O. Azouaoui

• PhD
• Directeur de recherche at Centre de Développement des Technologies Avancées

Future autonomous driving and next-generation vehicle control require further advances in vehicle dynamic state estimation. This paper describes a new localization approach based on the development of a personalized dynamic model for an autonomous car-like vehicle called “Robucar”. The proposed approach is used within a framework for the developmen...

In this paper, an evolutionary scan-matching approach is proposed to solve an optimization issue in simultaneous localization and mapping (SLAM). A rich literature has been invested in this direction, however, most of the proposed approaches lack fast convergence and simplicity regarding the optimization process, which should directly affect the ac...

This paper deals with the problem of simultaneous localization and mapping (SLAM). Providing both accurate environment’s map and pose estimation is necessary to correctly navigate, which is a key issue for a mobile robot interacting with human beings. It is a line of research that is always active, offering various solutions to this issue. Neverthe...

Robots are now among us and even though they compete with human beings in terms of performance and efficiency, they still fail to meet the challenge of performing a task optimally while providing strict motion safety guarantees. It is therefore necessary that the future generation of robots evolves in this direction. Generally, in robotics state-of...

This paper solves a motion planning problem, where a variant of the classical Rapidly exploring Random Tree (RRT) approach called p-safe RRT is proposed. The exploration of the search space is similar to RRT but a passive motion safety is introduced. The basic principle of this safety level is to guarantee that the system can brake down and stop be...

This paper implements an automated transportation system allowing passengers on a car-like robot to reach their destination safely, intelligently and autonomously. Different components of the system are described such as localization based on Hector SLAM, humans detection based on HOG (Histograms of Oriented Gradient) descriptor and intelligent nav...

The last few years, research in the area of path planning for mobile robots has been focusing on dynamic environments. Most of methods proposed in this topic need to re-plan the remaining path of the robot, every time when new information come in which significantly increase the computation time and make real-time implementation of these methods di...

This paper describes the implementation of a local ICP-SLAM (Iterative Closest Point - Simultaneous Localization and Mapping) to improve the method presented in [1] to become faster. The ICP algorithm is known as a method that requires more computation time when the environment grows leading to poor results for both localization and mapping. Theref...

This paper describes the implementation of a local ICP-SLAM (Iterative Closest Point - Simultaneous Localization and Mapping) to improve the method presented in [1]. The ICP algorithm is known as a method that requires more computation time when the environment grows leading to poor results for both localization and mapping. To overcome this proble...

A necessary condition to perform a fully autonomous driving system in urban environment is to detect object types in real scenes. Visual object recognition is a key solution, but multi-object detection still remain unsolved. In this paper, we present a fast and efficient multi-object detection system built to recognize, at the same time, pedestrian...

In this paper, a hierarchical multi-classification approach using support vector machines (SVM) has been proposed for road intersection detection and classification. Our method has two main steps. The first involves the road detection. For this purpose, an edge-based approach has been developed using the bird’s eye view image which is mapped from t...

This paper addresses the problem of planning the motion of a mobile robot with a limited sensory field-of-view in an unknown dynamic environment. In such a situation, the upper-bounded planning time prevents from computing a complete motion to the goal, partial motion planning is in order. Besides the presence of moving obstacles whose future behav...

This paper describes a hierarchical multi-classification approach for road intersection detection and classification using Support Vector Machines. Our approach has been organized in two main parts. The first involves the road detection in the bird's eye view image using an edge-based approach. In this step, we have exploited the concept of vertica...

Scan matching is a popular way of recovering a mobile robot's motion and constitutes the basis of many localization and mapping approaches. The work presented in this paper consists in the implementation of a SLAM method (Simultaneous Localization and Mapping) based on the Iterative Closest Points (ICP) algorithm on the Robucar, a car-like mobile r...

The automatic place recognition problem is one of the key challenges in SLAM approaches for loop closure detection. Most of the appearance-based solutions to this problem share the idea of image feature extraction, memorization, and matching search. The weakness of these solutions is the storage and computational costs which increase drastically wi...

In this paper, we investigate the feasibility of building a visual memory for robotics spatial cognition without saving visual information. The proposed approach captures some properties of primate brain and especially the view cells. Our bio-inspired visual memory for cognitive map building consists of two main parts. One part is visual informatio...

This paper presents an implementation of an intelligent navigation approach on a bi-steerable mobile robot Robucar. This approach is based on Neural Networks (NN) and Fuzzy Logic (FL) paradigms to provide Robucar with capability to acquire the obstacle avoidance, target localization, decision-making and action behaviors after learning and adaptatio...

The use of hybrid intelligent systems (HISs) is necessary to bring the behavior of intelligent autonomous vehicles (IAVs) near the human one in recognition, learning, adaptation, generalization, decision making, and action. First, the necessity of HIS and some navigation approaches based on fuzzy ArtMap neural networks (FAMNNs) are discussed. Indee...

This paper deals with a problem which is basic to mobile robot navigation in dynamic environments: the perception of dynamic objects. This required as a first step the extraction of mobile objects present in the environment tacking into account the effect of the embedded sensor motion. In the proposed algorithm, the laser data are used in the devel...

This paper investigates the use of a modified Dynamic Window Approach (DWA) to solve the collision avoidance problem for a Bi-steerable mobile robot. In this paper, our main interest is to adapt the DWA to a bi-steerable robot Robucar. This latter can move with high speed. It is characterized by its bounded steering angle imposed by its mechanical...

This paper presents an implementation of a Fuzzy Motion Controller (FMC) to endow the mobile robot Robucar with capability to achieve the action behavior allowing smooth motion generation with intelligence in real-time. The robot state space (velocity and distances) is modeled in discrete intervals leading to linguistic variables. The fuzzy motion...

This paper presents a mobile robot self localization method used to determine the position of the mobile robot Robucar. The localization approach is based on using both grids matching method and extended Kalman filter (EKF) method. The grids matching method provides accurate results but requires a large computational time that is why the EKF is int...

Purpose The purpose of this paper is to present an implementation of a soft‐computing (SC) based navigation approach on a bi‐steerable mobile robot, Robucar. This approach must provide Robucar with capability to acquire the obstacle avoidance, target localization, decision‐making and action behaviors after learning and adaptation. This approach use...

This paper investigates the use of dynamic window approach (DWA) to solve the high speed autonomous navigation problem for mobile robots. In this paper, our main interest is to adapt the DWA to a car-like robot Robucar. This latter can moves with high speed. It is characterized also by its bounded steering angle imposed by its mechanical structure...

This paper presents an implementation of a neural-based navigation approach on two types of mobile robots namely the Robucar and Robuter. This approach must provide these robots with capability to acquire the obstacle avoidance and target localization behaviors after learning. Then, this approach uses three Neural Networks (NN) to achieve the desir...

In several complex applications, the use of multiple autonomous robotic systems (ARS) becomes necessary to achieve different tasks, such as foraging and transport of heavy and large objects, with less cost and more efficiency. They have to achieve a high level of flexibility, adaptability and efficiency in real environments. In this paper, a reinfo...

Ce travail traite du problème de la commande de mouvements d'un robot mobile de type voiture à double braquage. Des algorithmes permettant le déplacement du robot d'une position initiale vers une quelconque destination désirée, tout en respectant ses contraintes cinématiques, ont été développés puis implémentés. D'une part, la localisation est fait...

In several complex applications, the use of multiple autonomous robotic systems (ARS) becomes necessary to achieve different tasks such as foraging and transport of heavy and large objects with less cost and more efficiency. They have to achieve a high level of flexibility, adaptability and efficiency in real environments. In this paper, a reinforc...

To ensure more autonomy and intelligence with real-time processing capabilities for the obstacle avoidance behavior of Intelligent Autonomous Vehicles (IAV), the use of Neural Networks (NN) is necessary to bring this behavior near to that of humans in the recognition, learning, adaptation, reasoning and decision-making, and action. In this paper, t...

To ensure more autonomy and intelligence with real-time processing capabilities for the obstacle avoidance behavior of Intelligent Autonomous Vehicles (IAV), the use of soft computing is necessary to bring this behavior near to that of humans in the recognition, learning, adaptation, generalization, reasoning and decision-making, and action. In thi...

Autonomous robotic systems have to achieve a high level of flexibility, adaptability and efficiency in real environments. To achieve this goal, they must particularly have the capability to avoid collisions with obstacles. In the paper, a fuzzy ArtMap neural network based collision avoidance approach for ARS is suggested. Indeed, each robot must le...

Current robotic systems are broadly applied in factory automation and surgery, underwater, space exploration, and also in human daily life in the coming future. First, the necessity of Autonomous Robotic Systems (ARS) in the interaction with dynamic worlds is shown. Second, criteria to satisfy by ARS to be intelligent and reactive are discussed. Th...

The real time resolution for manipulators inverse dynamics is difficult to achieve. To solve this problem, we first start an algorithmic study which leads us to choice the resolution algorithm. A parallelization methodology is then developed using a hierarchical approach which conducts to a decomposition scheme. This methodology is applied to a man...