Gabriele Valentini

A Design Pattern for Decentralised Decision Making.

The engineering of large-scale decentralised systems requires sound methodologies to guarantee the attainment of the desired macroscopic system-level behaviour given the microscopic individual-level implementation. While a general-purpose methodology is currently out of reach, specific solutions can be given to broad classes of problems by means of well-conceived design patterns. We propose a design pattern for collective decision making grounded on experimental/theoretical studies of the nest-site selection behaviour observed in honeybee swarms (Apis mellifera). The way in which honeybee swarms arrive at consensus is fairly well-understood at the macroscopic level. We provide formal guidelines for the microscopic implementation of collective decisions to quantitatively match the macroscopic predictions. We discuss implementation strategies based on both homogeneous and heterogeneous multiagent systems, and we provide means to deal with spatial and topological factors that have a bearing on the micro-macro link. Finally, we exploit the design pattern in two case studies that showcase the viability of the approach. Besides engineering, such a design pattern can prove useful for a deeper understanding of decision making in natural systems thanks to the inclusion of individual heterogeneities and spatial factors, which are often disregarded in theoretical modelling.

The Best-of-n Problem in Robot Swarms: Formalization, State of the Art, and Novel Perspectives

The ability to collectively choose the best among a finite set of alternatives is a fundamental cognitive skill for robot swarms. In this paper, we propose a formal definition of the best-of-n problem and a taxonomy that details its possible variants. Based on this taxonomy, we analyze the swarm robotics literature focusing on the decision-making problem dealt with by the swarm. We find that, so far, the literature has primarily focused on certain variants of the best-of-n problem while other variants have been the subject of only a few isolated studies. Additionally, we consider a second taxonomy about the design methodologies used to develop collective decision-making strategies. Based on this second taxonomy, we provide an in-depth survey of the literature that details the strategies proposed so far and discusses the advantages and disadvantages of current design methodologies.

Majority Rule with Differential Latency: An Absorbing Markov Chain to Model Consensus

We study collective decision-making in a swarm of robots. We consider the majority rule with differential latency: robots randomly form teams, make a decision following the majority rule, and then turn in a latent state whose duration depends on the decision made. While latent, robots do not participate in the decision mechanism, thus, the differential latency provides a positive feedback that favors the decision with the shortest latency. We analyze the dynamics using a discrete, time-homogeneous, absorbing Markov chain.

Evoptool: An extensible toolkit for evolutionary optimization algorithms comparison

This paper presents Evolutionary Optimization Tool (Evoptool), an optimization toolkit that implements a set of meta-heuristics based on the Evolutionary Computation paradigm. Evoptool provides a common platform for the development and test of new algorithms, in order to facilitate the performance comparison activity. The toolkit offers a wide set of benchmark problems, from classical toy examples to complex tasks, and a collection of implementations of algorithms from the Genetic Algorithms and Estimation of Distribution Algorithms paradigms. Evoptool is flexible and easy to extend, also with algorithms based on other approaches that go beyond Evolutionary Computation.

Introducing ℓ 1 -regularized logistic regression in Markov Networks based EDAs

Estimation of Distribution Algorithms evolve populations of candidate solutions to an optimization problem by introducing a statistical model, and by replacing classical variation operators of Genetic Algorithms with statistical operators, such as estimation and sampling. The choice of the model plays a key role in the evolutionary process, indeed it strongly affects the convergence to the global optimum. From this point of view, in a black-box context, especially when the interactions among variables in the objective function are sparse, it becomes fundamental for an EDA to choose the right model, able to encode such correlations. In this paper we focus on EDAs based on undirected graphical models, such as Markov Networks. To learn the topology of the graph we apply a sparse method based on ℓ1-regularized logistic regression, which has been demonstrated to be efficient in the high-dimensional case, i.e., when the number of observations is much smaller than the sample space. We propose a new algorithm within the DEUM framework, called DEUMℓ1, able to learn the interactions structure of the problem without the need of prior knowledge, and we compare its performance with other popular EDAs, over a set of well known benchmarks.

Collective Perception of Environmental Features in a Robot Swarm

In order to be effective, collective decision-making strategies need to be not only fast and accurate, but sufficiently general to be ported and reused across different problem domains. In this paper, we propose a novel problem scenario, collective perception, and use it to compare three different strategies: the DMMD, DMVD, and DC strategies. The robots are required to explore their environment, estimate the frequency of certain features, and collectively perceive which feature is the most frequent. We implemented the collective perception scenario in a swarm robotics system composed of 20 e-pucks and performed robot experiments with all considered strategies. Additionally, we also deepened our study by means of physics-based simulations. The results of our performance comparison in the collective perception scenario are in agreement with previous results for a different problem domain and support the generality of the considered strategies.

Time-variant feedback processes in collective decision-making systems: influence and effect of dynamic neighborhood sizes

Self-organizing systems rely on positive feedback (amplification of perturbations). In particular, in swarm systems, positive feedback builds up in a transient phase until maximal positive feedback is reached and the system converges temporarily on a state close to consensus. We investigate two examples of swarm systems showing time-variant positive feedback: alignment in locust swarms and adaptive aggregation of swarms. We identify an influencing bias in the spatial distribution of agents compared to a well-mixed distribution and two features, percentage of aligned swarm members and neighborhood size, that allow us to model the time variance of feedbacks. We report an urn model that is capable of qualitatively representing all these relevant features. The increase in neighborhood sizes over time enables the swarm to lock in a highly aligned state but also allows for infrequent switching between lock-in states. We report similar occurrences of time-variant feedback in a second collective system to indicate the potential for generality of this phenomenon. Our study is concluded by applications of methods from renormalization group theory that allow us to focus on the neighborhood dynamics as scale transformations. Correlation lengths and critical exponents are determined empirically.

Spatially targeted communication in decentralized multirobot systems

Spatially targeted communication (STC) allows a message sender to choose message recipients based on their location in space. Currently, STC in multirobot systems is limited to centralized systems. In this paper, we propose a novel communication protocol that enables STC in decentralized multirobot systems. The proposed protocol dispenses with the many aspects that underpin previous approaches, including external tracking infrastructure, a priori knowledge, global information, dedicated communication devices or unique robot IDs. We show how off-the-shelf hardware components such as cameras and LEDs can be used to establish ad-hoc STC links between robots. We present a Markov chain model for each of the two constituent parts of our proposed protocol and we show, using both model-based analysis and experimentation, that the proposed protocol is highly scalable. We also present the results of extensive experiments carried out on an autonomous, heterogeneous multirobot system composed of one aerial robot and numerous ground-based robots. Finally, two real world application scenarios are presented in which we show how spatial coordination can be achieved in a decentralized multirobot system through STC.

Derivation of a Micro-Macro Link for Collective Decision-Making Systems

Relating microscopic features (individual level) to macroscopic features (swarm level) of self-organizing collective systems is challenging. In this paper, we report the mathematical derivation of a macroscopic model starting from a microscopic one for the example of collective decision-making. The collective system is based on the application of a majority rule over groups of variable size which is modeled by chemical reactions (micro-model). From an approximated master equation we derive the drift term of a stochastic differential equation (macro-model) which is applied to predict the expected swarm behavior. We give a recursive definition of the polynomials defining this drift term. Our results are validated by Gillespie simulations and simulations of the locust alignment.

Kilogrid: a novel experimental environment for the Kilobot robot

We present the Kilogrid, an open-source virtualization environment and data logging manager for the Kilobot robot, Kilobot for short. The Kilogrid has been designed to extend the sensory-motor abilities of the Kilobot, to simplify the task of collecting data during experiments, and to provide researchers with a tool to fine-control the experimental setup and its parameters. Based on the design of the Kilobot and compatible with existing hardware, the Kilogrid is a modular system composed of a grid of computing nodes, or modules that provides a bidirectional communication channel between the Kilobots and a remote workstation. In this paper, we describe the hardware and software architecture of the Kilogrid system as well as its functioning to accompany its release as a new open hardware tool for the swarm robotics community. We demonstrate the capabilities of the Kilogrid using a 200-module Kilogrid, swarms of up to 100 Kilobots, and four different case studies: exploration and obstacle avoidance, site selection based on multiple gradients, plant watering, and pheromone-based foraging. Through this set of case studies, we show how the Kilogrid allows the experimenter to virtualize sensors and actuators not available to the Kilobot and to automatize the collection of data essential for the analysis of the experiments.