Intelligent Autonomous Robots are the new generation

225362-286769At least six fields of research today advanced robotics structure: one that relates the robot with its environment, the behavioral, cognitive, or developmental epigenetics, the evolutionary and biorrobótica. It’s a big field of interdisciplinary study that relies on the mechanical, electrical, electronics and informatics, as well as physical science, anatomy, psychology, biology, zoology and ethology, among others. The basis of this research is embodied Cognitive Science and the New AI. Its purpose: lighting intelligent and autonomous robots that reason, behave, evolve and act like people. By Sergio Moriello.multidisciplinary study, which relies largely on the engineering (mechanical, electrical, electronics and computers) and science (physics, anatomy, psychology, biology, zoology, ethology, etc.).. Refers to highly complex automated systems that have an articulated mechanical structure, governed by an electronic control system, and characteristics of autonomy, reliability, versatility and mobility.

In essence, the “autonomous intelligent robots” are dynamic systems consisting of an electronic controller coupled to a mechanical body. Thus, these machines require adequate sensory systems (to perceive the environment in which they operate), a precise mechanical structure adaptable (to have a certain physical skills of locomotion and manipulation) of complex effector systems (for running the assignments) and sophisticated control systems (to carry out corrective actions when necessary) [Moriello, 2005, p. 172].

Situated Robotics (Situated Robotics)
This approach deals with robots that are embedded in complex and often dynamically changing [Mataric, 2002]. It is based on two central ideas [Florian, 2003] [Muñoz Moreno, 2000] [Innocenti Badano, 2000]: robots) “are embodied” (embodiment), ie, have a suitable physical body to experience its environment so direct where their actions have immediate feedback on their own perceptions, and b) are situated “(situatedness), ie, they are embedded within an environment, interact with the world, which directly influences-its-on behavior.

Obviously, the complexity of the environment has a close relationship with the complexity of the control system. Indeed, if the robot has to react quickly and intelligently in a dynamic and challenging environment, the problem of control becomes very difficult. If the robot, however, need not answer quickly, reducing the complexity required to develop control.
Within this paradigm, there are several subparadigmas: the “Behavior-based robotics,” the “cognitive robotics”, the “epigenetic robotics”, the “evolutionary robotics” and “biomimetic robotics.

Behavior-Based Robotics and Behavior (Behavior-Base Robotics)
This approach uses behavioral principles: robots generate a behavior only when stimulated, ie respond to changes in their local environment (as when someone accidentally touches a hot object). Here, the designer divides tasks into many different basic behaviors, each of which runs on a separate layer of the control system of the robot.

Typically, these modules (behaviors) may be to avoid obstacles, walking, lifting, etc.. The intelligent features of the system, such as perception, planning, modeling, learning, etc.. emerge from interaction between the various modules and the physical environment where the robot is immersed. The system-control-Fully distributed incrementally builds, layer by layer, through a process of trial and error, and each layer is only responsible for basic behavior [Moriello, 2005, p. 177 / 8].

The behavior-based systems are capable of reacting in real time, as calculated directly from the actions of perceptions (through a set of correspondence rules “situation-action). It is important to note that the number of layers increases the complexity of the problem. Thus, a very complex task may be beyond the ability of the designer (it was hard to define all the layers, their interrelationships and dependencies) [Pratiharas, 2003].

Another drawback is that due to the presence of several individual behavior and dynamics of interaction with the world, it is often difficult to say that a series of actions in particular has been the product of a particular behavior. Sometimes several behaviors simultaneously working or are exchanging rapidly.

Although intelligence may reach the insect, probably built systems from this approach have limited skills, as they have internal representations [Dawson, 2002]. Indeed, this type of robots present a great difficulty to execute complex tasks and in the simplest, no guarantee the best solution as optimal.

Cognitive Robotics (Cognitive Robotics)

This approach uses techniques from the field of Cognitive Science. It deals with deploying robots that perceive, reason and act in dynamic environments, unknown and unpredictable. Such robots must have cognitive functions that involve high-level reasoning, for example, about goals, actions, time, cognitive states of other robots, when and what to perceive, learn from experience, and so on.

For that, they must possess an internal symbolic model and their local environment, and sufficient capacity for logical reasoning to make decisions and to perform the tasks necessary to achieve its objectives. In short, this line of work is responsible for implementing cognitive characteristics in robots, such as perception, concept formation, attention, learning, memory, short and long term, etc.. [Bogner, Maletic, Franklin, 2000].

If we achieve that the robots themselves develop their cognitive abilities, is avoid the “hand” for every conceivable contingency task or [Kovacs, 2004]. Also, if the robots is achieved using representations and reasoning mechanisms similar to that of humans, could improve human-computer interaction and collaborative work. However, it needs a high processing power (especially if the robot has many sensors and actuators) and lots of memory (to represent the state space).

Epigenetic Robotics and Development
This approach is characterized in that tries to implement control systems of general purpose through a long process of development or self-autonomous organization. As a result of interaction with their environment, the robot is able to develop different-and increasingly complex-perceptual skills, cognitive and behavioral.

This is a research area that integrates developmental neuroscience, developmental psychology and robotics located. Initially the system can be equipped with a small set of behaviors or innate knowledge, but, thanks to the experience-is able to create more complex representations and actions. In short, this is the machine to independently develop the skills appropriate for a given particular environment transiting through the different stages of their “autonomous mental development.

The difference between robotics and robotics development epigenetic-sometimes grouped under the term “ontogenetic robotics (ontogenetic robotics) – is a subtle thing, as regards the type of environment. Indeed, while the former refers only to the physical environment, the second takes into account the social environment.

The term epigenetic (beyond the genetic) was introduced in psychology, “by Swiss psychologist Jean Piaget to describe his new field of study that emphasizes the individual sensorimotor interaction with the physical environment, rather than take into account only to genes. Moreover, the Russian psychologist Lev Vygotsky supplemented this idea with the importance of social interaction.

The robots create a record that they will learn to follow

Robot2Researchers in robotics from the University of California at Davis have developed a control system which allows the robots to collect evidence suggesting that their leader is about to turn, predict where it will go and then follow him. “This is a fundamental problem in robotics,” says Sanjay Joshi, associate professor mechanical and aeronautical engineering at UC davis. Indeed, whether walking down the street, while driving on the highway or in many other situations, the man often collected deliberate signals and unconscious clues in order to predict what the others and act accordingly. The robots, however, more difficult to coordinate so, for example when the leader of a group turns a corner and disappears from the field of vision of its congeners. Studies in behavioral psychology have shown that a person is about to turn unconsciously a brief nod in the direction it is preparing to borrow so that others can follow.

A system inspired by studies of behavioral psychology
Humans use different signals and to build a predictive model of where their peers are going. It is on this model that Sanjay Joshi and his team have developed an integrated control system for robots to take into account such behavioral cues in their decision making. The research team tested the system on a small robot called marketed Evolution Robotics Scorpion. The camera of the robot was able to identify his sidekick who was ahead. Its computer system could then combine this information with behavioral indices. Rather than having programmed the robot leader to send signals directly saying we should follow, the research team sent “behavioral cues” to the second robot via a wireless device.

Teaching robots to follow men
The index said that the leader was running but did not specify in what direction. To decide the direction it should take, the robot has developed its own prediction by combining the index with other parameters such as speed and direction of the leader. The researchers concluded that robots capable of incorporating behavioral information into their decision parameters acted more efficiently by following the leader that others. “Robots increasingly capable of integrating behavioral and follow them would be easier to integrate a human work. In a hospital robot could follow such a doctor in the hallways.” A paper describing this work is published in the August 2008 issue of IEEE Transactions on Industrial Electronics.

Robo Nexus 2004: spam is now but where’s Aibo?

robosapienThe robot exhibition held from October 21 to 23, 2004 in Santa Clara, the heart of Silicon Valley is always a good opportunity to take stock of a rapidly evolving sector. This exhibition, the largest of its kind, brings together the Convention Center of town all that is best in robots. Thousands of professionals and enthusiasts jostle for three days on the stands and conferences to identify trends and see the main news. Continue reading

The snake robot slips through the barriers

serpent-robot-imageThe realization of robots inspired snake occupies a prominent place in robotics over the past decade. But it has hitherto been difficult to accurately reproduce the movements of the reptile. A skill that researchers SINTEF have managed to emulate a system composed of Aiko robot and a virtual double of the snake which allows experiments on computer. Unlike most of its predecessors, this snake does not need wheels to be able to move with ease. For if the recent addition facilitate the removal, by converting the twisting motion in a continuous slip, he moves better on smooth surfaces. But the interest is to use this type of machine in the affected areas. “In a collapsed building where there are lots of rubble, for example after an earthquake, a wheeled snake would probably be stuck,” says Aksel Transeth researcher at SINTEF.

A virtual double
A robot more versatile, more closely reproducing the twisting of the snake could he, pushing the obstacles he encounters, like stones. This new model is called Aiko, measuring 1.5 meters and is composed of segments of PVC tubes with motors connecting joints. It is capable of pushing objects that barrier in his path at 15 cm per second. Moreover, a virtual double of Aiko, which accurately reflects the movements and reactions of a snake in real life, will be used to guide system development. Thus, if this is not the first attempt to move obstacles to snake robot Aiko is the first to have a virtual double that simulates previously movement. However the research team qualifies: “It is easier to simulate on a computer than to build a robot and to experience in reality.”

Promising advances
It therefore remains difficult to construct a model of robot that accurately reproduced the movements of real reptile. It is still too complex to master perfectly the reflexes of robot and some segments of Aiko can interact with the ground or obstacles he encounters in different ways in different situations. For example, a segment can slide over an obstacle while another will try to climb over. To help achieve these movements, researchers have launched a program transcript of gestures algorithms to mathematically translate their observations. They also explore new ways to improve those movements. One of them is the integration of sensors.

Nanobots gifted sense of touch

nanorobotRobots at the nanoscale can already do some basic manipulation. But it remained difficult to control to enable them to perform more complex actions such as handling of nanoelectronic components or cells. To overcome this problem, a team of University of Toronto (Canada) announced having developed a pair of robotic grippers can move independently in the middle of a microscopic environment, without damaging the surrounding components. The principle is simple: these micro robots are endowed with the sense of touch. “The robots are equipped with load cells that allow them to perceive their environment through touch,” said L’Atelier Philippe Bidaud, director of the Institute for Intelligent Systems and Robotics (ISIR). “And therefore include data such as weight of an object, the resistance of a membrane,” says he.

Reuse of robotic processes common
According to Sun Yu, project manager, they would be the first to really feel the pressure with which they are grabbing an object and can be taken into account in their operations. “We apply the concepts already used in traditional robotic, but at the nanoscale, announced the researcher. “The experiments above do not elicit such a return. The pliers broke things they took, or break themselves,” says he. Also new: they can feel the proximity of an object and either take it or avoid it to prevent any damage. So many functions that, if the clamps are connected to a computer program containing operations to follow, let them act without human intervention. In tests on animal cells, these would nanobras robotic manipulation and made a damage rate of only 15%. “Giving these nanobots sense of touch makes it possible to achieve the nano-scale manipulation of space we do can do today to the human scale, “adds Philippe Bidaud.

Build sensors premium
And numerous applications, including industrial level. Such systems would in fact assemble microelectronic devices and sensors designed to integrate health appliance or high-tech (PC, mobile phone). A process that can be achieved with conventional manufacturing techniques. Especially it should not be expensive: the clips are made from conventional silicon wafers. If they were mass produced, they may well be offered at 10 dollars fifty pairs. Another use, not least, they have the opportunity to help rebuild body tissues. Finally, recalls Philippe Bidaud, they could play a significant role in therapy. “These systems will know certain properties of cells that we are currently inaccessible,” he concludes. From a technical perspective, each “arm” is about three millimeters long. They can catch cells and components of only ten micrometers wide. This, in less than a second.

Robopolis “spam utilities have been big hits”

robotJerome Damelincourt leader Store Robopolis, has agreed to give us his vision of robotics, he meets a world forever. At a time when Asimo serving coffee and talk to the caller, one wonders in what direction the robot moves. The Workshop – Hello Jerome Damelincourt. People come from afar to come and visit Robopolis. How did you get the idea to open a shop dedicated to robots? Jerome Damelincourt – In 2000, I decided to create the website vieartificielle.com. Before his success and public enthusiasm, I opened Robopolis in 2003. To mark the opening of the shop, I also created a website dedicated robopolis.com. It was a dream that I never let go.
What kind of audience greet you at Robopolis?

There are two types of targets: the active and passionate fans liabilities. The first category includes those who want to assemble their own robots and see them run while the second brings together children and fans of science fiction. Now that the books and films of science fiction describe the technique can actually create it. Basically, some things are no longer science fiction but reality.
You sell all kinds of robots: figurines, utilitarian robots, gadgets, etc.. What products work best?

Robots have been big commercial successes. The Roomba vacuum cleaner is autonomous (see our list). The RC 3000 lava soil alone and silently. We also have a robot mower etc.. Today, these products are not much more expensive than their conventional counterparts. The Roomba is the price of a good vacuum cleaner and these robots are technically very good.I think both will grow together. People will continue to buy robots to perform functions specific to the house such as cleaning the floor, sending messages or control of television. At a horizon of five years, although I imagine that robots are hired by large enterprises for example by distributing catalogs in salons. The rental and sale will be used with different objectives: we will not rent Asimo home for the price it costs.
If Sony, with its very large capacity, had to stop production of its Aibo, it does show there that the robots are too expensive to develop and they are not profitable enough? Is there a risk that companies are abandoning research on robotics?

Aibo was a technological showcase for Sony. The Japanese had to create everything Aibo, since the technology that allows a robot to move forward on all fours. Aibo was to continuous innovation. Expenditures made by Sony in research are well established. Companies can now sell robots like Aibo much cheaper. It’s like computers, initially the production of computers was very expensive and the products were not very profitable, but the combined effect of learning and economies of scale allowed the computing of ‘reach the point where it is today. Within fifteen years, the robotics market has exceeded that of the computer.
I went to check on the site of a large appliance store, they do not sell Roomba. I was very surprised. Do not you find that surprising? When the consumer robots they integrate supermarkets?
No, I do not find that surprising. Major distributors include new standards only when they are sure to sell a certain number. Today, the Roomba is still the preserve of specialist shops whose role is precisely to take risks. However, I think the robots like Roomba will be sold in general stores within a year. In the U.S., they already are.
Do you think that robotics will grow to more B2B or B2C?

I do not think that one of them will still dominate there. B2B is already highly developed medicine, security, military … But B2C is also a research topic for many businesses.
Do you think the robot will he ever be the equal of man?
Do not get carried away. The robot’s capabilities are far from achieving those rights. Today, we can not create a robot that knows how to clear a table. We do not know how to explain to a machine what a glass because it can take many forms and be made of several components. Creators of robot soccer players have launched a challenge: in 2050 (not before), a robot soccer team will beat the world champion team football.
The public will he not reluctant to move with humanoid robots?
People love the humanoid robots. Plus it looks like a man, more like it. Until the robot becomes too human. Surveys have shown that nobody wants a robot to hair or skin. But Asimo, for cons, everyone would have at home!

The robots develop a sense of direction

TartaloThe Flame robot was already able to move through a process similar to that of a human being. A team of University of the Basque country now wants to give the droids ability to move independently and adapt to their environment. His robot Tartalo, has a navigation system allowing it to move freely in confined spaces such as apartments, even if it has not been scheduled for housing in particular. This is indeed able to adapt to changes in space. A camera positioned at eye level allows him to perceive its environment. The computer is equipped has been programmed to recognize four different areas: bedroom, corridor, lobby and no door. When placed in a new environment, he made several Featured order to identify and memorize the location of each piece.

Identify and control his environment
This, by creating a topological map and transmit it to its owner – for display and voice recognition solution – appoint the pieces. When approaching a place, it is also able to calculate the width and length of it to identify it: a long narrow space is perceived as a corridor, another broader as the lounge. It is the same for the doors if it can not open one, it will take the initiative to knock several times against the panel to show its presence. Tartalo, which measures about 1.5 meters, is finally able to overcome the obstacles he encounters on his journey. Indeed, it is equipped with sensors that emit and detect ultrasounds, infrared lights and laser beams capable of estimating distance. His movements are still random: the robot wanders through the corridors of the university without any real purpose or destination.

Objectives of assistance to the person and for military
But what is interesting is not so much what he can do yet, but the possibility of carrying forward the tasks spontaneously and independently. Their goal is to render their droid can achieve by itself a number of applications becoming more important. And you can move on any surface in any location, from the moment he received the order. It could then be used in very different areas, assistance to seniors and disabled military applications. Several improvements, however, must still be made before any marketing or approximation of the robot with other projects. Indeed, if he knows differentiate a human from another inanimate, it is not yet able to identify faces, voices, or a particular object it should return to its owner.

Ryobot by Patrick Brennan

imagesRYOBOT is my Rug-Warrior-based robot. It stands about 8 inches high, and is about 6 inches in diameter. It has two-motor differential drive, a 360 degree bump skirt, and the full complement of sensors from the Rug Warrior design. The logic is powered by 4 AA cells and the drive is powered by six rechargeable 2-volt D cells in two batteries of three each.

The mechanical design and fabrication was largely dictated by the resources available to me at the time I began this project. The main structure is made of three flat, circular levels, supported by steel threaded rod beams. The levels are made out of sheet styrene, which was chosen because it is easy to cut and relatively strong.

The cylindrical bump skirt is constructed out of aluminum sheet stock, and is supported from the second level of RYOBOT by 18-gauge wire. When complete, RYOBOT’s extrernal appearance should be essentially that of a festive little trash can on wheels.

The wheels of RYOBOT are connected directly to the output shafts of my gearmotors, since I do not expect RYOBOT to carry much weight. The gearmotors are surplus which I got through Edmund Scientific. They are bolted to an aluminum bracket which is in turn bolted to the first level of the robot. The output shaft is connected via a shaft coupler (not shown in the drawings) to a MECCANO shaft which mates with the MECCANO wheel. The shaft couplers were far and away the most difficult part to source for this design, and finding them delayed the construction for a long time. The drive assembly on the first level is the bulk of RYOBOT’s weight.

The Rug Warrior board is the brain of RYOBOT. It is a commercially available circuit board, available from various vendors in various forms:

  • (a) as a bare PC board and parts list with assembly directions,
  • (b) in kit form with all parts, or
  • (c) fully assembled.

D-bot by Roger

imagesThis robot is built using a 6-wheel motorized toy called the BOSS (Battery Operated Spin System) which was available at toy stores a few years ago for about $120. The unit included wheels, gear motors, batteries and charger. The plastic stuff was tossed and a wooden base was attached and multiple decks made from sheets of aluminum. The power electronics including the batteries and motor drivers are on the bottom deck, the computer is on the second deck and the sensors are on the top deck. The computer is a standard PC/XT motherboard which controls the 2 DC motors using the parallel printer port. Software resides on a 3.5″ floppy and automatically boots the robot program which was written in Quick-Basic V4.5. The software has the ability to read a joystick and record motions then play back the motions. This teach-and-repeat technique works well for short robot competitions where the task is well defined. Overall cost was about $450. Email questions to roger@robotics.com

Robby by Oualid Burström

imagesRobot arm commanded by a microcomputer. 
The processor used is a onechip processor from Intel (80c196KB).
12 MHz, 32K RAM, 16K ROM.I use an IBM PC compatible computer to send command sequences via the serial port (LPT1) 
to the 80c196KB.The 80c196KB transformes the command sequences to pulses which make the robot to move.
The PC program is a simulator. You can make a simulation and see how the robot will move.
The program language I used to program the 80c196KB and the PC was C.
I used Playwood and aluminium to build the robot.
The robot is commanded by four R/C servos (HITEC HS-300). 

Specifications: 
- A robot-arm commanded by a microcomputer.
- The microcomputer: 16Kb ROM, 32Kb RAM, 12MHz.
- The microprocessor in the microcomputer is a onechip processor made by Intel, 80C196KB. 
- A program in a compatible IBM PC simulates the robot-arm's movement. 
- The command-sequences are then sended to the micorcomputer via COM port. 
- The microcomputer translates the commands and makes the robot to move. 
- The robot: Is a four axes robot and has four R/C servos (HITEC HS300). 
- One for the base, one for the choulder, one for elbow and 
  one for the gripper. 
- Materials used to build the robot are Playwood and Aluminium.
Hardware required for the PC's program :
- A PC based on Intel 80386 or highter.
- SVGA graphic for GUI.
- MS_DOS as OP.
- Communication port (COM1 or COM2) for sending/taking data 
  to/from the robot.