Biological models have inspired algorithm designers at least since 1954, with Barricelli’s simulations of evolution [1]. His motive as a biologist was the study of evolution, but engineers realized that many practical problems could be addressed with these mechanisms, leading to a plethora of evolutionary models including genetic algorithms, genetic programming, evolution strategies, and evolutionary programming. Other biologists used the computer to study how a collection of animals, such as an ant colony, could organize its behavior without central direction. Again, what began as an exercise in synthetic biology proved to have applications in decentralized control for engineering applications, leading to a field that since Beni and Wang’s work on cellular robotics in 1989 (NATO Advanced Workshop on Robots and Biological Systems, Tuscany, Italy, June 26–30) has been known as swarm intelligence. Early applications were inspired by pheromone coordination in ants, but both biologists and engineers have continued to articulate and exploit other coordination methods from nature. This volume offers a survey of some of the more recent examples.

Of the seven algorithms in the book, five originate in the research of Xin-She Yang and his associates, and four of these are discussed in much more detail in Yang’s 2014 volume [2]. Yang situates these algorithms, as well as algorithms inspired by physics and evolution that the present volume does not consider, in a general model of optimization. The added value of this book is in discussing three algorithms not presented by Yang, and in collecting some references to more recent applications and developments in the field. The biological models in the present volume include bats, fish swarms, cuckoos, fireflies, flower pollination, bee colony coordination, and the search behavior of wolves.

The book’s introductory chapter discusses nature-inspired algorithms in general, narrows the focus to swarming, and provides high-level mathematical background. After a chapter on each of the biological models, a final chapter classifies them along eight different dimensions: how the agents share information, what probability distributions are used, the number of different behaviors across the agents in an algorithm, how agents exploit their positional distribution, the number of control parameters, whether and how agents are replaced during execution, how agent velocity is exploited, and the type of exploration and exploitation used. Not all of these distinctions are intrinsic to the basic algorithm. For example, most algorithms are agnostic to the probability distributions that drive stochastic choices, and different implementations could very well use different distributions without changing the fundamental flavor of the algorithm.

The book provides a convenient literature update on a range of biological models for search and optimization. However, it has one major weakness. A biologist who is studying a given species needs no justification to build a simulation of that species’ behavior, as a way of testing hypotheses about the mechanisms that are operating in an animal community. But an engineer who proposes to use a biological model to solve real problems owes it to his readers to describe the kinds of problems for which the algorithm is particularly well suited, and to show that it outperforms alternatives, including the extensive body of work on swarm intelligence that has been done over the last quarter-century. This volume provides no guidance on either of these important questions, and readers intrigued by some of the innovative algorithms it offers will need to look elsewhere for reliable engineering guidance.