Adaptive Action Selection Mechanisms for Evolutionary Multimodular Robotics
نویسندگان
چکیده
This paper focuses on the well-known problem in behavioral robotics – “what to do next”. The problem addressed here lies in the selection of one activity to be executed from multiple regulative, homeostatic and developmental processes running onboard a reconfigurable multi-robot organism. We consider adaptive hardware and software frameworks and argue the non-triviality of action selection for evolutionary robotics. The paper overviews several deliberative, evolutionary and bio-inspired approaches for such an adaptive action selection mechanism.
منابع مشابه
Selective attention enables action selection: evidence from evolutionary robotics experiments
In this paper we investigate whether selective attention enables the development of action selection (i.e. the ability to select among conflicting actions afforded by the current agent/environmental context). By carrying out a series of experiments in which neuro-robots have been evolved for the ability to forage so to maximize the energy that can be extracted from ingested substances we observ...
متن کاملRobotic Organisms - Artificial Homeostatic Hormone System and Virtual Embryogenesis as Examples for Adaptive Reaction-Diffusion Controllers
Generating artificial organisms from robotic modules (cells) is a challenging and fascinating task. In this article, we discuss the most important challenges that such systems pose to the engineer and how these challenges are met by introducing bio-inspired mechanisms into the core functionality of such robotic systems. We develop a list of key functionalities that allow such a system to overco...
متن کاملUtvikling av altruisme med gjenkjennelse av slektskap i evolusjonære roboter
Swarm robotics provides ways of solving problems using groups of autonomous robots that display emergent intelligent behaviour. Control mechanisms for these robots can be designed by using mechanisms from evolutionary robotics. In order to ensure that the controllers are adaptive and capable of learning new behaviour controllers are artificially evolved using techniques inspired by natural evol...
متن کاملFrom natural to artificial life: Biomimetic mechanisms in animat designs
This paper describes several models that incorporate some biomimetic mechanisms into the control architecture of an animat that has to survive in a changing environment. The adaptive capacities of these mechanisms are compared to engineering solutions, in application domains that involve navigation, action selection, and evolutionary design. It is argued that, although impressive adaptive capac...
متن کاملDesigning adaptive humanoid robots through the FARSA open-source framework
We introduce FARSA, an open-source Framework for Autonomous Robotics Simulation and Analysis, that allows us to easily set up and carry on adaptive experiments involving complex robot/environmental models. Moreover, we show how a simulated iCub robot can be trained, through an evolutionary algorithm, to display reaching and integrated reaching and grasping behaviours. The results demonstrate ho...
متن کامل