Search

The federated learning (FL) paradigm aims to distribute the computational burden of the training process among several computation units, usually called agents or workers, while preserving private local training datasets. This is generally achieved by resorting to a server–worker architecture where agents iteratively update local models and communicate local parameters to a server that aggregates and returns them to the agents. However, the presence of adversarial agents, which may intentionally exchange malicious parameters or may have corrupted local datasets, can jeopardize the FL process. Therefore, we propose selective trimmed average (SETA), which is a resilient algorithm to cope with the undesirable effects of a number of misbehaving agents in the global model. SETA is based on properly filtering and combining the exchanged parameters. We mathematically prove that the proposed algorithm is resilient against data and local model poisoning attacks. Most resilient methods presented so far in the literature assume that a trusted server is in hand. In contrast, our algorithm works both in server–worker and shared memory architectures, where the latter excludes the necessity of a trusted server. The theoretical findings are corroborated through numerical results on MNIST dataset and on multiclass weather dataset (MWD).

In this article, we address a time-varying quadratic optimization problem over a graph under the assumption that the problem shares the same sparsity pattern as the static graph encoding the undirected network topology over which the multiagent systems interact. Notably, this framework allows us to effectively model scenarios in which the optimization problem is inherently embedded within the network topology, e.g., flow balancing, electrical power system management, or packet routing problems. In this regard, we propose a finite-time distributed algorithm, which allows the multiagent system to track the time-varying optimal solution over time. Specifically, we first solve the frozen-time optimization problem, providing a necessary and sufficient condition for a solution to be globally optimal. Then, based on such conditions, a continuous-time distributed nonsmooth algorithm is developed. Numerical simulations are provided to corroborate the theoretical findings.

In this article, we develop a general open multiagent systems (OMAS) framework over undirected graphs where the agents' interaction is, in general, nonlinear, time-varying, and heterogeneous, in that the agents interact with different pairwise interaction rules for each link, possibly nonlinear, which may change over time. In particular, assuming the agents interact by exchanging flows, which modify their states, our framework guarantees that the sum of the states of agents participating in the network is preserved. To this end, agents maintain a state variable for each of their neighbors. Upon the disconnection of a neighbor, such a variable is used to completely eliminate the effect of previous interaction with disconnected agents from the overall system. In order to demonstrate the effectiveness of the proposed OMAS framework, we provide a case study focused on average consensus, and, specifically, we develop a sufficient condition on the structure of the agents' interaction guaranteeing asymptotic convergence under the assumption that the network becomes fixed. The article is complemented by simulation results that numerically demonstrate the effectiveness of the proposed method.

In this work, we propose a route planning strategy for heterogeneous mobile robots in Precision Agriculture (PA) settings. Given a set of agricultural tasks to be performed at specific locations, we formulate a multi-Steiner Traveling Salesman Problem (TSP) to define the optimal assignment of these tasks to the robots as well as the respective optimal paths to be followed. The optimality criterion aims to minimize the total time required to execute all the tasks, as well as the cumulative execution times of the robots. Costs for travelling from one location to another, for maneuvering and for executing the task as well as limited energy capacity of the robots are considered. In addition, we propose a sub-optimal formulation to mitigate the computational complexity by leveraging the fact that generally in PA settings only a few locations require agricultural tasks in a certain period of interest compared to all possible locations in the field. A formal analysis of the optimality gap between the optimal and the sub-optimal formulations is provided. The effectiveness of the approach is validated in a simulated orchard where three heterogeneous aerial vehicles perform inspection tasks. Note to Practitioners—This paper aims at providing an efficient solution to PA needs by deploying a team of robots able to perform agricultural tasks at given locations in large-scale orchards. In particular, a novel general optimization problem is proposed that, given a set of mobile and possibly heterogeneous robots and a set of agricultural tasks to carry out, defines the assignment of these tasks to the robots as well as the routes to follow, while minimizing the total and the cumulative execution times of the robots. Existing approaches for route optimization in PA generally involves complete coverage of the field by one or multiple robots and do not account for maneuvering costs with general layouts of the field. We consider costs for travelling from one location to another, for executing the task and for maneuvering without any restriction on the layout of the plants as well as we take into account the limited energy capacity of the robots. We also provide a sub-optimal formulation which reduces the computational burden by relaxing the optimization of the maneuvering costs at the locations where agricultural tasks are carried out and formally derive the optimality gap. The proposed approach is flexible and can be easily adapted to any PA setting involving multiple mobile robots that are required to accomplish given tasks in an area of interest. We validate its effectiveness in a realistic simulated setup composed of three heterogeneous aerial vehicles performing inspection tasks. In future research, we aim to design algorithms to solve the proposed optimization problems in an efficient manner as well as to validate the formulations on real-world robotic platforms.

In this article, we propose a load balancing problem formulation where agents cooperate with the aim of simultaneously minimizing both the workload disparity among the agents and the overall workload transfer, under network capacity constraints. Notably, in our computational setting, the network is not just a device for the distributed solution of an optimization problem; on the contrary, the problem shares the same sparsity pattern as the network, and this aspect allows us to solve it without the need for the agents to store large amount of data. In particular, while the load balancing process occurs over directed links, agents' communication is assumed to be bidirectional. For this optimization setting, first, an optimality condition is derived; then, a provably convergent distributed algorithm to compute the optimal solution is developed, and an upper bound on the convergence rate is characterized. Simulation results are provided to corroborate the validity and performance of our theoretical findings.

Identifying pests and treating them in a timely manner is a crucial aspect of the Precision Agriculture (PA) paradigm. Driven by the needs of the H2020 European Project Pantheon, focused on precision farming in hazelnut orchards, we propose a pest management system for gall-mites, which cause severe symptoms on generative and vegetative buds. We develop a data-driven monitoring system based on You Only Look Once (YOLO) framework enabling the early detection of the pest infestation with mean average precision of 86.7% on a holdout dataset. We perform a thorough analysis on its performance using several data augmentation methods as well as validate its real-time computation capability on a NVIDIA Jetson Xavier, which can be easily integrated into any robotic platform. Finally, we contextualize the role of the proposed detection framework in a comprehensive pest management system.

Water management is an important aspect in modern agriculture. Irrigation systems are becoming more and more complex, trying to minimize the water consumption while ensuring the necessities of the plants. A fundamental requirement to define efficient irrigation policies is to be able to estimate the water status of the plants and of the soil. In this context, precision agriculture addresses this problem by using the latest technological advancements. In particular, most of the works in the literature aim to develop highly accurate estimations under the assumption of the availability of a dense network of sensors. Although this assumption may be adequate for intensive farming (e.g. greenhouses), it becomes quite unrealistic in the context of large-scale scenarios. In this work, we propose a novel observer-based architecture for the water management of large-scale (hazelnut) orchards which relies on a network of sparsely deployed soil moisture sensors along with a weather station and on remote sensing measurements carried out by drones with a pre-defined periodicity. The contribution is twofold: i) First a novel model of the water dynamics in an hazelnut orchard is proposed, which includes the water dynamics in the soil and in the plants, and ii) then, on the basis of this model and of the available measurements, the use of a Kalman filter with intermittent observations is proposed, taking also into account the availability of the weather station measurements. The effectiveness of the proposed solution is validated through simulation.

This article proposes a swarm-based path-following guidance system for an autonomous marine multi-vehicle system. In particular, a team of unmanned surface vehicles is required to converge to and navigate along a desired reference path, while at the same time aggregating and maintaining a range-based formation configuration. First, a separate description is given for a swarm methodology, initially developed for small ground mobile robots and exploited to aggregate the robot team, and a virtual target–based path-following guidance system developed for unmanned surface vehicles, exploited to drive not the single vehicles but the robot formation as a whole. Then, the integration of the two proposed methodologies is reported and proven, in order to guarantee the feasibility and stability of the overall guidance framework. Simulative results are proposed to validate the effectiveness of the proposed methodology and to evaluate the system performances.

In this paper, we consider the problem of identifying a bipartition of a planar multi-robot network, such that the resulting two sub-teams are rigid networks. As opposed to approaching the network splitting problem constructively, we instead determine the existence conditions for rigidity-preserving bipartitions, and provide an iterative algorithm that identifies such partitions in polynomial time. In particular, the relationship between rigid graph partitions and the previously identified Z-link edge structure is given, yielding a direction towards which a graph search is applied. Adapting a supergraph search mechanism from the set generation literature, we then provide a methodology for discerning graphs cuts that represent valid rigid bipartitions. Finally, full algorithm details and pseudocode are provided, together with simulation results that verify correctness and demonstrate complexity.

This paper focuses on the integration of a cooperative swarm methodology initially developed for small ground mobile robots, with a virtual target based path-following guidance system developed for marine surface vehicle systems. After separately describing the approaches, for swarm aggregation and path-following guidance respectively, the paper reports the effortless procedure to perform the integration of the proposed techniques. Simulative and experimental results are then proposed to prove the effectiveness of the proposed methodology.

Abstract: Localization, i.e., estimating a robot pose relative to a map of an environment, is one of the most relevant problems in mobile robotics. The research community has devoted a big effort to provide solutions for the localization problem. Several methodologies have been proposed, among them the Kalman filter and Monte Carlo Localization filters. In this paper, the Clustered Evolutionary Monte Carlo filter (CE-MCL) is presented. This algorithm, taking advantage of an evolutionary approach along with a clusterization method, is able to overcome classical MCL filter drawbacks. Exhaustive experiments, carried on the robot ATRV-Jr manufactured by iRobot, are shown to prove the effectiveness of the proposed CE-MCL filter.