Robot Evolution

Nyhet

From Evolutionary Robotics to Physical AI

Inbunden, Engelska, 2026

AvA.E. Eiben,Karine Miras,Emma Hart,Agoston E Eiben

1 229 kr

Kommande

This book provides a synthesis of current, state-of-the-art scientific and technological knowledge concerning robot evolution. It explores the potential and limitations of the principle “If evolution can create intelligence, artificial evolution can create artificial intelligence”. It focusses on embodied artificial intelligence, where intelligence emerges from the interaction between a physical body and its environment as exemplified by robots. The long-term vision behind the book is that of robots that evolve, learn, and improve their bodies and brains ‘on the job’.The book consists of three parts. Part I elaborates on natural evolution, its digital counterpart, evolutionary computing, and then reviews the history of Evolutionary Robotics, highlighting significant milestones. Part II discusses how evolutionary robotics can be implemented. It lays the groundwork by outlining the fundamental components of any Evolutionary Robotics system and then turns to increasingly sophisticated systems for controller evolution, the simultaneous evolution of morphology and control, and the augmentation of such systems with lifetime learning to improve evolved controllers. Part III looks toward the future, examining the prospect of the Evolution of Things, evolving physical robots in real time and real space, and highlighting emerging paradigms that are reshaping the field.The book is an entry point and a reference work, rich in case studies, design patterns, and adaptable algorithmic recipes that researchers and advanced students can reuse and extend. It is written as a didactic guide, structured to support both self-study and teaching. To enable hands-on exploration, the book is accompanied by open-source software, supporting experimentation, prototyping, and reproducible research.This volume serves as an accessible introduction to students in artificial intelligence, computer science, robotics, or related fields. AI experts, AI practitioners, roboticists, biologists and researchers in life science will find the book as a valuable reference work and an inspiration for their research.

Produktinformation

Utgivningsdatum2026-07-18
Mått155 x 235 x undefined mm
FormatInbunden
SpråkEngelska
SerieNatural Computing Series
FörlagSpringer Verlag, Singapore
ISBN9789819585816

Tillhör följande kategorier

Agoston E. Eiben is Professor of Computational Intelligence at Vrije Universiteit Amsterdam. He is a world-leading expert in evolutionary computing and evolutionary robotics, and (co-)author of the best-selling Introduction to Evolutionary Computing. His current research explores robots that can reproduce, evolve, and learn. He has published in leading venues such as Nature, Science Robotics, and Nature Machine Intelligence, and his work has received extensive international media coverage in more than ten countries. His long-term goal is to show how artificial evolution can yield physical artificial intelligence, and to deepen our understanding of the interplay between the body and the brain.Karine Miras is an Assistant Professor of Computational Intelligence at Vrije Universiteit Amsterdam, working at the intersection of Evolutionary Robotics and Artificial Life. Her research has the goal of advancing autonomous systems and also deepening our understanding of biological evolution. At the core of her work is exploring how environments shape the traits of evolving creatures, and how developmental processes can be embedded directly into the genetic architectures of artificial life forms. “I’m interested in understanding what allows for the evolution of complex life through the study of artificial life.”Emma Hart is a Professor of Computational Intelligence at Edinburgh Napier University, Scotland, UK. Her research interests lie in the cross-cutting areas of evolutionary robotics, artificial intelligence, and optimisation, with a particular focus on combining methods from evolutionary computing and machine-learning to develop systems that learn autonomously over time and improve their own performance with experience, whether in robotics or in solving real-world optimisation problems. She is Fellow of the Royal Society of Edinburgh, and a Senior Member of ACM.