Genetic transposition inspired incremental genetic programming for efficient coevolution of locomotion and sensing of simulated snake-like robot

摘要

Genetic transposition (GT) is a process of moving sequences of DNA to different positions within the genome of a single cell.It is recognized that the transposons (the jumping genes) facilitate the evolution of increasingly complex forms of life by providing the creative playground for the mutation where the latter could experiment with developing novel genetic structures without the risk of damaging the already existing, well-functioning genome.In this work we investigate the effect of a GT-inspired mechanism on the efficiency of genetic programming (GP) employed for coevolution of locomotion gaits and sensing of the simulated snake like robot (Snakebot).In the proposed approach, the task of coevolving the locomotion and the sensing morphology of Snakebot in a challenging environment is decomposed into two subtasks, implemented as two consecutive evolutionary stages.At first stage we employ GP to evolve a pool of simple, sensorless bots that are able to move fast in a smooth, open terrain.Then, during the second stage, we use these Snakebots to seed the initial population of the bots that are further subjected to coevolution of their locomotion control and sensing in a more challenging environment.For the second phase the seed is used as it is to create only part of a new individual, and the rest of the new individual's genetic makeup is created by a mutant copy of the seed.Experimental results suggest that the proposed two-staged GT inspired incremental evolution contributes to significant increase in the efficiency of the evolution of fast moving and sensing Snakebots.