Genetic Team Composition and Level of Selection in the Evolution of Cooperation

In cooperative multiagent systems, agents interact to solve tasks. Global dynamics of multiagent teams result from local agent interactions, and are complex and difficult to predict. Evolutionary computation has proven a promising approach to the design of such teams. The majority of current studies use teams composed of agents with identical control rules (“genetically homogeneous teams”) and select behavior at the team level (“team-level selection”). Here we extend current approaches to include four combinations of genetic team composition and level of selection. We compare the performance of genetically homogeneous teams evolved with individual-level selection, genetically homogeneous teams evolved with team-level selection, genetically heterogeneous teams evolved with individual-level selection, and genetically heterogeneous teams evolved with team-level selection. We use a simulated foraging task to show that the optimal combination depends on the amount of cooperation required by the task. Accordingly, we distinguish between three types of cooperative tasks and suggest guidelines for the optimal choice of genetic team composition and level of selection.      

Designing emergent cooperation: a pursuit–evasion game case study

This research concerns a comparison of two neuroevolution approaches for the design of cooperative behavior in a group of simulated mobile robots. The first approach, termed single pool (SP), was characterized by robot neural controllers that were derived from a single genotype. The second approach, termed multiple pools (MP), was characterized by robot neural controllers that were derived from many different genotypes. The application domain implemented a pursuit–evasion game in which teams of robots of various sizes, termed predators, collectively worked to capture (immobilize) other robots, termed prey. The MP and SP approaches were tested, with and without lifetime learning, for the design of cooperative prey capture behavior within teams of predators. Results indicated that the MP approach was superior to the SP approach in terms of measures defined for prey-capture performance. Specifically, the MP approach facilitated behavioral specializations in the predator team facilitating emergent cooperative prey capture strategies that proved effective for the various team sizes tested.     

Encouraging behavioral diversity in evolutionary robotics: an empirical study

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.     

RoboCup Rescue as multiagent task allocation among teams: experiments with task interdependencies

This paper addresses distributed task allocation among teams of agents in a RoboCup Rescue scenario. We are primarily concerned with testing different mechanisms that formalize issues underlying implicit coordination among teams of agents. These mechanisms are developed, implemented, and evaluated using two algorithms: Swarm-GAP and LA-DCOP. The latter bases task allocation on a comparison between an agent’s capability to perform a task and the capability demanded by this task. Swarm-GAP is a probabilistic approach in which an agent selects a task using a model inspired by task allocation among social insects. Both algorithms were also compared to another one that allocates tasks in a greedy way. Departing from previous works that tackle task allocation in the rescue scenario only among fire brigades, here we consider the various actors in the RoboCup Rescue, a step forward in the direction of realizing the concept of extreme teams. Tasks are allocated to teams of agents without explicit negotiation and using only local information. Our results show that the performance of Swarm-GAP and LA-DCOP are similar and that they outperform a greedy strategy. Also, it is possible to see that using more sophisticated mechanisms for task selection does pay off in terms of score.     

Exploiting Spatial Locality and Heterogeneity of Agents for Search and Rescue Teamwork*

We propose coordination mechanisms for multiple heterogeneous physical agents that operate in city‐scale disaster scenarios, where they need to find and rescue people and extinguish fires. Large‐scale disasters are characterized by limited and unreliable communications; dangerous events that may disable agents; uncertainty about the location, duration, and type of tasks; and stringent temporal constraints on task completion times. In our approach, agents form teams with other agents that are in the same geographical area. Our algorithms either yield stable teams formed up front and never change, fluid teams where agents can change teams as need arises, or teams that restrict the types of agents that can belong to the same team. We compare our teaming algorithms against a baseline algorithm in which agents operate independently of others and two state‐of‐the‐art coordination mechanisms. Our algorithms are tested in city‐scale disaster simulations using the RoboCup Rescue simulator. Our experiments with different city maps show that, in general, forming teams leads to increased task completion and, specifically, that our teaming method that restricts the types of agents in a team outperforms the other methods.     

Towards efficient multiagent task allocation in the RoboCup Rescue: a biologically-inspired approach

This paper addresses team formation in the RoboCup Rescue centered on task allocation. We follow a previous approach that is based on so-called extreme teams, which have four key characteristics: agents act in domains that are dynamic; agents may perform multiple tasks; agents have overlapping functionality regarding the execution of each task but differing levels of capability; and some tasks may depict constraints such as simultaneous execution. So far these four characteristics have not been fully tested in domains such as the RoboCup Rescue. We use a swarm intelligence based approach, address all characteristics, and compare it to other two GAP-based algorithms. Experiments where computational effort, communication load, and the score obtained in the RoboCup Rescue aremeasured, show that our approach outperforms the others.     

Challenges in cooperative coevolution of physically heterogeneous robot teams

Heterogeneous multirobot systems have shown significant potential in many applications. Cooperative coevolutionary algorithms (CCEAs) represent a promising approach to synthesise controllers for such systems, as they can evolve multiple co-adapted components. Although CCEAs allow for an arbitrary level of team heterogeneity, in previous works heterogeneity is typically only addressed at the behavioural level. In this paper, we study the use of CCEAs to evolve control for a heterogeneous multirobot system where the robots have disparate morphologies and capabilities. Our experiments rely on a simulated task where a simple ground robot must cooperate with a complex aerial robot to find and collect items. We first show that CCEAs can evolve successful controllers for physically heterogeneous teams, but find that differences in the complexity of the skills the robots need to learn can impair CCEAs’ effectiveness. We then study how different populations can use different evolutionary algorithms and parameters tuned to the agents’ complexity. Finally, we demonstrate how CCEAs’ effectiveness can be improved using incremental evolution or novelty-driven coevolution. Our study shows that, despite its limitations, coevolution is a viable approach for synthesising control for morphologically heterogeneous systems.

     

A set partitioning based goal programming model for the team formation problem

In this paper, we offer a multi-objective set-partitioning formulation for team formation problems using goal programming. Instead of selecting team members to teams, we select suitable teams from a set of teams. This set is generated using a heuristic algorithm that uses the social network of potential team members. We then utilize the proposed multi-objective formulation to select the desired number of teams from this set that meets the skill requirements. Therefore, we ensure that selected teams include individuals with the required skills and effective communication with each other. Two real datasets are used to test the model. The results obtained with the proposed solution are compared with two well-known approaches: weighted and lexicographic goal programming. Results reveal that weighted and lexicographic goal programming approaches generate almost identical solutions for the datasets tested. Our approach, on the other hand, mostly picks teams with lower communication costs. Even in some cases, better solutions are obtained with the proposed approach. Findings show that the developed solution approach is a promising approach to handle team formation problems.     

计算机专业大学生创新团队的建设与实践

建立大学生创新团队是高校培养学生创新意识的有效途径,其作用日益凸显。通过计算机专业创新团队的建设,介绍了大学生创新团队建设的途径及所取得的成绩,指出大学生创新团队建设是高校教学的一项重要任务,必须建立健全创新团队建设的长效机制,才能有力地促进高校对创新人才的培养。     

Group development in virtual teams: An experimental reexamination

Virtual teams are thought to be experienced differently and to have poor outcomes because there is little or no face-to-face interaction and a tendency for virtual team members to use different communication techniques for forming relationships. However, the expanding use of virtual teams in organizations suggests that virtual teams in real world contexts are able to overcome these barriers and be experienced in much the same way as face-to-face teams. This paper reports the result of an experiment in which virtual teams participated in an exercise where they completed an information-sharing task ten times as a team. The results suggest that, contrary to one-shot, ad hoc virtual teams, longer-lived virtual teams follow a sequential group development process. Virtual team development appears to differ from face-to-face teams because the use of computer-mediated communication heightens pressure to conform when a virtual team is first formed, meaning trust is most strongly linked with feeling that the team was accomplishing the task appropriately. As the virtual teams developed, trust in peers was more strongly linked with goal commitment. Once the teams were working together effectively, accomplishing the task appropriately was the strongest link with trust in peers. I suggest that virtual team managers should cultivate virtual workspaces that are similar to those proven to work in face-to-face contexts: (1) teams should have clear, specific goals, (2) members should be encouraged or even required to communicate with each other, and (3) team members should feel that they might work with the other team members again.     

From roles to teamwork: A framework and architecture

Teamwork requires organization, strategies, and coordination. The design of a multiagent system should support these conceptual properties for constructing effective teams. The advantage of a teamwork approach is the reduction in complexity of the task through distribution of responsibilities, resulting in better utilization of resources, robust behaviors, and a greater variety of behaviors against competitors. In this article a framework for building teams of responsible agents using roles, responsibilities, and strategies is described. Its application to the domain of soccer is used to design a high-performance team of soccer agents. The architecture for these agents utilizes a reactive planning system with support for teamwork. The team of soccer agents will be tested in a series of competitions against other teams in the real-time soccer simulator proposed for Robocup-97, which provides an uncertain, resource bounded world.     

The effect of governance on global software development: An empirical research in transactive memory systems

ContextThe way global software development (GSD) activities are managed impacts knowledge transactions between team members. The first is captured in governance decisions, and the latter in a transactive memory system (TMS), a shared cognitive system for encoding, storing and retrieving knowledge between members of a group.ObjectiveWe seek to identify how different governance decisions (such as business strategy, team configuration, task allocation) affect the structure of transactive memory systems as well as the processes developed within those systems.MethodWe use both a quantitative and a qualitative approach. We collect quantitative data through an online survey to identify transactive memory systems. We analyze transactive memory structures using social network analysis techniques and we build a latent variable model to measure transactive memory processes. We further support and triangulate our results by means of interviews, which also help us examine the GSD governance modes of the participating projects. We analyze governance modes, as set of decisions based on three aspects; business strategy, team structure and composition, and task allocation.ResultsOur results suggest that different governance decisions have a different impact on transactive memory systems. Offshore insourcing as a business strategy, for instance, creates tightly-connected clusters, which in turn leads to better developed transactive memory processes. We also find that within the composition and structure of GSD teams, there are boundary spanners (formal or informal) who have a better overview of the network’s activities and become central members within their network. An interesting mapping between task allocation and the composition of the network core suggests that the way tasks are allocated among distributed teams is an indicator of where expertise resides.ConclusionWe present an analytical method to examine GSD governance decisions and their effect on transactive memory systems. Our method can be used from both practitioners and researchers as a “cause and effect” tool for improving collaboration of global software teams.     

A Constraint-Based Robotic Soccer Team

It is a challenging task for a team of multiple fast-moving robots to cooperate with each other and to compete with another team in a dynamic, real-time environment. For a robot team to play soccer successfully, various technologies have to be incorporated including robotic architecture, multi-agent collaboration and real-time reasoning. A robot is an integrated system, with a controller embedded in its plant. A robotic system is the coupling of a robot to its environment. Robotic systems are, in general, hybrid dynamic systems, consisting of continuous, discrete and event-driven components. Constraint Nets (CN) provide a semantic model for modeling hybrid dynamic systems. Controllers are embedded constraint solvers that solve constraints in real-time. A controller for our robot soccer team, UBC Dynamo98, has been modeled in CN, and implemented in Java, using the Java Beans architecture. A coach program using an evolutionary algorithm has also been designed and implemented to adjust the weights of the constraints and other parameters in the controller. The results demonstrate that the formal CN approach is a practical tool for designing and implementing controllers for robots in multi-agent real-time environments. They also demonstrate the effectiveness of applying the evolutionary algorithm to the CN-modeled controllers.     

Dynamic task allocation for multi-robot search and retrieval tasks

Many application domains require search and retrieval, which is also known in the robotic domain as foraging. For example, in a search and rescue domain, a disaster area needs to be explored and transportation of survivors to a safe area needs to be arranged. Performing such a search and retrieval task by more than one robot increases performance if they are able to distribute their workload efficiently and evenly. In this work, we study the Multi-Robot Task Allocation (MRTA) problem in the search and retrieval domain, where a team of robots is required to cooperatively search for targets of interest in an environment and also retrieve them back to a home base. In comparison with typical foraging tasks, we look at a more general search and retrieval task in which the targets are distinguished with various types, and task allocation also requires taking into account temporal constraints on the team goal. As usual, robots have no prior knowledge about the location of targets in the environment but in addition they need to deliver targets to the home base in a specific order according to their types, which significantly increases the complexity of a foraging problem. We first use a graph-based model to analyse the search and retrieval problem and the dynamics of exploration and retrieval within a cooperative team. We then proceed to present an extended auction-based approach, as well as a prediction approach. The essential difference between these two approaches is that the task allocation and execution procedures in the auction approach are running in parallel, whereas a robot in the prediction approach only needs to choose a task to perform when it has no thing to do. The auction approach uses a winner determination mechanism to allocate tasks to each robot, whereas the robots in the prediction approach implicitly coordinate their activities by team reasoning that leads to consensuses about task allocation. We use the Blocks World for Teams (BW4T) simulator to evaluate the two approaches in our experimental study.     

分布式约束满足问题及其在MAS任务分配中的应用*

针对多智能体系统(MAS)任务分配问题中多个任务与MAS两者的分布式特征,将任务分配问题形式化为分布式约束满足问题(DCSP)进行求解,分别建立了以任务为中心和以agent为中心两种MAS任务分配模型,基于改进的DCSP分布式并行求解算法,提出了基于DCSP的MAS任务分配问题求解框架。该方法适合求解agent间通信有随机延迟以及agent间存在多约束的问题,应用实例的求解表明了其实用性与有效性。     

Sharing in teams of heterogeneous,collaborative learning agents

This paper is focused on the effects of sharing knowledge and collaboration of multiple heterogeneous, intelligent agents (hardware or software) which work together to learn a task. As each agent employs a different machine learning technique, the system consists of multiple knowledge sources and their respective heterogeneous knowledge representations. Collaboration between agents involves sharing knowledge to both speed up team learning, as well as refine the team's overall performance and group behavior. Experiments have been performed that vary the team composition in terms of machine learning algorithms, learning strategies employed by the agents, and sharing frequency for a predator‐prey cooperative pursuit task. For lifelong learning, heterogeneous learning teams were more successful than homogeneous learning counterparts. Interestingly, sharing increased the learning rate, but sharing with higher frequency showed diminishing results. Lastly, knowledge conflicts are reduced over time the more sharing takes place. These results support further investigation of the merits of heterogeneous learning. © 2008 Wiley Periodicals, Inc.     

Visualising Errors in Animal Pedigree Genotype Data

Genetic analysis of a breeding animal population involves determining the inheritance pattern of genotypes for multiple genetic markers across the individuals in the population pedigree structure. However, experimental pedigree genotype data invariably contains errors in both the pedigree structure and in the associated individual genotypes, introducing inconsistencies into the dataset, rendering them useless for further analysis. The resolution of these errors requires consideration of genotype inheritance patterns in the context of the pedigree structure. Existing pedigree visualisations are typically more suited to human pedigrees and are less suitable for large complex animal pedigrees which may exhibit cross generational inbreeding. Similarly, table‐based viewers of genotype marker data can highlight where errors become apparent but lack the functionality and interactive visual feedback to allow users to locate the origin of errors within the pedigree. In this paper, we detail a design study steered by biologists who work with pedigree data, and describe successive iterations through approaches and prototypes for viewing genotyping errors in the context of a displayed pedigree. We describe how each approach performs with real pedigree genotype data and why eventually we deemed them unsuitable. Finally, a novel prototype visualisation for pedigrees, which we term the ‘sandwich view’, is detailed and we demonstrate how the approach effectively communicates errors in the pedigree context, supporting the biologist in the error identification task.     

基于能力匹配的异地敏捷开发任务分配方法

为了能够快速准确地获得异地敏捷软件开发团队任务分配的全局最优解,提出了一种基于能力匹配的异地敏捷开发任务分配方法.该方法强调子任务能力需求和团队能力的匹配关系,构建了能力匹配的效用函数,对效用矩阵进行求解,全局效用值最大时获得最优分配方案.算例仿真结果表明,所提出的方法可以有效得到能力匹配较优的任务分配方案.     

NDM-Based Cognitive Agents for Supporting Decision-Making Teams

Naturalistic decision making (NDM) focuses on how people actually make decisions in realistic settings that typically involve ill-structured problems. Taking an experimental approach, we investigate the impacts of using an NDM-based software agent (R-CAST) on the performance of human decision-making teams in a simulated C3I (Communications, Command, Control and Intelligence) environment. We examined four types of decision-making teams with mixed human and agent members playing the roles of intelligence collection and command selection. The experiment also involved two within-group control variables: task complexity and context switching frequency. The result indicates that the use of an R-CAST agent in intelligence collection allows its team member to consider the latest situational information in decision making but might increase the team member's cognitive load. It also indicates that a human member playing the role of command selection should not rely too much on the agent serving as his or her decision aid. Together, it is suggested that the roles of both humans and cognitive agents are critical for achieving the best possible performance of C3I decision-making teams: Whereas agents are superior in computation-intensive activities such as information seeking and filtering, humans are superior in projecting and reasoning about dynamic situations and more adaptable to teammates' cognitive capacities. This study has demonstrated that cognitive agents empowered with NDM models can serve as the teammates and decision aids of human decision makers. Advanced decision support systems built upon such team-aware agents could help achieve reduced cognitive load and effective human-agent collaboration.     

WESTT (workload, error, situational awareness, time and teamwork): an analytical prototyping system for command and control

Modern developments in the use of information technology within command and control allow unprecedented scope for flexibility in the way teams deal with tasks. These developments, together with the increased recognition of the importance of knowledge management within teams present difficulties for the analyst in terms of evaluating the impacts of changes to task composition or team membership. In this paper an approach to this problem is presented that represents team behaviour in terms of three linked networks (representing task, social network structure and knowledge) within the integrative WESTT software tool. In addition, by automating analyses of workload and error based on the same data that generate the networks, WESTT allows the user to engage in the process of rapid and iterative “analytical prototyping”. For purposes of illustration an example of the use of this technique with regard to a simple tactical vignette is presented.