scholarly journals Augmenting Scalable Communication-Based Role Allocation for a Three-Role Task

2020 ◽  
Vol 22 (64) ◽  
pp. 152-165
Author(s):  
Gustavo Martins ◽  
Paulo Urbano ◽  
Anders Lyhne Christensen
Keyword(s):  
Fitness Function ◽  
A Priori ◽  
Evolutionary Process ◽  
Evolutionary Robotics ◽  
Cognitive Architectures ◽  
Local Conditions ◽  
Fitness Functions ◽  
Signal Differentiation ◽  
Scalable Communication ◽  
Role Allocation

In evolutionary robotics role allocation studies, it is common that the role assumed by each robot is strongly associated with specific local conditions, which may compromise scalability and robustness because of the dependency on those conditions. To increase scalability, communication has been proposed as a means for robots to exchange signals that represent roles. This idea was successfully applied to evolve communication-based role allocation for a two-role task. However, it was necessary to reward signal differentiation in the fitness function, which is a serious limitation as it does not generalize to tasks where the number of roles is unknown a priori. In this paper, we show that rewarding signal differentiation is not necessary to evolve communication-based role allocation strategies for the given task, and we improve reported scalability, while requiring less a priori knowledge. Our approach for the two-role task puts fewer constrains on the evolutionary process and enhances the potential of evolving communication-based role allocation for more complex tasks. Furthermore, we conduct experiments for a three-role task where we compare two different cognitive architectures and several fitness functions and we show how scalable controllers might be evolved.

Engineering the Evolution of Self-Organizing Behaviors in Swarm Robotics: A Case Study

2011 ◽  
Vol 17 (3) ◽  
pp. 183-202 ◽  
Author(s):  
Vito Trianni ◽  
Stefano Nolfi
Keyword(s):  
Communication System ◽  
Swarm Robotics ◽  
A Priori ◽  
Evolutionary Process ◽  
Deep Understanding ◽  
Evolutionary Robotics ◽  
Powerful Approach ◽  
Large Groups ◽  
Self Organizing

Evolutionary robotics (ER) is a powerful approach for the automatic synthesis of robot controllers, as it requires little a priori knowledge about the problem to be solved in order to obtain good solutions. This is particularly true for collective and swarm robotics, in which the desired behavior of the group is an indirect result of the control and communication rules followed by each individual. However, the experimenter must make several arbitrary choices in setting up the evolutionary process, in order to define the correct selective pressures that can lead to the desired results. In some cases, only a deep understanding of the obtained results can point to the critical aspects that constrain the system, which can be later modified in order to re-engineer the evolutionary process towards better solutions. In this article, we discuss the problem of engineering the evolutionary machinery that can lead to the desired result in the swarm robotics context. We also present a case study about self-organizing synchronization in a swarm of robots, in which some arbitrarily chosen properties of the communication system hinder the scalability of the behavior to large groups. We show that by modifying the communication system, artificial evolution can synthesize behaviors that scale properly with the group size.

scholarly journals Evolving Self-Assembly in Autonomous Homogeneous Robots: Experiments with Two Physical Robots

2009 ◽  
Vol 15 (4) ◽  
pp. 465-484 ◽  
Author(s):  
Christos Ampatzis ◽  
Elio Tuci ◽  
Vito Trianni ◽  
Anders Lyhne Christensen ◽  
Marco Dorigo
Keyword(s):  
Self Assembly ◽  
Motor Coordination ◽  
Fitness Function ◽  
A Priori ◽  
Research Work ◽  
Evolutionary Robotics ◽  
Direct Access ◽  
Necessary Condition ◽  
Perceptual Cues ◽  
Dynamic Neural Networks

This research work illustrates an approach to the design of controllers for self-assembling robots in which the self-assembly is initiated and regulated by perceptual cues that are brought forth by the physical robots through their dynamical interactions. More specifically, we present a homogeneous control system that can achieve assembly between two modules (two fully autonomous robots) of a mobile self-reconfigurable system without a priori introduced behavioral or morphological heterogeneities. The controllers are dynamic neural networks evolved in simulation that directly control all the actuators of the two robots. The neurocontrollers cause the dynamic specialization of the robots by allocating roles between them based solely on their interaction. We show that the best evolved controller proves to be successful when tested on a real hardware platform, the swarm-bot. The performance achieved is similar to the one achieved by existing modular or behavior-based approaches, also due to the effect of an emergent recovery mechanism that was neither explicitly rewarded by the fitness function, nor observed during the evolutionary simulation. Our results suggest that direct access to the orientations or intentions of the other agents is not a necessary condition for robot coordination: Our robots coordinate without direct or explicit communication, contrary to what is assumed by most research works in collective robotics. This work also contributes to strengthening the evidence that evolutionary robotics is a design methodology that can tackle real-world tasks demanding fine sensory-motor coordination.

scholarly journals Encouraging Behavioral Diversity in Evolutionary Robotics: An Empirical Study

2012 ◽  
Vol 20 (1) ◽  
pp. 91-133 ◽  
Author(s):  
J.-B. Mouret ◽  
S. Doncieux
Keyword(s):  
Empirical Study ◽  
Infinite Number ◽  
Evolutionary Process ◽  
Evolutionary Robotics ◽  
Premature Convergence ◽  
Multi Objective ◽  
Behavioral Diversity ◽  
Fitness Functions ◽  
Common Framework ◽  
Or Task

Evolutionary robotics (ER) aims at automatically designing robots or controllers of robots without having to describe their inner workings. To reach this goal, ER researchers primarily employ phenotypes that can lead to an infinite number of robot behaviors and fitness functions that only reward the achievement of the task—and not how to achieve it. These choices make ER particularly prone to premature convergence. To tackle this problem, several papers recently proposed to explicitly encourage the diversity of the robot behaviors, rather than the diversity of the genotypes as in classic evolutionary optimization. Such an approach avoids the need to compute distances between structures and the pitfalls of the noninjectivity of the phenotype/behavior relation; however, it also introduces new questions: how to compare behavior? should this comparison be task specific? and what is the best way to encourage diversity in this context? In this paper, we review the main published approaches to behavioral diversity and benchmark them in a common framework. We compare each approach on three different tasks and two different genotypes. The results show that fostering behavioral diversity substantially improves the evolutionary process in the investigated experiments, regardless of genotype or task. Among the benchmarked approaches, multi-objective methods were the most efficient and the generic, Hamming-based, behavioral distance was at least as efficient as task specific behavioral metrics.

scholarly journals Wankelmut: A Simple Benchmark for the Evolvability of Behavioral Complexity

2021 ◽  
Vol 11 (5) ◽  
pp. 1994
Author(s):  
Thomas Schmickl ◽  
Payam Zahadat ◽  
Heiko Hamann
Keyword(s):  
Neural Networks ◽  
Evolutionary Computation ◽  
A Priori ◽  
Evolutionary Process ◽  
Evolutionary Robotics ◽  
Sensor Data ◽  
Control Software ◽  
Main Claim ◽  
Behavioral Complexity ◽  
Model Free

In evolutionary robotics, an encoding of the control software that maps sensor data (input) to motor control values (output) is shaped by stochastic optimization methods to complete a predefined task. This approach is assumed to be beneficial compared to standard methods of controller design in those cases where no a priori model is available that could help to optimize performance. For robots that have to operate in unpredictable environments as well, an evolutionary robotics approach is favorable. We present here a simple-to-implement, but hard-to-pass benchmark to allow for quantifying the “evolvability” of such evolving robot control software towards increasing behavioral complexity. We demonstrate that such a model-free approach is not a free lunch, as already simple tasks can be unsolvable barriers for fully open-ended uninformed evolutionary computation techniques. We propose the “Wankelmut” task as an objective for an evolutionary approach that starts from scratch without pre-shaped controller software or any other informed approach that would force the behavior to be evolved in a desired way. Our main claim is that “Wankelmut” represents the simplest set of problems that makes plain-vanilla evolutionary computation fail. We demonstrate this by a series of simple standard evolutionary approaches using different fitness functions and standard artificial neural networks, as well as continuous-time recurrent neural networks. All our tested approaches failed. From our observations, we conclude that other evolutionary approaches will also fail if they do not per se favor or enforce the modularity of the evolved structures and if they do not freeze or protect already evolved functionalities from being destroyed again in the later evolutionary process. However, such a protection would require a priori knowledge of the solution of the task and contradict the “no a priori model” approach that is often claimed in evolutionary computation. Thus, we propose a hard-to-pass benchmark in order to make a strong statement for self-complexifying and generative approaches in evolutionary computation in general and in evolutionary robotics specifically. We anticipate that defining such a benchmark by seeking the simplest task that causes the evolutionary process to fail can be a valuable benchmark for promoting future development in the fields of artificial intelligence, evolutionary robotics, and artificial life.

Beyond Interactive Evolution: Expressing Intentions Through Fitness Functions

Leonardo ◽  
2016 ◽  
Vol 49 (3) ◽  
pp. 251-256 ◽  
Author(s):  
Penousal Machado ◽  
Tiago Martins ◽  
Hugo Amaro ◽  
Pedro H. Abreu
Keyword(s):  
Fitness Function ◽  
Interactive Art ◽  
Support Tool ◽  
Generative Art ◽  
Fitness Functions ◽  
Meta Level ◽  
Evolutionary Art ◽  
Creativity Support ◽  
The Creation ◽  
The Impact

Photogrowth is a creativity support tool for the creation of nonphotorealistic renderings of images. The authors discuss its evolution from a generative art application to an interactive evolutionary art tool and finally into a meta-level interactive art system in which users express their artistic intentions through the design of a fitness function. The authors explore the impact of these changes on the sense of authorship, highlighting the range of imagery that can be produced by the system.

Dynamic Correcting Dispersion Parameters of Lagrangian Puff Model in Atmospheric Tracer Experiments

2014 ◽  
Author(s):  
Yuanwei Ma ◽  
Dezhong Wang ◽  
Zhilong Ji ◽  
Nan Qian
Keyword(s):  
Numerical Simulation ◽  
Dispersion Model ◽  
Fitness Function ◽  
Atmospheric Dispersion ◽  
Correction Procedure ◽  
Model Errors ◽  
Dispersion Parameters ◽  
Fitness Functions ◽  
Atmospheric Dispersion Models ◽  
Atmospheric Tracer

In atmospheric dispersion models of nuclear accident, the empirical dispersion coefficients were obtained under certain experiment conditions, which is different from actual conditions. This deviation brought in the great model errors. A better estimation of the radioactive nuclide’s distribution could be done by correcting coefficients with real-time observed value. This reverse problem is nonlinear and sensitive to initial value. Genetic Algorithm (GA) is an appropriate method for this correction procedure. Fitness function is a particular type of objective function to achieving the set goals. To analysis the fitness functions’ influence on the correction procedure and the dispersion model’s forecast ability, four fitness functions were designed and tested by a numerical simulation. In the numerical simulation, GA, coupled with Lagrange dispersion model, try to estimate the coefficients with model errors taken into consideration. Result shows that the fitness functions, in which station is weighted by observed value and by distance far from release point, perform better when it exists significant model error. After performing the correcting procedure on the Kincaid experiment data, a significant boost was seen in the dispersion model’s forecast ability.

A Multiple Objective Fitness Function Design Scheme for GA-Based Cognitive Engines

2015 ◽  
Vol 764-765 ◽  
pp. 444-447
Author(s):  
Keun Hong Chae ◽  
Hua Ping Liu ◽  
Seok Ho Yoon
Keyword(s):  
Genetic Algorithm ◽  
Parameter Optimization ◽  
Fitness Function ◽  
Multiple Objective ◽  
Cognitive Engine ◽  
Fitness Functions ◽  
Design Scheme ◽  
Transmission Parameter ◽  
Function Design ◽  
Single Objective

In this paper, we propose a multiple objective fitness function for cognitive engines by using the genetic algorithm (GA). Specifically, we propose four single objective fitness functions, and finally, we propose a multiple objective fitness function based on the single objective fitness functions for transmission parameter optimization. Numerical results demonstrate that we can obtain transmission parameter sets optimized for given transmission scenarios with the GA-based cognitive engine incorporating the proposed objective fitness function.

scholarly journals Pixel-Based Approach for Generating Original and Imitating Evolutionary Art

Electronics ◽  
2020 ◽  
Vol 9 (8) ◽  
pp. 1311
Author(s):  
Yuchen Wang ◽  
Rong Xie
Keyword(s):  
Fitness Function ◽  
Visual Complexity ◽  
Generation Process ◽  
Single Object ◽  
Von Neumann ◽  
Modified Particle Swarm Optimization ◽  
Fitness Functions ◽  
Evolutionary Art ◽  
Artistic Styles ◽  
The Aesthetic

We proposed a pixel-based evolution method to automatically generate evolutionary art. Our method can generate diverse artworks, including original artworks and imitating artworks, with different artistic styles and high visual complexity. The generation process is fully automated. In order to adapt to the pixel-based method, a von Neumann neighbor topology-modified particle swarm optimization (PSO) is employed to the proposed method. The fitness functions of PSO are well prepared. Firstly, we come up with a set of aesthetic fitness functions. Next, the imitating fitness function is designed. Finally, the aesthetic fitness functions and the imitating fitness function are weighted into one single object function, which is used in the modified PSO. Both the original outputs and imitating outputs are shown. A questionnaire is designed to investigate the subjective aesthetic feeling of proposed evolutionary art, and the statistics are shown.

Genetic multistart algorithm for the design of fault-tolerant systems

2008 ◽  
Vol 222 (1) ◽  
pp. 17-29
Author(s):  
K Echtle ◽  
I Eusgeld ◽  
D Hirsch
Keyword(s):  
Genetic Algorithm ◽  
Fault Tolerance ◽  
Fault Tolerant ◽  
Design Method ◽  
Fitness Function ◽  
Evolutionary Process ◽  
Mechatronic Systems ◽  
Fast Evaluation ◽  
Pareto Solution ◽  
Fault Tolerant Systems

This paper presents a new approach to the multiobjective design of fault-tolerant systems. The design objectives are fault tolerance and cost. Reducing the cost is of particular importance for fault-tolerant systems because the overhead caused by redundant components is considerable. The new design method consists of a special genetic algorithm that is tailored to the particular issues of fault-tolerant systems. The interface of the present tool ePADuGA (elitist and Pareto-based Approach to Design fault-tolerant systems using a Genetic Algorithm) allows for adaptation to various fields of application. The degree of fault tolerance is measured by the number of tolerated faults rather than traditional reliability metrics, because reliability numbers are mostly unknown during early design phases. The special features of the genetic algorithm comprise a graph-oriented representation of systems (which are the individuals during the evolutionary process), a simple yet expressive fault model, a very efficient procedure for fault-tolerance evaluation, and a Pareto-oriented fitness function. In a genetic algorithm generating thousands of individuals, a very fast evaluation of each individual is mandatory. For this purpose, state-space-oriented evaluation methods have been cut down to an extremely simple function which is still sufficient to assess the fault tolerance of individuals. An innovative aspect is also a multistart technique to find a Pareto solution set, which is independent of any parameters. In this paper, experimental results are presented showing the feasibility of the approach as well as the usefulness of the final fault-tolerant architectures, particularly in the field of mechatronic systems.

The eleventh R. A. Fisher Memorial Lecture - Kin selection and altruism

1983 ◽  
Vol 219 (1216) ◽  
pp. 327-353 ◽  
Keyword(s):  
Kin Selection ◽  
Group Composition ◽  
Group Structure ◽  
Fitness Function ◽  
Memorial Lecture ◽  
Marginal Benefit ◽  
Fitness Functions ◽  
Short And Long Term ◽  
Cost Parameters

Fisher (1930), Haldane (1932), and others discussed short and long term fitness relationships of the biological basis of social behaviour. Hamilton (1964 a , b ) proposed the inequality b / c > 1/ r ( b and c are marginal benefit and cost parameters, respectively, r is an appropriate kinship coefficient) as an essential concomitant of the evolution of altruism. Virtually all current kin selection models take the marginal benefit and cost parameters as primitive concepts and combine them in various ways to determine population fitness values. We offer an intrinsic ‘fitness function’ approach to modelling the theory of kin selection. The components of the model involve: ( a ) the delineation of the basic group structure specifying individual relationships; ( b ) the specification of local fitness functions that depend on group composition; ( c ) the determination of average fitness functions for the different phenotypes with respect to the population at large. We then derive a pair of benefit and cost functions, which are functions of the group composition and the numbers of altruist and selfish phenotypes. In this new framework the quantitative validity of the Hamilton criterion for the evolution of altruism are assessed and reinterpreted.

Sign in / Sign up

Export Citation Format

Share Document