On the cumulative effect of bloat and genetic transposition on the efficiency of incremental evolution of snake-like robot

Abstract

We present a study on the cumulative effect of the bloat and the seeding of the initial population, inspired by genetic transposition (GT), on the efficiency of incremental evolution of simulated snake-like robot (Snakebot). In the proposed incremental genetic programming (IGP), the task of coevolving the locomotion gaits and sensing of the bot in a challenging environment is decomposed into two sub-tasks, implemented as two consecutive evolutionary stages. First, we use genetic programming (GP) with two ways of bloat management, (i) parsimony pressure which penalizes the bloat and (ii) no bloat control, to evolve two pools of sensor-less Snakebots. During the second stage of IGP, we use these pools to seed the initial population of Snakebots applying two methods of seeding: canonical seeding and GT-inspired seeding. The empirical results indicate that the efficiency of the first stage of IGP for both bloat control techniques is similar. However, the bloated bots contribute to a much more efficient second stage of evolution. Compared to the canonical seeding with parsimony bots, the GT-inspired seeding with bloated Snakebots yields about five times higher probability of success and similar decrease of computational effort of the second stage of IGP.