Papers by NourEddine Djedi
Rendu De Surfaces Par Deformation De Texels
lesialab.net
In this paper a virtual ecosystem environment with basic physical law and energy concept has been... more In this paper a virtual ecosystem environment with basic physical law and energy concept has been proposed, this ecosystem is populated with 3D virtual creatures that are living in this environment in order to forage food. Artificial behaviours are developed to control virtual creatures. A genetic algorithm with an artificial neural network were implemented together to guarantee some of these behaviours like searching food. Foods are presented in different locations in the virtual ecosystem. The evolutionary process uses the physical properties of the virtual creatures and an external fitness function that will conduct to the expected behaviours. The experiment evolving locomoting virtual creatures show that these virtual creatures try to obtain at least one of the food sources presented in its trajectory. Our bestevolved creatures are able to reach multiple food sources during the simulation time.
無線ボディセンサネットワークにおける公平と高速データ収集のための動的協力マッチングゲームアプローチ【Powered by NICT】
IEEE Conference Proceedings, 2017
Predator-prey coevolution in a physically simulated environment
ABSTRACT This paper presents a predator-prey model that is the key element of ecological systems.... more ABSTRACT This paper presents a predator-prey model that is the key element of ecological systems. We present the results of coevolving artificial creatures of two species (predator and prey) in a 3D physically simulated environment. The evolutionary process uses the physical properties of virtual creatures and two opposite fitness functions to drive the process to the behaviors we are looking for. Our best-evolved creatures are able to reach their adversary during the simulation time.
Modelisation en synthese d'images : utilisation d'une methodologie declarative
... Auteur(s) / Author(s). Djedi Noureddine ; Caubet René (Directeur de thèse) ; Affiliation(s) d... more ... Auteur(s) / Author(s). Djedi Noureddine ; Caubet René (Directeur de thèse) ; Affiliation(s) du ou des auteurs / Author(s) Affiliation(s). Université de Toulouse 3, Toulouse, FRANCE (Université de soutenance) Résumé / Abstract. ...
Recognition of 3D Faces with Missing Parts Based on SIFT and LBP Methods
Signal processing for security technologies, Dec 24, 2016
Presently, 3D face recognition researched solutions confronted the problem of recognizing 2D. In ... more Presently, 3D face recognition researched solutions confronted the problem of recognizing 2D. In our contribution, we specifically discuss major difficulties further to propose and test a novel solution of 3D face recognition that is significantly capable to perform the recognition subject, in cases where the analysis of only a part of the face. With the proposed approach, the distinctive features of the face are captured by first extracting SIFT keypoints on the face of analysis and measure how the face changes along profiles built between pairs of keypoints, second we applied the operator SIFT on LBPP,R images, separately. Following the work of Faltemier and al. then Tang and al., we can better detect a number of keypoints by using SIFT on LBPP, R images, than using SIFT on the original images. The contribution is tested using the whole of the Face Recognition Grand Challenge FRGC v1.0 data. Finally, we perform a classification based on SVM process.
Computer Networks, Dec 1, 2017
NEAT neural networks to control and simulate virtual creature's locomotion
ABSTRACT In this paper, we present the results of our experimentation of evolutionary reinforceme... more ABSTRACT In this paper, we present the results of our experimentation of evolutionary reinforcement learning technique for evolving a variety of autonomous virtual creatures in a 3D physically simulated environment. Evolutionary reinforcement learning uses a neuroevolution NEAT technique, which based upon three main ideas (Genetic Encoding with Historical Markings, Speciation and Minimizing Dimensionality). The morphology of creatures is completely predefined before the evolution. Laws of physics are simulated using the ODE library. Experiments have shown that our model based upon NEAT neural networks is capable to allow locomotion strategies to develop optimal behaviours. The obtained results have demonstrated that NEAT algorithm outperform genetic algorithm for evolving autonomous virtual creatures problems.
Springer eBooks, 2012
This article describes a bio-inspired system and the associated series of experiments, for the ev... more This article describes a bio-inspired system and the associated series of experiments, for the evolution of walking behavior in a simulated humanoid robot. A previous study has demonstrated the potential of this approach for evolving controllers based on simulated humanoid robots with a restricted range of movements. The development of anthropomorphic bipedal locomotion is addressed by means of artificial evolution using a genetic algorithm. The proposed task is investigated using full rigid-body dynamics simulation of a bipedal robot with 15 degrees of freedom. Stable bipedal gait with a velocity of 0.94 m/s is realized. Locomotion controllers are evolved from scratch, for example neither does the evolved controller have any a priori knowledge on how to walk, nor does it have any information about the kinematics structure of the robot. Instead, locomotion control is achieved based on intensive use of sensory information. In this work, the emergence of non-trivial walking behaviors is entirely due to evolution.
Journal of Automation, Mobile Robotics & Intelligent Systems, Jul 18, 2019
In order to be fully autonomous, robots have to be resilient so that they can recover from damage... more In order to be fully autonomous, robots have to be resilient so that they can recover from damages and operate for a long period of time with no human assistance. To be resilient, existing approaches propose to change the robots' behavior using a different control system when a hardware fault or damage occurs. These approaches are used for robots which have fixed morphologies. However, we cannot assume which morphology would be optimal for a given problem and which morphology allows resilience. In the present paper, we introduce a new approach that generates resilient artificial modular robots by evolving the robot morphology along with its controller. We used a multi-objective evolutionary algorithm to optimize two objectives at a time, which are the traveled distance of a damage-free robot and the traveled distance of the same robot with damaged parts. The result of preliminary experiments demonstrates that during evaluation, when robots are deliberately faced to motor failures, the evolution process can optimize and generate new morphologies for which the robot's behavior is less affected by damage. This makes the robot capable to recover its ability to move forward.
This paper describes a bacterial system that reproduces a population of bacteria that behave by s... more This paper describes a bacterial system that reproduces a population of bacteria that behave by simulating the internal reactions of each bacterial cell. The chemotaxis network of a cell is modulated by a hybrid approach that uses an algebraic model for the receptor clusters activity and an ordinary differential equation for the adaptation dynamics. The experiments are defined in order to simulate bacterial growth in an environment where nutrients are regularly added to it. The results show analysis of the motion obtained by some bacteria and their effects on the population behaviors generated by evolution. This evolution allows bacteria to have the ability to adapt themselves to better growth in the available food existed in its environment and to survive.
Artificial Life and Robotics, Nov 19, 2014
Dans le problème de l'estimation de mouvement dans une séquence d'images, deux performances criti... more Dans le problème de l'estimation de mouvement dans une séquence d'images, deux performances critiques sont à améliorer, le temps de calcul et la précision des résultats. La méthode itérative multi-échelles de Lucas et Kanade est l'une des meilleures méthodes dans ce domaine. Dans cet article on a introduit le réseau de neurones de Zhang pour accélérer le processus d'estimation suite à la nature parallèle de ces derniers. Du coté précision, on a proposé d'utiliser la transformée en curvelettes pour le pré-filtrage des images. Les simulations numériques sur des séquences d'images artificielles et réelles montrent la bonne performance de la méthode proposée.
International Journal of Computers and Applications, 2013
The aim of this study is to present a fast parallel implementation of the Horn and Schunck method... more The aim of this study is to present a fast parallel implementation of the Horn and Schunck method using a new kind of recurrent neural network called discrete Zhang neural networks. This network is characterized by a few iterations to converge which make it very suitable for real-time motion estimation. To compute the optical flow, we propose to solve directly the system of equations utilizing the discrete Zhang neural networks for matrix inversion instead of the original iterative method. The simulation results for synthetic and real image sequences show that the proposed algorithm is faster than the Horn and Schunck method.
Quantum Genetic Algorithm for Evolving Neural Controllers
Advanced Science Letters, 2018
PEvoRNN: Extended CUDA-based hybrid parallel evolutionary robot controller
2017 5th International Conference on Electrical Engineering - Boumerdes (ICEE-B), 2017
Development and simulation of artificial biological models require a considerable computational c... more Development and simulation of artificial biological models require a considerable computational complexity which can be surmounted using Modern Graphic Processing Units (GPUs), since these devices adopt parallel multi-core architecture which favors the inherently data-parallel nature. In this paper, we present an extended version of our hybrid model presented in [1], [2], based on a high tuning CUDA parameter (Compute Unified Device Architecture) for the parallel evolutionary strategy (ES) and a recurrent neural network (RNN), called PEvoRNN. This proposal takes the advantages of the GPU memory hierarchy effective use, the best parallelism management by investigating tuning CUDA parameters and the optimal GPU-based coding, to generate optimal trajectories of a humanoid robot using GPU Accelerator at multiple levels. PEvoRNN represents a controller which monitors the movement and the evolution of a 3D humanoid robot simulated on the open dynamic engine simulator (ODE). Moreover, this...
Dynamic coalitional matching game approach for fair and swift data-gathering in wireless body sensor networks
2017 International Conference on Wireless Networks and Mobile Communications (WINCOM), 2017
Wireless Sensor Networks are deployed in different fields of application to gather data on the mo... more Wireless Sensor Networks are deployed in different fields of application to gather data on the monitored environment. The Wireless Body Sensor Network (WBSN) is a wireless sensor network designed to monitor a human body vital and environment parameters. The design and development of such WBSN systems for health monitoring have been motivated by costly healthcare and propelled by the development of miniature health monitoring devices. This paper presents the architecture design of a preventive health care monitoring system. This architecture is designed for monitoring multiple patients in a hospital. It is based on a set of mobile data collectors and static sensors for analysis of various patient's parameters. The data collectors need to cooperate together in order to gather the data from the sensor nodes. The point of this paper is how to dynamically and effectively appoint and deploy several data collectors in the hospital to gather the measured data in minimal time. We formula...
A Predator-Prey Scenario in a Virtual Ecosystem
European Conference on Artificial Life, Jul 1, 2015
One of the main topics in artificial life is the design of systems that exhibit some characterist... more One of the main topics in artificial life is the design of systems that exhibit some characteristics of living organisms. Among the great variety of biological systems that inspire and guide these researches and according to Bedau et al. (2003), three broad areas can be identified depending on their basic elements: (a) At the microscopic scale, chemical, cellular and tissular systems; Wet ALife synthesizes living systems out of biochemical substances, (b) At the mesoscopic scale, organismal and architecture systems; or the Soft ALife that uses simulations or other purely digital constructions that exhibit lifelike behavior, (c) At the macroscopic scale, collective and societal systems. In our model we try to blend at least (b) and (c) in the same simulation. In this abstract, we propose architecture to simulate a virtual ecosystem and present extended results. This ecosystem is populated with 3D artificial creatures that have to use a simple predator-prey scenario. Artificial behaviors are developed in order to control artificial creatures. The artificial creatures living in the ecosystem are
Nowadays, morphogenetic engineering (ME) [1] is inspired by biological systems (embryogenesis) to... more Nowadays, morphogenetic engineering (ME) [1] is inspired by biological systems (embryogenesis) to export their self-formation capabilities to engineered autonomous systems. As cells are intelligent by nature, researchers of ME are trying to recreate this intelligence in artificial systems, so that these cells know how and when to act in order to accomplish a specific function (e.g. Build an organism).
This paper describes a bacterial system that reproduces a population of bacteria that behave by s... more This paper describes a bacterial system that reproduces a population of bacteria that behave by simulating the internal reactions of each bacterial cell. The chemotaxis network of a cell is modulated by a hybrid approach that uses an algebraic model for the receptor clusters activity and an ordinary differential equation for the adaptation dynamics. The experiments are defined in order to simulate bacterial growth in an environment where nutrients are regularly added to it. The results show analysis of the motion obtained by some bacteria and their effects on the population behaviors generated by evolution. This evolution allows bacteria to have the ability to adapt themselves to better growth in the available food existed in its environment and to survive.
Uploads
Papers by NourEddine Djedi