Emergence of Collective Behavior in Evolving Populations of Flying Agents

We demonstrate the emergence of collective behavior in two evolutionary computation systems, one an evolutionary extension of a classic (highly constrained) flocking algorithm and the other a relatively un-constrained system in which the behavior of agents is governed by evolved computer programs. The first system demonstrates the evolution of a form of multicellular organization, while the second demonstrates the evolution of a form of altruistic food sharing. In this article we describe both systems in detail, document the emergence of collective behavior, and argue that these systems present new opportunities for the study of group dynamics in an evolutionary context. We also provide a brief overview of the breve simulation environment in which the systems were produced, and of breve’s facilities for the rapid, exploratory development of visualization strategies for artificial life.     

Morphogenic inspired self-organization control of unmanned aerial swarm robots

A central issue in distributed formation of swarm is enabling agents with only a local view of their environment to take actions that advance global system objectives (emergence of collective behavior). This paper describes a bio-inspired control algorithm using pheromone for coordinating a swarm of identical flying agents to spatially self-organize into arbitrary shapes using local communication maintaining a certain level of density. Development of morphological features such as the fine formation of particles, hollow particles, cracks in particles, and the evolution of the structure are used as illustrative examples. Distributed formation of swarming with no coordinated agreement or positioning information is an interesting research area because the global behaviors must emerge from many diverse local interactions. Different from most existing distributed control, the proposed approach considers the topological structure of the organization, supports dynamic reconfiguration and self-organization. This paper presents the experimental results on simulating in the forming of arbitrary shape, and also describes a analytic aspect of our model comparing with behavioral performances.

     

Towards the evolutionary emergence of increasingly complex advantageous behaviours

The generation of complex entities with advantageous behaviours beyond our manual design capability requires long-term incremental evolution with continuing emergence. In this paper, we argue that artificial selection models, such as traditional genetic algorithms, are fundamentally inadequate for this goal. Existing natural selection systems are evaluated, revealing both significant achievements and pitfalls. Thus, some requirements for the perpetuation of evolutionary emergence are established. An (artificial) environment containing simple virtual autonomous organisms with neural controllers has been created to satisfy these requirements and to aid in the development of an accompanying theory of evolutionary emergence. Resulting behaviours are reported alongside their neural correlates. In a particular example, the collective behaviour of one species provides a selective force which is overcome by another species, demonstrating the incremental evolutionary emergence of advantageous behaviours via naturally arising coevolution. While the results fall short of the ultimate goal, experience with the system has provided some useful lessons for the perpetuation of emergence towards increasingly complex advantageous behaviours.     

Evolved swarming without positioning information: an application in aerial communication relay

In most swarm systems, agents are either aware of the position of their direct neighbors or they possess a substrate on which they can deposit information (stigmergy). However, such resources are not always obtainable in real-world applications because of hardware and environmental constraints. In this paper we study in 2D simulation the design of a swarm system which does not make use of positioning information or stigmergy. This endeavor is motivated by an application whereby a large number of Swarming Micro Air Vehicles (SMAVs), of fixed-wing configuration, must organize autonomously to establish a wireless communication network (SMAVNET) between users located on ground. Rather than relative or absolute positioning, agents must rely only on their own heading measurements and local communication with neighbors. Designing local interactions responsible for the emergence of the SMAVNET deployment and maintenance is a challenging task. For this reason, artificial evolution is used to automatically develop neuronal controllers for the swarm of homogenous agents. This approach has the advantage of yielding original and efficient swarming strategies. A detailed behavioral analysis is then performed on the fittest swarm to gain insight as to the behavior of the individual agents.

Emergence of memory-driven command neurons in evolved artificial agents

Using evolutionary simulations, we develop autonomous agents controlled by artificial neural networks (ANNs). In simple lifelike tasks of foraging and navigation, high performance levels are attained by agents equipped with fully recurrent ANN controllers. In a set of experiments sharing the same behavioral task but differing in the sensory input available to the agents, we find a common structure of a command neuron switching the dynamics of the network between radically different behavioral modes. When sensory position information is available, the command neuron reflects a map of the environment, acting as a location-dependent cell sensitive to the location and orientation of the agent. When such information is unavailable, the command neuron's activity is based on a spontaneously evolving short-term memory mechanism, which underlies its apparent place-sensitive activity. A two-parameter stochastic model for this memory mechanism is proposed. We show that the parameter values emerging from the evolutionary simulations are near optimal; evolution takes advantage of seemingly harmful features of the environment to maximize the agent's foraging efficiency. The accessibility of evolved ANNs for a detailed inspection, together with the resemblance of some of the results to known findings from neurobiology, places evolved ANNs as an excellent candidate model for the study of structure and function relationship in complex nervous systems.     

Emerging Cooperation With Minimal Effort: Rewarding Over Mimicking

This paper compares supervised and unsupervised learning mechanisms for the emergence of cooperative multiagent spatial coordination using a top-down approach. By observing the global performance of a group of homogeneous agents-supported by a nonglobal knowledge of their environment-we attempt to extract information about the minimum size of the agent neurocontroller and the type of learning mechanism that collectively generate high-performing and robust behaviors with minimal computational effort. Consequently, a methodology for obtaining controllers of minimal size is introduced and a comparative study between supervised and unsupervised learning mechanisms for the generation of successful collective behaviors is presented. We have developed a prototype simulated world for our studies. This case study is primarily a computer games inspired world but its main features are also biologically plausible. The two specific tasks that the agents are tested in are the competing strategies of obstacle-avoidance and target-achievement. We demonstrate that cooperative behavior among agents, which is supported only by limited communication, appears to be necessary for the problem's efficient solution and that learning by rewarding the behavior of agent groups constitutes a more efficient and computationally preferred generic approach than supervised learning approaches in such complex multiagent worlds     

Representations of space and time in the maximization of information flow in the perception-action loop

Sensor evolution in nature aims at improving the acquisition of information from the environment and is intimately related with selection pressure toward adaptivity and robustness. Our work in the area indicates that information theory can be applied to the perception-action loop. This letter studies the perception-action loop of agents, which is modeled as a causal Bayesian network. Finite state automata are evolved as agent controllers in a simple virtual world to maximize information flow through the perception-action loop. The information flow maximization organizes the agent's behavior as well as its information processing. To gain more insight into the results, the evolved implicit representations of space and time are analyzed in an information-theoretic manner, which paves the way toward a principled and general understanding of the mechanisms guiding the evolution of sensors in nature and provides insights into the design of mechanisms for artificial sensor evolution.     

A study of evolution strategy based cooperative behavior in collective agents

The following paper introduces an evolution strategy on the basis of cooperative behaviors in each group of agents. The evolution strategy helps each agent to be self-defendable and self-maintainable. To determine an optimal group behavior strategy under dynamically varying circumstances, agents in same group cooperate with each other. This proposed method use reinforcement learning, enhanced neural network, and artificial life. In the present paper, we apply two different reward models: reward model 1 and reward model 2. Each reward model is designed as considering the reinforcement or constraint of behaviors. In competition environments of agents, the behavior considered to be advantageous is reinforced as adding reward values. On the contrary, the behavior considered to be disadvantageous is constrained as subtracting the values. And we propose an enhanced neural network to add learning behavior of an artificial organism-level to artificial life simulation. In future, the system models and results described in this paper will be applied to the framework of healthcare systems that consists of biosensors, healthcare devices, and healthcare system.     

群集系统的软控制

考虑欧氏空间中群集动态系统的"软控制"问题,即在不改变群集系统中个体间现有规则的条件下,通过增加一个或多个可控智能体干扰群集个体的行为,使群集中心转移到期望的位置.给出了可控智能体的控制规则,并利用粒子群(PSO)算法讨论了群集系统软控制的最优收敛问题.实例研究表明了"软控制"在群集系统中的可行性.     

To the Analytical Investigation of the Convergence Conditions of the Processes of Reflexive Collective Behavior in Oligopoly Models

Automation and Remote Control - A study of the dynamics of the collective behavior of interconnected rational agents under conditions of incomplete information is presented. Statements are proved...     

An optimization algorithm for multimodal functions inspired by collective animal behavior

Interest in multimodal function optimization is expanding rapidly as real-world optimization problems often demand locating multiple optima within a search space. This article presents a new multimodal optimization algorithm named as the collective animal behavior. Animal groups, such as schools of fish, flocks of birds, swarms of locusts, and herds of wildebeest, exhibit a variety of behaviors including swarming about a food source, milling around a central location, or migrating over large distances in aligned groups. These collective behaviors are often advantageous to groups, allowing them to increase their harvesting efficiency to follow better migration routes, to improve their aerodynamic, and to avoid predation. In the proposed algorithm, searcher agents are a group of animals which interact with each other based on the biologic laws of collective motion. Experimental results demonstrate that the proposed algorithm is capable of finding global and local optima of benchmark multimodal optimization problems with a higher efficiency in comparison with other methods reported in the literature.     

The proportion of leaders needed for the expected consensus

In order to have a self-organized multi-agent system exhibit some expected collective behavior, it is necessary to add some special agents with information (called leaders) to intervene the system. Then a fundamental question is: how many such leaders are needed? Naturally the answer depends on the model to be studied. In this paper a typical model proposed by Vicsek et al. is used for answering the question. By estimating the characteristics concerning the initial states of all agents and analyzing the system dynamics, we provide lower bounds on the ratio of leaders needed to guarantee the expected consensus.     

Self-organized coordinated motion in groups of physically connected robots.

An important goal of collective robotics is the design of control systems that allow groups of robots to accomplish common tasks by coordinating without a centralized control. In this paper, we study how a group of physically assembled robots can display coherent behavior on the basis of a simple neural controller that has access only to local sensory information. This controller is synthesized through artificial evolution in a simulated environment in order to let the robots display coordinated-motion behaviors. The evolved controller proves to be robust enough to allow a smooth transfer from simulated to real robots. Additionally, it generalizes to new experimental conditions, such as different sizes/shapes of the group and/or different connection mechanisms. In all these conditions the performance of the neural controller in real robots is comparable to the one obtained in simulation.     

Neural network-based adaptive consensus tracking control for multi-agent systems under actuator faults

In this paper, a distributed output feedback consensus tracking control scheme is proposed for second-order multi-agent systems in the presence of uncertain nonlinear dynamics, external disturbances, input constraints, and partial loss of control effectiveness. The proposed controllers incorporate reduced-order filters to account for the unmeasured states, and the neural networks technique is implemented to approximate the uncertain nonlinear dynamics in the synthesis of control algorithms. In order to compensate the partial loss of actuator effectiveness faults, fault-tolerant parts are included in controllers. Using the Lyapunov approach and graph theory, it is proved that the controllers guarantee a group of agents that simultaneously track a common time-varying state of leader, even when the state of leader is available only to a subset of the members of a group. Simulation results are provided to demonstrate the effectiveness of the proposed consensus tracking method.     

An Evolutionary Model of Price Competition Among Spatially Distributed Firms

Various studies have shown the emergence of cooperative behavior in evolutionary models with repeated local interaction among spatially distributed agents. We investigate to what extent these findings generalize to evolutionary models of price competition among spatially distributed firms. We consider both one- and two-dimensional models, and we vary the amount of information firms have about competitors in their neighborhood. Our computer simulations show that the emergence of cooperative behavior depends strongly on the amount of information available to firms. Firms tend to behave most cooperatively if they have only a very limited amount of information about their competitors. We provide an intuitive explanation for this phenomenon. Our simulations further indicate that three other factors in our models, namely the accuracy of firms’ information, the probability of experimentation, and the spatial distribution of consumers, have little effect on the emergence of cooperative behavior.     

On the evolution of homogeneous two-robot teams: clonal versus aclonal approaches

This study compares two different evolutionary approaches (clonal and aclonal) to the design of homogeneous two-robot teams (i.e. teams of morphologically identical agents with identical controllers) in a task that requires the agents to specialise to different roles. The two approaches differ mainly in the way teams are formed during evolution. In the clonal approach, a team is formed from a single genotype within one population of genotypes. In the aclonal approach, a team is formed from multiple genotypes within one population of genotypes. In both cases, the goal is the synthesis of individual generalist controllers capable of integrating role execution and role allocation mechanisms for a team of homogeneous robots. Our results diverge from those illustrated in a similar comparative study, which supports the superiority of the aclonal versus the clonal approach. We question this result and its theoretical underpinning, and we bring new empirical evidence showing that the clonal outperforms the aclonal approach in generating homogeneous teams required to dynamically specialise for the benefit of the team. The results of our study suggest that task-specific elements influence the evolutionary dynamics more than the genetic relatedness of the team members. We conclude that the appropriateness of the clonal approach for role allocation scenarios is mainly determined by the specificity of the collective task, including the evaluation function, rather than by the way in which the solutions are evaluated during evolution.     

Evolving Receptive-Field Controllers for Mobile Robots

The use of evolutionary methods to generate controllers for real-world autonomous agents has attracted recent attention. Most of the pertinent research has employed genetic algorithms or variations thereof. Recent research has applied an alternative evolutionary method, evolution strategies, to the generation of simple Braitenberg vehicles. This application accelerates the development of such controllers by more than an order of magnitude (a few hours compared to more than two days). Motivated by this useful speedup, this paper investigates the evolution of more complex architectures, receptive-field controllers, that can employ nonlinear interactions and, therefore, can yield more complex behavior. It is interesting to note that the evolution strategy yields the same efficacy in terms of function evaluations, even though the second class of controllers requires up to 10 times more parameters than the simple Braitenberg architecture. In addition to the speedup, there is an important theoretical reason for preferring an evolution strategy over a genetic algorithm for this problem, namely the presence of epistasis.     

Evolution of cartesian genetic programs for development of learning neural architecture

Although artificial neural networks have taken their inspiration from natural neurological systems, they have largely ignored the genetic basis of neural functions. Indeed, evolutionary approaches have mainly assumed that neural learning is associated with the adjustment of synaptic weights. The goal of this paper is to use evolutionary approaches to find suitable computational functions that are analogous to natural sub-components of biological neurons and demonstrate that intelligent behavior can be produced as a result of this additional biological plausibility. Our model allows neurons, dendrites, and axon branches to grow or die so that synaptic morphology can change and affect information processing while solving a computational problem. The compartmental model of a neuron consists of a collection of seven chromosomes encoding distinct computational functions inside the neuron. Since the equivalent computational functions of neural components are very complex and in some cases unknown, we have used a form of genetic programming known as Cartesian genetic programming (CGP) to obtain these functions. We start with a small random network of soma, dendrites, and neurites that develops during problem solving by repeatedly executing the seven chromosomal programs that have been found by evolution. We have evaluated the learning potential of this system in the context of a well-known single agent learning problem, known as Wumpus World. We also examined the harder problem of learning in a competitive environment for two antagonistic agents, in which both agents are controlled by independent CGP computational networks (CGPCN). Our results show that the agents exhibit interesting learning capabilities.     

Modeling human behavior during emergency evacuation using intelligent agents: A multi-agent simulation approach

It is costly and takes a lot of time for disaster employees to execute several evacuation drills for a building. One cannot glean information to advance the plan and blueprint of forthcoming buildings without executing many drills. We have developed a multi-agent system simulation application to aid in running several evacuation drills and theoretical situations. This paper combines the genetic algorithm (GA) with neural networks (NNs) and fuzzy logic (FL) to explore how intelligent agents can learn and adapt their behavior during an evacuation. The adaptive behavior focuses on the specific agents changing their behavior in the environment. The shared behavior of the agent places an emphasis on the crowd-modeling and emergency behavior in the multi-agent system. This paper provides a fuzzy individual model being developed for realistic modeling of human emotional behavior under normal and emergency conditions. It explores the impact of perception and emotions on the human behavior. We have established a novel intelligent agent with characteristics such as independence, collective ability, cooperativeness, and learning, which describes its final behavior. The contributions of this paper lie in our approach of utilizing a GA, NNs, and FL to model learning and adaptive behavior of agents in a multi-agent system. The planned application will help in executing numerous evacuation drills for what-if scenarios for social and cultural issues such as evacuation by integrating agent characteristics. This paper also compares our proposed multi-agent system with existing commercial evacuation tools as well as real-time evacuation drills for accuracy, building traffic characteristics, and the cumulative number of people exiting during evacuation. Our results show that the inclusion of GA, NNs, and fuzzy attributes made the evacuation time of the agents closer to the real-time evacuation drills.     

Predictor-based neural dynamic surface control for distributed formation tracking of multiple marine surface vehicles with improved transient performance基于预估器神经网络动态面的海洋航行器分布式队形跟踪

In this paper, we investigate the distributed formation tracking problem of multiple marine surface vehicles with model uncertainty and time-varying ocean disturbances induced by wind, waves, and ocean currents. The objective is to achieve a collective tracking with a time-varying trajectory, which can only be accessed by a fraction of follower vehicles. A novel predictor-based neural dynamic surface control design approach is proposed to develop the distributed adaptive formation controllers. We use prediction errors, rather than tracking errors, to construct the neural adaptive laws, which enable the fast identification of the vehicle dynamics without incurring high-frequency oscillations in control signals. We establish the stability properties of the closed-loop network via Lyapunov analysis, and quantify the transient performance by deriving the truncated L₂ norms of the derivatives of neural weights, which we demonstrate to be smaller than the classical neural dynamic surface control design approach. We also extend the above result to the distributed formation tracking using the relative position information of vehicles, and the advantage is that the velocity information of neighbors and leader are required. Finally, we give the comparative studies to illustrate the performance improvement of the proposed method.