Evolutionary robotics
- Exclusive ContentScan the cover in the app
- CommunityDiscuss & Share
About this book
Rather than programming robots step by step, evolutionary robotics builds them automatically, borrowing the logic of natural selection. Its designers treat each robot as a self-sufficient artificial organism that learns its abilities through its surroundings, with no human hand guiding the process. The field leans heavily on biology and animal behaviour research and combines neural networks, genetic algorithms, dynamical systems, and biomorphic design. Robots built this way tend to resemble simple living creatures: they are sturdy, uncomplicated, compact, adaptable, and made of interchangeable parts.
The method starts with a group of artificial chromosomes, each one encoding the control system of a robot. These are placed into an environment, where every machine moves, observes, and handles objects under the direction of its genetic controller, and its success on several tasks is measured automatically. The best performers pass on their genetic material by exchanging segments with one another, and small random mutations are introduced along the way. This cycle repeats until a robot appears that meets the required performance standards.
The book lays out the core ideas and methods of evolutionary robotics and reviews the results obtained so far. Its main strength is a sequence of empirical experiments that grow progressively more complex. Accompanying graphical software, available for free on a website, lets readers reproduce most of these experiments and modify them, both in simulation and on physical robots.
Reader reviews
No reviews yet. Share your thoughts on this book in the Appilog app.
Book Details
- ISBN
- 9789812568700
- Author
- Lingfeng Wang
- Publisher
- Unknown Publisher
- Genres
- Science & MathematicsTechnology & Computing
Media & Extras
This book has 0 videos that play on its cover. Scan the cover with the Appilog app to watch.
Get the app



