This book presents various contributions to the construction of life-like artifacts that globally show adaptation and evolution, as in the artificial life paradigm. Here, the meaning of hardware is extended beyond the typical meaning in computer science; it also considers mechanical and chemical components, to obtain real objects.

The book consists of 11 chapters, written by various authors, that are quite different in organization and approach. Chapter 1, “The History and Future of Stiquito: A Hexapod Insectoid Robot,” reports on a small robot built with nitinol muscles, controlled by a microprocessor, and sold as an educational kit. The chapter also reports on an extended analog computer (EAC)-based controller, used to implement the retina functions, and on developing a Stiquito colony and the reasons it failed.

Chapter 2, “Learning Legged Locomotion,” approaches one of the most debated problems in robotics: namely, how walking bipeds interact with the environment using their muscle-skeletal system and sensory information. The authors explore self-organization processes, not only in terms of the capability to induce physically meaningful patterns, but also to produce more intelligent behavior that adapts to the environment. This chapter is about methods, not hardware.

Chapter 3, “Salamandra Robotica: A Biologically Inspired Amphibious Robot that Swims and Walks,” is also about walking. It states that robots are useful tools for biology, and it illustrates the construction of an artificial salamander that is controlled by a central pattern generator (CPG), which enables it to swim, walk, and crawl. The mechanical and electrical hardware is illustrated, and the experiments are described.

Chapter 4, “Multilocomotion Robot: Novel Concept, Mechanism, and Control of Bio-inspired Robot,” suggests that bio-inspired robots will be our companions; therefore, they will need to move autonomously in every environment. In the animal world, different locomotion systems have been developed for different environments, so why not try different ways of walking for the same robot? The chapter illustrates the development of a gorilla-like robot that walks like both a biped and a quadruped.

Chapter 5, “Self-regulatory Hardware: Evolutionary Design for Mechanical Passivity on a Pseudo Passive Dynamic Walker,” focuses on biped walking and investigates how the mechanical structure of the legs may have evolved in morphology, the number of actuated degrees of freedom, and compliance. Experiments in simulation illustrate the evolutive design.

Chapter 6, “Perception for Action in Roving Robots: A Dynamical System Approach,” exploits chaotic dynamic systems to create the control architecture of a mobile robot that can integrate perception and action. The controller is implemented in a field-programmable gate array (FPGA), and real experiments are discussed.

Chapter 7, “Nature-inspired Single-electron Computers,” moves away from robotics and explores a computational paradigm that can be based on single-electron circuits and can perform computations inspired by chemical and biological systems. The proposed circuit is discussed for four cases: computation of a Voronoi diagram, spike detection of neurons, weak signal transmission, and dendritic pattern generation. Nanotechnology will make the real construction of such computers possible.

Chapter 8, “Tribolon: Water-based Self-assembly Robots,” explores the self-assembly of devices and robots--specifically, stochastically self-reconfigurable robots. The Tribolon project is an experiment to separately solve the three main needs of robotics--actuation, battery, and connector--by constructing tile-like elements, either passive or self-propelled, that move in water and produce different shapes.

Chapter 9, “Artificial Symbiosis in EcoBots,” explores the problem of energy autonomy, taking examples from the symbiosis mechanism of real living. The authors’ system uses a bio-electrochemical transducer to convert biochemical energy (from bacteria metabolism) into electricity. The system was tested on real conventional robots; it gave them significant autonomy, without the need for other energy sources.

Chapter 10, “The Phi-Bot: A Robot Controlled by a Slime Mould,” reports on a robot controller that is based on Physarum polycephalum plasmodium, a unicellular organism that changes rhythms according to the stimulus, thus behaving like a simple computational mechanism. The chapter presents different generations of controllers for exapod and wheeled robots. Moreover, it discusses the computational system’s syntax and semantics, including how the information is pragmatically represented as the behavioral structure of the physical system.

Chapter 11, “Reaction-diffusion Controllers for Robots,” presents another approach to creating a controller: Belousov-Zhabotinsky chemical reactions that have a spontaneous oscillatory behavior or, in altered conditions, show excitability. Chemical reaction-diffusion processors can be used as robot controllers, and the chapter shows how to combine a chemical processor--encapsulated in the Physarum polycephalum--with conventional hardware. The plasmodium can be considered a reaction-diffusion robot, since it can compute and modify the environment. A short index concludes the book.

The chapters present different systems and very different methodologies. A few chapters focus mainly on hardware, and hardware is often more mechanical than electrical. Some of the presentations are based only on the integration of previous papers; in general, they offer a new synthesis or new data. The end-of-chapter discussions nicely analyze the current state and future of the approaches.

The book is unique in that it attempts to mix various ideas on building artificial life in hardware--a hot topic that is still in its infancy. However, my initial excitement turned to cautious optimism. Regarding the experiments, some are quite old, some have a limited time horizon, and some are only weakly integrated to the design of a new generation of artifacts. While the beginning chapters present more classic and stable ideas, the book moves toward new methods and tools that represent the cutting edge of research. How fast this edge will move depends on advancements in other fields than just artificial life, so this picture is only a snapshot. Currently, this book is appropriate for research; in the long term, it can serve as a reference.