Evolutionary robotics is a new technique for the automatic creation of autonomous robots. Inspired by the Darwinian principle of selective reproduction of the fittest, it views robots as autonomous artificial organisms that develop their own skills in close interaction with the environment and without human intervention. Drawing heavily on biology and ethology, it uses the tools of neural networks, genetic algorithms, dynamic systems, and biomorphic engineering. The resulting robots share with simple biological systems the characteristics of robustness, simplicity, small size, flexibility, and modularity.
In evolutionary robotics, an initial population of artificial chromosomes, each encoding the control system of a robot, is randomly created and put into the environment. Each robot is then free to act (move, look around, manipulate) according to its genetically specified controller while its performance on various tasks is automatically evaluated. The fittest robots then "reproduce" by swapping parts of their genetic material with small random mutations. The process is repeated until the "birth" of a robot that satisfies the performance criteria.
This book describes the basic concepts and methodologies of evolutionary robotics and the results achieved so far. An important feature is the clear presentation of a set of empirical experiments of increasing complexity. Software with a graphic interface, freely available on a Web page, will allow the reader to replicate and vary (in simulation and on real robots) most of the experiments.
For the past three decades, the author and his colleagues in the MIT Man-Machine Systems Laboratory have been carrying out experimental research in the area of teleoperation, telerobotics, and supervisory control - a new form of technology that allows humans to work through machines in hazardous environments and control complex systems such as aircraft and nuclear power plants. This timely reference brings together a variety of theories and technologies that have emerged in a number of fields of application, describing common themes, presenting experiments and hardware embodiments as examples, and discussing the advantages and the drawbacks of this new form of human-machine interaction.
There are many places - such as outer space, the oceans, and nuclear, biologically, and chemically toxic environments - that are inaccessible or hazardous to humans but in which work needs to be done. Telerobotics - remote supervision by human operators of robotic or semiautomatic devices - is a way to enter these difficult environments. Yet it raises a host of problems, such as the retrieval of sensory information for the human operator, and how to control the remote devices with sufficient dexterity. In its complete coverage of the theoretical and technological aspects of telerobotics and human-computer cooperation in the control of complex systems, this book moves beyond the simplistic notion of humans versus automation to provide the necessary background for exploring a new and informed cooperative relationship between humans and machines. Thomas B. Sheridan is Professor of Engineering and Applied Psychology at the Massachusetts Institute of Technology.
Contents: Introduction. Theory and Models of Supervisory Control: Frameworks and Fragments. Supervisory Control of Anthropomorphic Teleoperators for Space, Undersea, and Other Applications. Supervisory Control in Transportation, Process, and Other Automated Systems. Social Implications of Telerobotics, Automation, and Supervisory Control.
Intelligence takes many forms. This exciting study explores the novel insight, based on well-established ethological principles, that animals, humans, and autonomous robots can all be analyzed as multi-task autonomous control systems. Biological adaptive systems, the authors argue, can in fact provide a better understanding of intelligence and rationality than that provided by traditional AI.
In this technically sophisticated, clearly written investigation of robot-animal analogies, McFarland and Bösser show that a bee's accuracy in navigating on a cloudy day and a moth's simple but effective hearing mechanisms have as much to teach us about intelligent behavior as human models. In defining intelligent behavior, what matters is the behavioral outcome, not the nature of the mechanism by which the outcome is achieved. Similarly, in designing robots capable of intelligent behavior, what matters is the behavioral outcome.
McFarland and Bösser address the problem of how to assess the consequences of robot behavior in a way that is meaningful in terms of the robot's intended role, comparing animal and robot in relation to rational behavior, goal seeking, task accomplishment, learning, and other important theoretical issues.
Foundations of Robotics presents the fundamental concepts and methodologies for the analysis, design, and control of robot manipulators. It explains the physical meaning of the concepts and equations used, and it provides, in an intuitively clear way, the necessary background in kinetics, linear algebra, and control theory. Illustrative examples appear throughout.
The author begins by discussing typical robot manipulator mechanisms and their controllers. He then devotes three chapters to the analysis of robot manipulator mechanisms. He covers the kinematics of robot manipulators, describing the motion of manipulator links and objects related to manipulation. A chapter on dynamics includes the derivation of the dynamic equations of motion, their use for control and simulation and the identification of inertial parameters. The final chapter develops the concept of manipulability.
The second half focuses on the control of robot manipulators. Various position-control algorithms that guide the manipulator's end effector along a desired trajectory are described Two typical methods used to control the contact force between the end effector and its environments are detailed For manipulators with redundant degrees of freedom, a technique to develop control algorithms for active utilization of the redundancy is described. Appendixes give compact reviews of the function atan2, pseudo inverses, singular-value decomposition, and Lyapunov stability theory.
What is 16 feet long, 10 feet high, weighs 6,000 pounds, has six legs, and can sprint at 8 mph and step over a 4 foot wall? The Adaptive Suspension Vehicle (ASV) described in this book.
Machines That Walk provides the first in depth treatment of the "statically stable walking machine" theory employed in the design of the ASV, the most sophisticated, self contained, and practical walking machine being developed today. Under construction at Ohio State University, the automatically terrain adaptive ASV has one human operator, can carry a 500 pound payload and is expected to have better fuel economy and mobility than that of conventional wheeled and tracked vehicles in rough terrain.
The development of the ASV is a milestone in robotics research, and Machines That Walk provides a wealth of research results in mobility, gait, static stability, leg design, and vertical geometry design. The authors' treatment of statically stable gait theory and actuator coordination is by far the most complete available.
The goal of neurotechnology is to confer the performance advantages of animal systems on robotic machines. Biomimetic robots differ from traditional robots in that they are agile, relatively cheap, and able to deal with real-world environments. The engineering of these robots requires a thorough understanding of the biological systems on which they are based, at both the biomechanical and physiological levels.
This book provides an in-depth overview of the field. The areas covered include myomorphic actuators, which mimic muscle action; neuromorphic sensors, which, like animal sensors, represent sensory modalities such as light, pressure, and motion in a labeled-line code; biomimetic controllers, based on the relatively simple control systems of invertebrate animals; and the autonomous behaviors that are based on an animal's selection of behaviors from a species-specific behavioral "library." The ultimate goal is to develop a truly autonomous robot, one able to navigate and interact with its environment solely on the basis of sensory feedback without prompting from a human operator.
Two important subproblems of computer vision are the detection and recognition of 2D objects in gray-level images. This book discusses the construction and training of models, computational approaches to efficient implementation, and parallel implementations in biologically plausible neural network architectures. The approach is based on statistical modeling and estimation, with an emphasis on simplicity, transparency, and computational efficiency.
The book describes a range of deformable template models, from coarse sparse models involving discrete, fast computations to more finely detailed models based on continuum formulations, involving intensive optimization. Each model is defined in terms of a subset of points on a reference grid (the template), a set of admissible instantiations of these points (deformations), and a statistical model for the data given a particular instantiation of the object present in the image. A recurring theme is a coarse to fine approach to the solution of vision problems. The book provides detailed descriptions of the algorithms used as well as the code, and the software and data sets are available on the Web.
The Simulation of Adaptive Behavior Conference brings together researchers from ethology, psychology, ecology, artificial intelligence, artificial life, robotics, computer science, engineering, and related fields to further understanding of the behaviors and underlying mechanisms that allow adaptation and survival in uncertain environments. The work presented focuses on robotic and computational experimentation with well-defined models that help to characterize and compare alternative organizational principles or architectures underlying adaptive behavior in both natural animals and synthetic animats.
The effort to explain the imitative abilities of humans and other animals draws on fields as diverse as animal behavior, artificial intelligence, computer science, comparative psychology, neuroscience, primatology, and linguistics. This volume represents a first step toward integrating research from those studying imitation in humans and other animals, and those studying imitation through the construction of computer software and robots.
Imitation is of particular importance in enabling robotic or software agents to share skills without the intervention of a programmer and in the more general context of interaction and collaboration between software agents and humans. Imitation provides a way for the agent—-whether biological or artificial—to establish a "social relationship" and learn about the demonstrator's actions, in order to include them in its own behavioral repertoire. Building robots and software agents that can imitate other artificial or human agents in an appropriate way involves complex problems of perception, experience, context, and action, solved in nature in various ways by animals that imitate.
Remote-controlled robots were first developed in the 1940s to handle radioactive materials. Trained experts now use them to explore deep in sea and space, to defuse bombs, and to clean up hazardous spills. Today robots can be controlled by anyone on the Internet. Such robots include cameras that not only allow us to look, but also go beyond Webcams: they enable us to control the telerobots' movements and actions.
This book summarizes the state of the art in Internet telerobots. It includes robots that navigate undersea, drive on Mars, visit museums, float in blimps, handle protein crystals, paint pictures, and hold human hands. The book describes eighteen systems, showing how they were designed, how they function online, and the engineering challenges they meet.