Papers by Nicolas Jouandeau
Swarm Intelligence and IoT-Based Smart Cities: A Review
Internet of Things
Smart cities are complex and large distributed systems characterized by their heterogeneity, secu... more Smart cities are complex and large distributed systems characterized by their heterogeneity, security, and reliability challenges. In addition, they are required to take into account several scalability, efficiency, safety, real-time responses, and smartness issues. All of this means that building smart city applications is extremely complex. Swarm Intelligence is a very promising paradigm to deal with such complex and dynamic systems. It presents robust, scalable and self-organized behaviors to deal with dynamic and fast changing systems. The intelligence of cities can be modeled as a swarm of digital telecommunication networks (the nerves), ubiquitously embedded intelligence (the brains), sensors and tags (the sensory organs), and software (the knowledge and cognitive competence). In this chapter, swarm intelligence-based algorithms and existing swarm intelligence-based smart city solutions will be analyzed. Moreover, a swarm-based framework for smart cities will be presented. Then, a set of trends on how to use swarm intelligence in smart cities, in order to make them flexible and scalable, will be investigated.
Sample-Label View Transfer Active Learning for Time Series Classification
Lecture Notes in Computer Science
Intelligent Computing, 2019
Farming has seen a number of technological transformations in the last decade, becoming more indu... more Farming has seen a number of technological transformations in the last decade, becoming more industrialized and technology-driven. This means use of Internet of Things(IoT), Cloud Computing(CC), Big Data (BD) and automation to gain better control over the process of farming. As the use of these technologies in farms has grown exponentially with massive data production, there is need to develop and use state-of-the-art tools in order to gain more insight from the data within reasonable time. In this paper, we present an initial understanding of Convolutional Neural Network (CNN), the recent architectures of stateof-the-art CNN and their underlying complexities. Then we propose a classification taxonomy tailored for agricultural application of CNN. Finally, we present a comprehensive review of research dedicated to applications of state-of-the-art CNNs in agricultural production systems. Our contribution is in twofold. First, for end users of agricultural deep learning tools, our benchmarking finding can serve as a guide to selecting appropriate architecture to use. Second, for agricultural software developers of deep learning tools, our in-depth analysis explains the state-ofthe-art CNN complexities and points out possible future directions to further optimize the running performance.
Swarm intelligence-based algorithms within IoT-based systems: A review
Journal of Parallel and Distributed Computing
Abstract IoT-based systems are complex and dynamic aggregations of entities (Smart Objects) which... more Abstract IoT-based systems are complex and dynamic aggregations of entities (Smart Objects) which usually lack decentralized control. Swarm Intelligence systems are decentralized, self-organized algorithms used to resolve complex problems with dynamic properties, incomplete information, and limited computation capabilities. This study provides an initial understanding of the technical aspects of swarm intelligence algorithms and their potential use in IoT-based applications. We present the existing swarm intelligence-based algorithms with their main applications, then we present existing IoT-based systems that use SI-based algorithms. Finally, we discuss trends to bring together swarm intelligence and IoT-based systems. This review will pave the path for future studies to easily choose the appropriate SI-based algorithm for IoT-based systems.
Journal of Agricultural Informatics
In this paper, we present an optimized Machine Learning (ML) algorithm for predicting land suitab... more In this paper, we present an optimized Machine Learning (ML) algorithm for predicting land suitability for crop (sorghum) production, given soil properties information. We setup experiments using Parallel Random Forest (PRF), Linear Regression (LR), Linear Discriminant Analysis (LDA), KNN, Gaussian Naïve Bayesian (GNB) and Support Vector Machine (SVM). Experiments were evaluated using 10 cross fold validation. We observed that, parallel random forest had a better accuracy of 0.96 and time of execution of 1.7 sec. Agriculture is the main stream of food security. Kenya relies on agriculture to feed its population. Land evaluation gives potential of land use, in this case for crop production. In the Department of Soil Survey in Kenya Agriculture and Livestock Research Organization (KALRO) and other soil research organizations, land evaluation is done manually, is stressful, takes a long time and is prone to human errors. This research outcomes can save time and improve accuracy in land evaluation process. We can also be able to predict land suitability for crop production from soil properties information without intervention of a soil scientist expert. Therefore, agricultural stakeholders will be able to efficiently make informed decisions for optimal crop production and soil management.
Complex Adaptive Systems Modeling
Background The foraging task is one of the canonical testbeds for cooperative robotics, in which ... more Background The foraging task is one of the canonical testbeds for cooperative robotics, in which a collection of robots has to search and transport objects to specific storage point(s). In this paper, we investigate the Multi-Agent Foraging (MAF) problem from several perspectives that we analyze in depth. Results First, we define the Foraging Problem according to literature definitions. Then we analyze previously proposed taxonomies, and propose a new foraging taxonomy characterized by four principal axes: Environment, Collective, Strategy and Simulation, summarize related foraging works and classify them through our new foraging taxonomy. Then, we discuss the real implementation of MAF and present a comparison between some related foraging works considering important features that show extensibility, reliability and scalability of MAF systems Conclusions Finally we present and discuss recent trends in this field, emphasizing the various challenges that could enhance the existing MA...
2015 Conference on Technologies and Applications of Artificial Intelligence (TAAI), 2015
Alpha-beta and Monte-Carlo Tree Search (MCTS) are two powerful paradigms useful in computer games... more Alpha-beta and Monte-Carlo Tree Search (MCTS) are two powerful paradigms useful in computer games. When considering imperfect information, the tree that represents the game has to deal with chance. When facing such games that presents increasing branching factor, MCTS may consider, as alpha-beta do, pruning to keep efficiency. We present a modified version of MCTS-Solver algorithm, called OPI-MCTS as Optimal Player Information MCTS, that adds game state information to exploit logical reasoning during backpropagation and that influences selection and expansion. OPI-MCTS is experimented in Chinese Dark Chess, which is a imperfect information game. OPI-MCTS is compared with classical MCTS.
S-MASA: A stigmergy based algorithm for multi-target search
2014 Federated Conference on Computer Science and Information Systems, Sep 1, 2014
This paper presents a new approach for collaborative multi-robot planning issues. The main proble... more This paper presents a new approach for collaborative multi-robot planning issues. The main problem that arises from multi-robot exploration is waiting situations. We consider that such problem involves two or more autonomous robots in an unknown environment. The mission objective is to explore the entire map, while trying to minimize its executing time. Moreover if each robot uses the same topological graph, then it uses the same exploration path that makes waiting situations arising. To solve this problem, we propose a new approach in this paper based on sampling iteratively maps to allow interactive multi-robot exploration. Our approach has been implemented in simulation and the experiments demonstrate that the overall completion time of an exploration task can be significantly reduced by our sampling-based method.
We present algorithms to compute and match maze specific deadlock tables at Sokoban. They enable ... more We present algorithms to compute and match maze specific deadlock tables at Sokoban. They enable to use a greedy search that can solve problems that are not solvable using IDA*.
Habilitation à Diriger des Recherches
... L'Université des Sciences et Technologies de Lille Discipline : Sciences Physiques par L... more ... L'Université des Sciences et Technologies de Lille Discipline : Sciences Physiques par LaurentThais Modélisation des écoulements complexes Application aux géo-fluides et aux fluides viscoélastiques Jury J.-F. AGASSANT Professeur, École des Mines de Paris Examinateur ...
Varying Complexity in CHINESE DARK CHESS Stochastic Game
RSRT: Rapidly Exploring Sorted Random Tree
ISR/ROBOTIK 2010, 2010
This paper presents a new approach for collaborative multi-robot planning issues. The main proble... more This paper presents a new approach for collaborative multi-robot planning issues. The main problem that arises from multi-robot exploration is waiting situations. We consider that such problem involves two or more autonomous robots in an unknown environment. The mission objective is to explore the entire map, while trying to minimize its executing time. Moreover if each robot uses the same topological graph, then it uses the same exploration path that makes waiting situations arising. To solve this problem, we propose a new approach in this paper based on sampling iteratively maps to allow interactive multi-robot exploration. Our approach has been implemented in simulation and the experiments demonstrate that the overall completion time of an exploration task can be significantly reduced by our sampling-based method.
Many problems have a huge state space and no good heuristic to order moves so as to guide the sea... more Many problems have a huge state space and no good heuristic to order moves so as to guide the search toward the best positions. Random games can be used to score positions and evaluate their interest. Random games can also be improved using random games to choose a move to try at each step of a game. Nested Monte-Carlo Search addresses the problem of guiding the search toward better states when there is no available heuristic. It uses nested levels of random games in order to guide the search. The algorithm is studied theoretically on simple abstract problems and applied successfully to three different games: Morpion Solitaire, SameGame and 16x16 Sudoku.
Algorithmique de la planification de mouvement probabiliste pour un robot mobile
Nous etudions dans ce memoire la planification de mouvement probabiliste incrementale. Nos travau... more Nous etudions dans ce memoire la planification de mouvement probabiliste incrementale. Nos travaux se concretisent par un nouvel algorithme de construction des arbres aleatoires d'exploration rapide et une nouvelle decomposition de l'espace des configurations en cellules irregulieres. Partant d'une synthese des methodes de planification incrementale probabiliste, nos travaux presentent un algorithme de construction accelerant l'exploration de l'espace de recherche. Partant des principales approches d'echantillonnage de l'espace de recherche utilisees dans les methodes probabilistes, l'analyse des proprietes associees a ces echantillonnages nous conduit a proposer une decomposition de l'espace en cellules irregulieres adaptee a la notion d'accessibilite. Apres definition de cette decomposition, ces algorithmes de construction sont evalues comparativement a un echantillonnage uniforme de l'espace de recherche.
Retinal Vessel Segmentation Based on Adaptive Random Sampling
Journal of Medical and Bioengineering, 2014
Improved trade-based multi-robot coordination
2011 6th IEEE Joint International Information Technology and Artificial Intelligence Conference, 2011
ABSTRACT Team work is essential to multiple mobile robot systems. An important question is, which... more ABSTRACT Team work is essential to multiple mobile robot systems. An important question is, which robot should imple- ment which action? In our previous work, we presented a trade-based task allocation approach for coordinated multi- robot exploration, which simulates the relationship between buyers and sellers in a business system, to achieve dynamic task allocation by using a mechanism of unsolicited bid. This paper still addresses the problem of coordinating multi-robot exploration while presents an improved trade-based approach to raise the efficiency of task allocation by using the Hungarian method. The proposed approach has been implemented and evaluated in simulation. The experimental results demonstrate the total exploration time can be significantly reduced by the improved trade-based approach compared to previous approaches.
Automatic Generation of Humanoid’s Geometric Model Parameters
Lecture Notes in Computer Science, 2014
Decentralized waypoint-based multi-robot coordination
2012 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER), 2012
ABSTRACT This paper describes a decentralized approach based on Alternative Waypoint Generating (... more ABSTRACT This paper describes a decentralized approach based on Alternative Waypoint Generating (AWG) for planning separate kinematic paths to multiple robots to alleviate the waiting situation problems. The basic thought of this approach is straight-forward: If a robot request a target waypoint that has already been assigned to another robot, then this robot will consider this waypoint as an obstacle and attempt to get an alternative one around it. The proposed strategy has been implemented and evaluated in simulation. The experimental results demonstrated that the waiting situation problems have been decreased and the overall system performance has been improved.
Uploads
Papers by Nicolas Jouandeau