18Evolutionary Methods
18.1 Introduction
In this chapter, we outline prominent evolutionary computation methods, including particle swarm optimization in Section 18.2, differential evolution in Section 18.3 and grammatical evolution in Section 18.4. For each one, we present a numerical example to illustrate the mechanics of implementing the method in any language and software environment.
We would like to point out the ever powerful parallelizable characteristics of evolutionary algorithms, which make them excellent candidates for very‐large‐scale problems. One can easily envision that, in the near future, millions of wetware computers running a particle swarm optimization algorithm in parallel can solve a space travel problem with low overhead. Similarly, the best flight control program for an autonomous spacecraft exploring galaxies can be generated by a grammatical evolution algorithm using a realistic grammar and running in parallel on a few million biological computers.
18.2 Particle Swarm Optimization
Based on mechanisms observed in biological organizations, swarm artificial intelligence (SAI) is used to control and analyse the collective behaviour of decentralized autonomous systems [132]. The agents in SAI systems interact with their nearest neighbours and their external environment to manifest a global intelligence.
Self‐organization can be defined as a capacity that emerges from interactions among lower‐level components that have not been explicitly designed to ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access