17 research outputs found
Adaptive exploration of a UAVs swarm for distributed targets detection and tracking
No full textThis paper focuses on the problem of coordinating multiple UAVs for distributed targets detection and tracking, in different technological and environmental settings. The proposed approach is founded on the concept of swarm behavior in multi-agent systems, i.e., a self-formed and self-coordinated team of UAVs which adapts itself to mission-specific environmental layouts. The swarm formation and coordination are inspired by biological mechanisms of flocking and stigmergy, respectively. These mechanisms, suitably combined, make it possible to strike the right balance between global search (exploration) and local search (exploitation) in the environment. The swarm adaptation is based on an evolutionary algorithm with the objective of maximizing the number of tracked targets during a mission or minimizing the time for target discovery. A simulation testbed has been developed and publicly released, on the basis of commercially available UAVs technology and real-world scenarios. Experimental results show that the proposed approach extends and sensibly outperforms a similar approach in the literature
A hyper-heuristic methodology for coordinating swarms of robots in target search
No full textTarget search aims to discover elements of various complexity in a physical environment, by minimizing the overall discovery time. Different swarm intelligence algorithms have been proposed in the literature, inspired by biological species. Despite the success of bio-inspired techniques (bio-heuristics), there are relevant algorithm selection and parameterization costs associated with every new type of mission and with new instances of known missions. In this paper, evolutionary optimization is proposed for achieving significant improvements of the mission performance. Although adaptive, the logic of bio-heuristics is nevertheless constrained by models of biological species. To generate more adaptable logics, a novel design approach based on hyper-heuristics is proposed, in which the differential evolution optimizes the aggregation and tuning of modular heuristics for a given application domain. A modeling and optimization testbed has been developed and publicly released. Experimental results on real-world scenarios show that the hyper-heuristics based on stigmergy and flocking significantly outperform the adaptive bio-heuristics
High Resolution Mapping of Vegetation Biodiversity by Hyperspectral Images and Convolutional Autoencoders
No full textA methodology is presented to map the vegetation biodiversity based on the hypothesis of the spectral variation (SV) which has been proposed to assess the forest biodiversity by means of Earth Observation (EO) data. Hyperspectral data acquired by the PRecursore Iperspettrale della Missione Applicativa (PRISMA) mission of the Italian Space Agency to spectral signature with a high spectral resolution. The NDVI is computed from PRISMA data and used to identify pixels corresponding to vegetation cover. The spectral signatures at those pixels are then clusterized using the convolutional autoenconders technique and the final map with the location of pixels belonging to the different classes is produce. The methodology is applied to assess the vegetation biodiversity in National Parks of Gargano, Alta Murgia, Cilento-Vallo di Diano-Alburni, Appennino Lucano Val D’Agri Lagonegrese and Pollino, all located in Southern Italy
Stock price forecasting over adaptive timescale using supervised learning and receptive fields
No full textPattern recognition in financial time series is not a trivial task, due to level of noise, volatile context, lack of formal definitions and high number of pattern variants. A current research trend involves machine learning techniques and online computing. However, medium-term trading is still based on human centric heuristics, and the integration with machine learning support remains relatively unexplored. The purpose of this study is to investigate potential and perspectives of a novel architectural topology providing modularity, scalability and personalization capabilities. The proposed architecture is based on the concept of Receptive Fields (RF), i.e., sub-modules focusing on specific patterns, that can be connected to further levels of processing to analyze the price dynamics on different granularities and different abstraction levels. Both Multilayer Perceptrons (MLP) and Support Vector Machines (SVM) have been experimented as a RF. Early experiments have been carried out over the FTSEMIB index
Managing the oceans cleanup via sea current analysis and bio-inspired coordination of USV swarms
No full textThis work presents the results of a simulated analysis concerning algorithms of self-coordination of a swarm of Unmanned Surface Vehicles (USV) for the mitigation of plastic pollution in oceans. The analysis is based on real scenarios provided by the Copernicus Marine Service. The scenario includes the localization of plastics on the sea surface and their movement in time based on the sea surface currents. A swarm intelligence algorithm is used for the decentralized coordination of the USV swarm. Results are presented on a study area located in the northern Tyrrhenian sea between Corsica and the Tuscan coast, in the period of July 2016
Using Artificial Immune System to Prioritize Swarm Strategies for Environmental Monitoring
No full textA Machine-Learning Approach for Generating Synthetic Prisma Hyperspectral Images from Multispectral Data
No full textThe scarcity of a sufficiently large and representative hyperspectral image dataset is a substantial obstacle to the effective development of algorithms for remote sensing applications. Hyperspectral images can provide rich spectral information for various tasks, such as land cover classification, vegetation monitoring, and environmental assessment. However, the limited availability of diverse and well-annotated hyperspectral datasets hinders the development and optimization of these models in this domain. For this purpose, the generation of synthetic hyperspectral images has emerged as a pivotal area of research.This paper aims to introduce a preliminary analysis of various AI-based methodologies specifically crafted to generate synthetic PRISMA hyperspectral images derived from Sentinel-2 data. By exploring innovative approaches, this study aims to develop novel techniques for creating synthetic datasets, providing valuable insights into the potential of synthetic hyperspectral imagery for algorithm training and evaluation in the absence of extensive real-world hyperspectral datasets
Using VLF time series from the INFREP network for the study of pre-seismic radio anomalies
No full textThis work presents an application of the Perceptually Important Points (PIP) technique for the analysis of VLF time series. The aim of the analysis is to detect anomalies with respect to the normal variations of the data trends. Such anomalies could reveal possible radio precursors of the earthquake. Since 2009, several radio receivers have been installed throughout Europe in order to realize the INFREP European radio network for studying the VLF (10-50 kHz) and LF (150-300 kHz) radio precursors of earthquakes. The time series used for experiments was collected during the Dodecanese islands earthquakes (MW=5.6 and MW=5.7) occurred on January 30, 2020
Progettazione e realizzazione di un algoritmo di coordinamento di sciami di droni basato sulla swarm intelligence per il rilevamento e l'inseguimento di target dinamici
No full textNegli ultimi anni i droni sono stati sempre più usati in applicazioni in ambito civile. Infatti, grazie ai progressi tecnologici nella sensoristica, queste piattaforme aeree possono essere equipaggiate con molteplici tipologie di payload, come camere visuali o termocamere. Inoltre, l’uso dei droni si rivela particolarmente adatto per missioni caratterizzate dalle 3D (Dull, Dirty and Dangerous).
Il problema del rilevamento e monitoraggio di fenomeni che evolvono nel tempo è molto indicato per essere affrontato attraverso l’uso di sciami di droni, perché garantisce robustezza, scalabilità e flessibilità. Nel presente lavoro di tesi, il controllo e coordinamento degli sciami di droni per la ricerca di target in ambienti non strutturati è realizzato mutuando meccanismi comportamentali tipici della swarm intelligence, come il flocking e la stigmergia, ed è emulato grazie ad un simulatore progettato ad hoc. Inoltre, il comportamento parametrico, sia dei droni che dell’ambiente, è specializzato rispetto ad uno scenario specifico per mezzo di un algoritmo evolutivo.
Grazie alla modifica della struttura logica del simulatore, si è potuto valutare la performance relativa al coordinamento di sciami di droni su scenari caratterizzati da target in movimento. Tra gli scenari in cui è auspicabile usare gli sciami di droni troviamo il rilevamento remoto e monitoraggio di incendi boschivi. A tal proposito si sono costruiti due scenari realistici, uno per la detection e l’altro per il tracking di incendi boschivi. Infine, si sono selezionati due droni commerciali, adeguatamente equipaggiati per la missione, e si è valutata la performance dello sciame sugli scenari dinamici costruiti
Coordination of swarms of robots in target search: from bio-inspired heuristics to hyper-heuristics
No full text
