Autonomous robots may be characterized as intelligent machines capable of performing tasks in unstructured environments without explicit or continuous human control over their movements. Concepts range from small insect-like machines to highly sophisticated humanoid robots with social intelligence and awareness of their environment.
An autonomous robot can sense and gain information about its surroundings, work and move either part or all of itself for an extended period without human assistance, and avoid situations that are harmful to people, property, or itself. It may also learn, or gain new capabilities, like adapting to changing conditions or adjusting strategies for accomplishing its tasks.
New categories of autonomous and mobile robots have been developed that can significantly expand the applications of robotics.
Cognitive robots are endowed with artificial reasoning skills to achieve complex goals in complex environments. Cognitive robots can be used in manufacturing and as home helpers, caregivers, or emergency and rescue aids. They are also useful for space missions.
Moving On Their Own – A flying robot stars in a Microsoft video for young people interested in computer science
A number of research projects are focused on cognitive robotic systems, including the European Union’s project CoSy—Cognitive Systems for Cognitive Assistants—aimed at developing robots that are more aware of their environment and better able to interact with humans. Another is provided by the cognitive robot companion in the Cogniron Project of the French National Center for Scientific Research. The project aims at developing a robot that would serve humans in their daily lives. It would exhibit cognitive capabilities for adapting its behavior to changing situations and for various tasks.
Neurorobotics couples neuroscience with robotics. The overall goals of the activity are to develop high-performance, human-centered robotic systems to serve as physical platforms for validating biological models. Current activities are focused on developing robotic devices with control systems that mimic the nervous system, such as brain-inspired algorithms and models of biological neural networks.
The field of evolutionary robotics emerged from the idea of allowing robots to evolve. Although the field shares many of the insights of artificial life, which pioneered the use of genetic algorithms in the 1970s and 1980s, evolutionary robotics is distinguished by its insistence on making the leap from computer animations to physical machines. Evolutionary robotics aims at developing robots that acquire their own skills through close interaction with the environment. Evolutionary computational tools like neural networks, genetic algorithms, and fuzzy logic are used in developing intelligent autonomous controllers for robots.
Life-like robots are biologically inspired robots that resemble living systems and biological organisms, from insects to humans. Mobility mechanisms are incorporated into their design that mimic biological mobility systems, and the resulting robots are referred to as biomimetic robots. A recent life-like robot project is the BigDog built by Boston Dynamics Inc. with funding from DARPA—a quadruped robot that can walk, run, or climb on rough terrain (and other places where accessibility is difficult), and carry heavy loads up to 340 pounds. The iCat robot platform for human-robot interaction research, developed by Philips labs in the Netherlands, can generate different facial expressions and talks to its users. The Amphibian Snake robot ACM-R5, built by the Hirose Fukushima Robotics Lab in Japan, can slither and swim under water for 30 minutes, can navigate in very confined spaces, and can search for earthquake victims.
Several humanoid and anthropomorphic robots were created to imitate some of the physical and mental functions of humans. They wrestle, skate, or play soccer. Honda’s Asimo, originally developed in 2000, has more recently been equipped with software and an array of eight microphones, which enable it to understand three humans shouting at once. Sony has a dancing robot, Sugoi. A home helper robot, HRP-2 or Promet, from Kawada Industries Inc., understands voice commands. HRP-3 from the same company can work in hazardous environments and carry out disaster relief. The Kansei robot, developed by Japan’s Robot and Science Institute, can make up 36 different facial expressions in response to words associated with different emotions.
Cyborg robots, in the form of hybrid biological/artificial assistive limbs and wearable robots, have been developed to expand and improve human capability. The robotic exoskeleton developed by Raytheon amplifies the wearer’s ability and enhances personal mobility. An integrated prosthetic arm prototype that can be controlled naturally has been developed by an international team led by Johns Hopkins University under DARPA sponsorship. The arm, Proto 1, provides sensory feedback and allows for eight degrees of freedom—a level of control far beyond the current state of the art for prosthetic limbs.
Advances in computing, sensing, networking, and communication technologies have led to the development of distributed robotics and multirobot systems for performing complex tasks in dynamic and challenging environments. Applications include search and rescue, reconnaissance, cleanups of toxic spills, firefighting, and planetary exploration.
Swarm robotics envisions large numbers of mostly simple robots. It is inspired by swarm intelligence, the principle of cooperation observed in colonies of ants and bees. The swarm may consist of heterogeneous robots, differing in the type of sensors, manipulators, and computational power. Robots can communicate by wireless transmission systems.
Potential applications for swarm robotics include tasks that demand miniaturization, like distributed sensing tasks in micromachinery or the human body; tasks that demand cheap design, such as mining or agricultural foraging; and tasks for which failure can be very costly, such as planetary exploration. A current swarm robotic project is the shape-shifting, or “claytronic,” robots of Carnegie Mellon University. The pocket-size, cylindrical wheeled robots of the swarm use electromagnetic forces to cling together, and to assume different shapes.
Inputs have been taken from “memagazine” and wiki.