Applied Soft Computing 11 (2011) 2286–2299 Contents lists available at ScienceDirect Applied Soft Computing journal homepage: www.elsevier.com/locate/asoc Evolutionary algorithm for a genetic robot’s personality Kang-Hee Lee ∗ Department of Global School of Media, Soongsil University, 511 Sando-dong, Dongjak-gu, Seoul 156-743, Republic of Korea article info Article history: Received 30 December 2009 Received in revised form 20 June 2010 Accepted 1 August 2010 Available online 7 August 2010 Keywords: Genetic robot Artificial creature Artificial chromosome Robot genome Robot personality Evolutionary algorithm Mobile phone abstract This paper proposes a new concept of the genetic robot characterized by its own robot genome in which each chromosome consists of many genes that contribute to defining the robot’s personality. The large number of genes allows for a highly complex system, however it becomes increasingly difficult and time- consuming to ensure reliability, variability and consistency for the robot’s personality while manually initializing values for the individual genes. To overcome this difficulty, this paper also proposes an evolu- tionary algorithm for evolving a genetic robot’s personality (EAGRP) in a mobile phone. EAGRP evolves a gene pool that customizes the robot’s genome so that it closely matches a simplified set of personality fea- tures desired by the user. It acts on a two-dimensional individual, composed of artificial chromosomes, upon which a new masking method, the Eliza–Meme scheme, is used to derive a plausible individual given the restricted preference settings. This paper also proposes a crossover method that enables repro- duction for the two-dimensional genome. Finally, an evaluation procedure for individuals is carried out in a virtual environment using tailored perception scenarios. © 2010 Elsevier B.V. All rights reserved. 1. Introduction Since “The Origin of Species” by Charles Darwin in 1859, the concepts of gene and its evolution have become widely accepted around the world. Richard Dawkins claimed that “We and other ani- mals are machines created by our genes [1]” and also emphasized the importance of genes by comparing the evolutionary design which is “blind” to the purpose of the animal to the design by a designer who specifies the job to be performed [2]. Sloman thought the gene as a very powerful thing with its implicit goals and the way it functions [3]. Similarly, Roll prescribed that “Genes can spec- ify the goal for action [4].” Thus, the genes are considered as key components in defining a creature’s personality and the essence of this research should be on the genome of various types of artificial creatures such as pet-type, humanoid-type, or head-type, which can be implemented in either a real hardware robot or a simulated software robot (sobot) [5–10]. When interaction with users is required, the personality of the artificial creatures needs to be taken into consideration as the personality is crucial in building a believable emotional creature. Having a diverse personality is important because, “Personality is the engine of behavior” [11]. It can be encoded as an inherited trait, which decides the behavior based on an internal state in response to the stimulus. It is characterized by the Big Five personality dimen- sions [12,13]. This allows for the creation of diverse personalities ∗ Tel.: +82 2 828 7270; fax: +82 2 828 7270. E-mail address: kanghee.lee@ssu.ac.kr. for the agent, e.g. expression of highly agreeable and also (at the other end of the scale), highly antagonistic characteristics. In spite of the importance of innate personality in deciding the behavior, the concretization of diverse personalities have not been investigated to any great extent in previous research ven- tures, compared to that of artificial intelligence that is accumulated through experience and knowledge as posterior information dur- ing its lifetime by means of learning or interaction with human. A general approach to doing this concretization would be to identify a generative mechanism that might have the power to compose a variety of particular states and behaviors simply by varying a few parameters [11]. It is necessary in producing a really com- plex, multifunctional design [14]. Genetic encoding and algorithm for modularity and reusability can serve as an engine of consis- tency and coherence economically [3]. Evolutionary techniques are being strongly recommended to encode the features (geno- type) that would lead to ‘emotional behavior’ (phenotype) and that would generate different emotional systems with the architecture which is totally hand-coded and therefore very difficult to tune [15]. Motivated by engineered concepts of genes above, this paper proposes a genetic robot defined as an artificial creature which has its own genome that is composed of multiple artificial chromo- somes. Each chromosome in a robot genome consists of many genes that contribute to defining the robot’s personality. They are also crucial factors in determining its internal state and external behav- iors at any moment in time. The structure employing the robot genome provides primary advantages for artificial reproduction, reusability among robots, and the ability to evolve. The large num- ber of genes also allows for a highly complex system, however it 1568-4946/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.asoc.2010.08.010