Ecological Robotics: A Schema-theoretic Approach

Abstract

An in-depth understanding and dynamic modeling of the relationship a robot has with its environment (i.e., the overall ecology) is important to ensure that fielded robotic systems are:Not competing with other agents that can do the task more effectively and hence prove themselves useless.Successful competitors within the ecological system and can potentially displace less efficient agents.Ecologically sensitive so that agent-environmental system dynamics are well-modeled and as predictable as possible whenever new robotic technology is introduced.Little emphasis to date has been placed on this ecological approach within mobile robotics research, although some related research has been conducted in the recent past in tile context of the artificial life community. All too often, however, these approaches lack both a strong biological basis for their working assumptions and any formal underpinnings (neural, behavioral, and computational) for the results they obtain. We address these problems directly using schema theory and neurophysiological and ethological models to provide credible, generalizable, and useful results in this domain. These systems are currently grounded in robotic simulations and ultimately in actual robotic hardware.The study of sensory guided behaviors in living animals has become of general significance not only for scientists working in neuroscience and computational neuroscience but also for scientists working in robotics and distributed artificial intelligence, who are using functional principles generated from the study of living animals as models to build computer based automata that display complex sensorimotor behaviors. Our research effort, which follows these lines, is tied together by software tools including: NSL, a neural simulation language; ASL, an abstract schema language; and MissionLab, a schema-based mission-oriented simulation and robot implementation environment.

What is intrinsic motivation? A typology of computational approaches

“…Inspired by these kinds of motivation and their understanding by (neuro-) ethologists, roboticists have built machines endowed with similar systems with the aim of providing them with autonomy and properties of life-like intelligence (Arkin, 2005 ). For example sowbug-inspired robots (Endo and Arkin, 2001 ), praying mantis robots (Arkin et al, 1998 ) dog-like robots (Fujita et al, 2001 ) have been constructed.…”

A single computational model for many learning phenomena

“…Today's robots either evolve in what we may call their natural environment or they learn in the equivalent of an experimental laboratory. This is a limitation that robots as scientific theories will have to overcome because real animals both evolve and learn in their natural environment and, therefore, robotics should be an ecological robotics (Arkin, Cervantes-Perez, & Weitzenfeld, 1998;Duchon, Kaelbling, & Warren, 1998;Parisi, Cecconi, & Nolfi, 1990). Individual robots that learn in the equivalent of a psychologist's experimental laboratory tend to exhibit different behaviours because the connection weights that have been assigned to their neural network at the beginning of learning are random, not because they have different genes or they have had different experiences.…”

Multiscale Enaction Model (MEM): the case of complexity and “context-sensitivity” in vision

“…The reported control-law is a linear combination of flow and of the perceived direction of the visual target, weighted by the magnitude of flow ( Warren et al, 2001 ). Among recent developments, and beyond modeling of human behavior and applications in neural networks, ecological psychology has also found an allied in behavior-based robotics ( Arkin, 1998 ), specifically in a Gibsonian trend ( Duchon and Warren, 1994 , 2002 ) sometimes so-called “ecological robotics” ( Arkin et al, 1998 ; Duchon et al, 1998 ). For instance, Duchon and Warren (1994) noted that laws of control were applicable to any moving agent (see, for recent instantiations of the principle in the critical context of autonomous airborne navigation, Serres et al, 2006 ; Franceschini et al, 2007 ).…”