This paper describes a study in evolutionary robotics conducted completely in hardware without using simulations. The experiments employ on-line evolution, where robot controllers evolve on-the-fly in the robots' ...
详细信息
ISBN:
(纸本)9783319311531
This paper describes a study in evolutionary robotics conducted completely in hardware without using simulations. The experiments employ on-line evolution, where robot controllers evolve on-the-fly in the robots' environment as the robots perform their tasks. The main issue we consider is the feasibility of tackling a non-trivial task in a realistic timeframe. In particular, we investigate whether a population of six robots can evolve foraging behaviour in one hour. The experiments demonstrate that this is possible and they also shed light on some of the important features of our evolutionary system. Further to the specific results we also advocate the system itself. It provides an example of a replicable and affordable experimental set-up for other researches to engage in research into on-line evolution in a population of real robots.
Continuous-time recurrent neural networks affected by random additive noise are evolved to produce phototactic behaviour in simulated mobile agents. The resulting neurocontrollers are evaluated after evolution against...
详细信息
ISBN:
(纸本)9783540691334
Continuous-time recurrent neural networks affected by random additive noise are evolved to produce phototactic behaviour in simulated mobile agents. The resulting neurocontrollers are evaluated after evolution against perturbations and for different levels of neural noise. Controllers evolved with neural noise are more robust and may still function in the absence of noise. Evidence from behavioural tests indicates that robust controllers do not undergo noise-induced bifurcations or if they do, the transient dynamics remain functional. A general hypothesis is proposed according to which evolution implicitly selects neural systems that operate in noise-resistant landscapes which are hard to bifurcate and/or bifurcate while retaining functionality.
This paper presents a new algorithm for distributed on-line evolutionary learning in swarm robotics. The challenge we address is to cope with the limited computation and communication capabilities of low cost robots, ...
详细信息
ISBN:
(纸本)9781728169293
This paper presents a new algorithm for distributed on-line evolutionary learning in swarm robotics. The challenge we address is to cope with the limited computation and communication capabilities of low cost robots, which are often used in swarm robotics. In order to do so, the algorithm decouples computation and communication and ensures learning of efficient control policies even when only a limited amount of information can be exchanged between neighbouring robots. We show experimentally that this algorithm is both remarkably robust with respect to its meta-parameter values, and able to adapt automatically to the available communication bandwidth.
This paper demonstrates how individual and collective behaviours of robots can be evolved using a novel technique of applying a genetic algorithm on a lookup table based chromosome. The evolved behaviours are: orbitin...
详细信息
ISBN:
(纸本)9781479984640
This paper demonstrates how individual and collective behaviours of robots can be evolved using a novel technique of applying a genetic algorithm on a lookup table based chromosome. The evolved behaviours are: orbiting a stationary robot;navigation between three robots;follow the leader using six robots;and robot dispersal where the robots move away from each other. These behaviours are based on those used by Harvard University when demonstrating some of the collective behaviours that can be implemented by the Kilobot robot. With careful selection of the lookup tables and fitness functions all the above behaviours can be successfully evolved.
Online evolution of controllers on real robots typically requires a prohibitively long evolution time. One potential solution is to distribute the evolutionary algorithm across a group of robots and evolve controllers...
详细信息
ISBN:
(纸本)9783319234854;9783319234847
Online evolution of controllers on real robots typically requires a prohibitively long evolution time. One potential solution is to distribute the evolutionary algorithm across a group of robots and evolve controllers in parallel. No systematic study on the scalability properties and dynamics of such algorithms with respect to the group size has, however, been conducted to date. In this paper, we present a case study on the scalability of online evolution. The algorithm used is odNEAT, which evolves artificial neural network controllers. We assess the scalability properties of odNEAT in four tasks with varying numbers of simulated e-puck-like robots. We show how online evolution algorithms can enable groups of different size to leverage their multiplicity, and how larger groups can: (i) achieve superior task performance, and (ii) enable a significant reduction in the evolution time and in the number of evaluations required to evolve controllers that solve the task.
Exploration of the search space through the optimisation of phenotypic diversity is of increasing interest within the field of evolutionary robotics. Novelty search and the more recent MAP-Elites are two state of the ...
详细信息
ISBN:
(纸本)9783319458236;9783319458229
Exploration of the search space through the optimisation of phenotypic diversity is of increasing interest within the field of evolutionary robotics. Novelty search and the more recent MAP-Elites are two state of the art evolutionary algorithms which diversify low dimensional phenotypic traits for divergent exploration. In this paper we introduce a novel alternative for rapid divergent search of the feature space. Unlike previous phenotypic search procedures, our proposed Spatial, Hierarchical, Illuminated Neuro-Evolution (SHINE) algorithm utilises a tree structure for the maintenance and selection of potential candidates. SHINE penalises previous solutions in more crowded areas of the landscape. Our experimental results show that SHINE significantly outperforms novelty search and MAP-Elites in both performance and exploration. We conclude that the SHINE algorithm is a viable method for rapid divergent search of low dimensional, phenotypic landscapes.
Spiking neural networks (SNNs) are neuroscience-inspired computational systems that carry out computation based on the biological modeling of neuron interactions. Current SNN studies have shown their ability to solve ...
详细信息
ISBN:
(纸本)9781728125473
Spiking neural networks (SNNs) are neuroscience-inspired computational systems that carry out computation based on the biological modeling of neuron interactions. Current SNN studies have shown their ability to solve a wide variety of machine learning problems. The temporal dynamics and future low-power neuromorphic implementations of SNNs also make them suitable controller candidates for embedded applications, especially for robotic platforms with very low payload and power budgets (e.g. Micro Air Vehicles). In this paper, we present a solution to simulate full control of a hexacopter UAV in 6 degrees of freedom using SNNs. By decomposing the neurocontroller into modules, we demonstrate that the development of UAV Bight control can be accomplished by an incremental evolutionary approach using a modified NEAT algorithm.
The presence of functional diversity within a group has been demonstrated to lead to greater robustness, higher performance and increased problem-solving ability in a broad range of studies that includes insect groups...
详细信息
ISBN:
(纸本)9781450356183
The presence of functional diversity within a group has been demonstrated to lead to greater robustness, higher performance and increased problem-solving ability in a broad range of studies that includes insect groups, human groups and swarm robotics. Evolving group diversity however has proved challenging within evolutionary robotics, requiring reproductive isolation and careful attention to population size and selection mechanisms. To tackle this issue, we introduce a novel, decentralised, variant of the MAP-Elites illumination algorithm which is hybridised with a well-known distributed evolutionary algorithm (mEDEA). The algorithm simultaneously evolves multiple diverse behaviours for multiple robots, with respect to a simple token-gathering task. Each robot in the swarm maintains a local archive defined by two pre-specified functional traits which is shared with robots it come into contact with. We investigate four different strategies for sharing, exploiting and combining local archives and compare results to mEDEA. Experimental results show that in contrast to previous claims, it is possible to evolve a functionally diverse swarm without geographical isolation, and that the new method outperforms mEDEA in terms of the diversity, coverage and precision of the evolved swarm.
Diverse, complex, and adaptive animal behaviors are achieved by organizing hierarchically structured controllers in motor systems. The levels of control progress from simple spinal reflexes and central pattern generat...
详细信息
Diverse, complex, and adaptive animal behaviors are achieved by organizing hierarchically structured controllers in motor systems. The levels of control progress from simple spinal reflexes and central pattern generators through to executive cognitive control in the frontal cortex. Various types of hierarchical control structures have been introduced and shown to be effective in past artificial agent models, but few studies have shown how such structures can self-organize. This study describes how such hierarchical control may evolve in a simple recurrent neural network model implemented in a mobile robot. Topological constraints on information flow are found to improve system performance by decreasing interference between different parts of the network. One part becomes responsible for generating lower behavior primitives while another part evolves top-down sequencing of the primitives for achieving global goals. Fast and slow neuronal response dynamics are automatically generated in specific neurons of the lower and the higher levels, respectively. A hierarchical neural network is shown to outperform a comparable single-level network in controlling a mobile robot.
Scaffolding-initially simplifying the task environment of autonomous robots-has been shown to increase the probability of evolving robots capable of performing in more complex task environments. Recently, it has been ...
详细信息
ISBN:
(纸本)9781450305570
Scaffolding-initially simplifying the task environment of autonomous robots-has been shown to increase the probability of evolving robots capable of performing in more complex task environments. Recently, it has been shown that changes to the body of a robot may also scaffold the evolution of non trivial behavior. This raises the question of whether two different kinds of scaffolding (environmental and morphological) synergize with one another when combined. Here it is shown that, for legged robots evolved to perform phototaxis, synergy can be achieved, but only if morphological and environmental scaffolding are combined in a particular way: The robots must first undergo morphological scaffolding, followed by environmental scaffolding. This suggests that additional kinds of scaffolding may create additional synergies that lead to the evolution of increasingly complex robot behaviors.
暂无评论