. These sensors work in the same way as the echolocation of animals. The robot sends a sound signal or infrared light beam and detects the signal reflection. It measures the distance of the obstacles based on the time taken by the signal to return.

 The most advanced robots use stereo vision to visualize the world around them. Two cameras give these robots depth perception, and image recognition software gives them the ability to locate and classify various objects. Robots can also use microphones and odor sensors to analyze the world around them.

 Some autonomous robots can only work in a familiar environment. Clipper robots, for example, depend on well-defined borders that define the boundaries of the garden. The most advanced robots, however, can analyze unknown environments and adapt to them, even in the most sinuous areas.

 These robots can associate certain terrain patterns with certain actions. For example, a rover can develop a map of the terrain that faces him based on his visual sensors. If the map shows hilly patterns, the robot knows how to move in a certain way. This type of system is very useful for exploratory robots responsible for discovering other planets. An alternative design is to rely on randomness. When a robot gets stuck, it moves its body in any way possible until one of them works.

 Artificial intelligence robot

 Artificial intelligence is probably the most exciting field of robotics, but also the most controversial. The debate is raging to determine whether robots must be intelligent. Just like the concept of robot, intelligence is difficult to define. The ultimate artificial intelligence would be the recreation of the process of human thought. A machine created by humans, with equal or superior intellectual abilities.

 Such intelligence would include the ability to learn, reason, use language and formulate original ideas. Roboticians are still far from reaching such a level of intelligence, but have made many advances in this area. Today, an artificial intelligence robot can replicate specific elements of our intellectual faculties.

 Computers can solve problems in some areas. The basic idea of ​​problem solving by an artificial intelligence robot is very simple, but its execution is complicated. First, the artificial intelligence robot or the computer gathers facts about a situation through sensors or human inputs. The computer compares this information with stored data and determines what that information means. They then launch a variety of possible actions and predict which actions will solve the problem. Of course, the computer can only solve the problems that it is programmed to solve. He does not have generalized analytical abilities.

 Some robots, however, have the ability to learn, to a certain extent. These robots recognize if a certain action achieves a desired result. The artificial intelligence robot then stores this information and tries this action to success the next time it encounters the same situation. Again, modern computers can only perform these tasks in limited situations. They can not absorb information as a human. Some robots can learn by imitating human actions. For example, in Japan, roboticists taught a robot artificial intelligence how to dance by demonstrating movements.

 Some robots can interact socially. For example, Kismet, an artificial intelligence robot from MIT, recognizes human body language and vocal intonation and responds accordingly. The creators of Kismet are interested in how humans and babies interact, based solely on the tone of the voice and the expression of the gaze. These interactions could be the basis of a learning system similar to that of humans.

 Kismet and the other humanoid robots of the MIT AI Lab operate using an unconventional control structure. Rather than performing each action using a central computer, robots control lower-level actions using lower-level computers. Program Director Rodney Brooks believes this is a model more like human intelligence. We perform most tasks automatically, without deciding to do them consciously.

 The real challenge of artificial intelligence is to understand how natural intelligence works. Developing an artificial intelligence robot is not the same thing as making an artificial heart. Scientists do not have a simple and concrete model to emulate. The brain contains billions of neurons, and we think by establishing electrical connections between these different neurons. However, we do not know exactly how these connections lead to higher reasoning or operations of a lower level. This system does not seem understandable.

 For this reason, research in artificial intelligence is mainly theoretical. Scientists make assumptions about how we learn and think and why we do it. They then experiment with these ideas using the artificial intelligence robot. Brooks and his team focus on humanoid robots when they feel that the ability to experience the world as a human is essential to developing human intelligence. These robots also allow easier interactions and thus make learning easier for the artificial intelligence robot.

 Just as physical robotics is a useful tool for understanding human and animal anatomy, artificial intelligence research is useful for understanding how natural intelligence works. For some roboticists, this is the ultimate goal of robot design. Others visualize a world in which we interact with the artificial intelligence robot and use a variety of inferior robots for manual work, health care and communication. A large number of experts predict that the evolution of robotics will turn us into cyborgs, namely humans combined with machines.

 In any case, the artificial intelligence robot will occupy an important place in our daily lives in the future. In the decades to come, robots will gradually move from the industrial and scientific world to everyday life, in the same way that computers took over homes in the 1980s.