Spatial, temporal, and modulatory factors affecting the evolvability of GasNets-a style of artificial neural network incorporating an analogue of volume signalling-are investigated. The focus of the article is a compa...
详细信息
Spatial, temporal, and modulatory factors affecting the evolvability of GasNets-a style of artificial neural network incorporating an analogue of volume signalling-are investigated. The focus of the article is a comparative study of variants of the GasNet, implementing various spatial, temporal, and modulatory constraints, used as control systems in an evolutionary robotics task involving visual discrimination. The results of the study are discussed in the context of related research. (C) 2010 Wiley Periodicals, Inc. Complexity 16: 35-44, 2010
Embodied evolution (EE) is a methodology in evolutionary robotics in which, without simulations on a host computer, real robots evolve on the basis of their interactions with the actual environment. However, when adop...
详细信息
Embodied evolution (EE) is a methodology in evolutionary robotics in which, without simulations on a host computer, real robots evolve on the basis of their interactions with the actual environment. However, when adopting EE, we had to accept robot behavior with a low fitness, especially in the early generations. This article introduces pre-evaluation into the EE framework for a biped robot in order to restrain the behavior of a robot of which the fitness is estimated to be low, especially falling down onto the ground. We provide a comparative discussion on the conventional simulate-and-transfer method, the original EE method, and the proposed one in terms of calculation time, cost of fitness evaluation, and cost of simulation or modeling based on the evaluation experiments. We believe that the EE framework with pre-evaluation is applicable to a wide variety of optimization tasks in which the cost or the risk of fitness evaluation is not negligible.
There is a growing trend in the cognitive sciences to conceive of cognitive behavior as being distributed across brain, body and environment. However, the implications of such distribution for our understanding of bio...
详细信息
ISBN:
(纸本)9781424481262
There is a growing trend in the cognitive sciences to conceive of cognitive behavior as being distributed across brain, body and environment. However, the implications of such distribution for our understanding of biological robustness, which so far has been related to individual-based mechanisms alone, has rarely been discussed in the literature. We used the evolutionary robotics technique to examine the relationship between distributed behavioral mechanisms and behavioral robustness. Two kinds of model agents were evolved for a mobile object-tracking task and tested to see whether they can sustain their behavior despite sensorimotor perturbations. The results indicate that a highly distributed realization of behavior can be (i) detrimental, if it is mostly based on factors that are necessary for the behavior, or (ii) beneficial, if it is mostly based on factors that are sufficient for the behavior. Accordingly, we suggest that future discussions of distributed cognition should take into account that there are at least two different possible modes of realizing distributed behavior and that these have a qualitatively different effect on behavioral robustness.
Genetic Algorithms and Genetic programming have been used extensively in evolutionary robotics (ER) with the goal of automatic programming of robotic controllers and has shown to be a promising approach. In this paper...
详细信息
ISBN:
(纸本)9781424481262
Genetic Algorithms and Genetic programming have been used extensively in evolutionary robotics (ER) with the goal of automatic programming of robotic controllers and has shown to be a promising approach. In this paper, we demonstrate the use of Gene Expression Programming, GEP, a newly developed evolutionary algorithm akin to GA and GP, to evolve robotic behaviours. We use the already well known obstacle avoidance behaviour for our initial work. The behaviour can be regarded as emergent when the main aim is to develop a wandering/exploratory behaviour. From our investigations, we show that GEP is able to learn controllers for a number of different environments. Moreover, standard GEP has never been used before in evolving robotic behaviours, however due to its reported good performances in other fields, we feel it has the capability to be used in ER.
This paper presents results from two sets of experiments which investigate how strategies used by embodied dynamical agents in a simple braking task are affected by the perceptual information that the agents receive. ...
详细信息
ISBN:
(纸本)9783642151927
This paper presents results from two sets of experiments which investigate how strategies used by embodied dynamical agents in a simple braking task are affected by the perceptual information that the agents receive. Agents are evolved in a simple 2D environment containing one stationary object. The task of the agents is to stop as close as possible to the object without hitting it. The results of these experiments demonstrate that most of the evolved agents use an impulsive braking strategy, in which deceleration is not controlled continuously. Potential causes of this impulsive braking strategy and possible future directions are discussed.
Designing robots and robot controllers is a highly complex and often expensive task. However, genetic programming provides an automated design strategy to evolve complex controllers based on evolution in nature. We sh...
详细信息
ISBN:
(纸本)9789896740214
Designing robots and robot controllers is a highly complex and often expensive task. However, genetic programming provides an automated design strategy to evolve complex controllers based on evolution in nature. We show that, even with limited computational resources, genetic programming is able to evolve efficient robot controllers for corridor following in a simulation environment. Therefore, a mixed and gradual form of layered learning is used, resulting in very robust and efficient controllers. Furthermore, the controller is successfully applied to real environments as well.
We study on artificial neural network-based controllers which are either trained or evolved by using the supervised or unsupervised learning approach. We employed backpropagation for the supervised method and the gene...
详细信息
ISBN:
(纸本)9780769542539
We study on artificial neural network-based controllers which are either trained or evolved by using the supervised or unsupervised learning approach. We employed backpropagation for the supervised method and the genetic algorithm for the unsupervised method. After training the controllers, we applied the controllers to our three newly designed mini-3D games. We performed a comprehensive study on the performance and weaknesses of the controllers. We emerged the controllers as fundamental tools for giving us more understanding about artificial neural network and its effectiveness in imitating players' behaviours.
Swarm robotics (SR) is the research field of multi-robot systems, which consist of many homogeneous autonomous robots without any types of global controllers. Generally, since a task given to this system cannot be ach...
详细信息
ISBN:
(纸本)9781424481262
Swarm robotics (SR) is the research field of multi-robot systems, which consist of many homogeneous autonomous robots without any types of global controllers. Generally, since a task given to this system cannot be achieved by a single robot, cooperative behavior is expected to emerge in a robotic swarm by a certain mechanism, which is through the interactions among robots or with an environment. In this paper, an evolutionary robotics approach, in which robot controllers are designed by evolving artificial neural networks, is adopted. Among the many approaches to evolving artificial neural networks, two approaches, NEAT and MBEANN are adopted for conducting computer simulations. Although a conventional neural network has a fixed topology and evolves only with its synaptic weights, NEAT and MBEANN evolve not only with their synaptic weights, but also with their topologies. As a benchmark for swarm robotics, cooperative package-pushing problems using ten autonomous robots are conducted to evaluate their performance. The behavioral characteristics that emerge are then discussed.
This talk will outline challenges and opportunities in translating evolutionary learning of autonomous robotics from simulation to reality. It covers evolution and adaptation of both morphology and control, hybrid co-...
详细信息
ISBN:
(纸本)9781586035952
This talk will outline challenges and opportunities in translating evolutionary learning of autonomous robotics from simulation to reality. It covers evolution and adaptation of both morphology and control, hybrid co-evolution of reality and simulation, handling noise and uncertainty, and morphological adaptation in hardware.
For the re-evolution of the mobile robot behavior in unknown environments, the mapping relation was constructed between input of sensors and output of actuators based on echo state network. An algorithm of adaptive be...
详细信息
ISBN:
(纸本)9781424467129
For the re-evolution of the mobile robot behavior in unknown environments, the mapping relation was constructed between input of sensors and output of actuators based on echo state network. An algorithm of adaptive behavior learning was presented based on echo state network for evolutionary robotics. The composite architecture with resposive behavior and behavior learning was adopted. The resposive behavior was drived by the samples composed with sensor information and decision. The weights of echo state network were optimized via (mu+lambda)-evolution strategy. The new control rules were generated via evolutionary algorithms, and new samples were added to the database constantly. The high intelligent behaviors of robot were transmitted to resposive behaviors. The experimental results indicate that the proposed approach has a better adaptability.
暂无评论