Amar Ramdane-Cherif

Generic multimedia multimodal agents paradigms and their dynamic reconfiguration at the architectural level

The multimodal fusion for natural human-computer interaction involves complex intelligent architectures which are subject to the unexpected errors and mistakes of users. These architectures should react to events occurring simultaneously, and possibly redundantly, from different input media. In this paper, intelligent agent-based generic architectures for multimedia multimodal dialog protocols are proposed. Global agents are decomposed into their relevant components. Each element is modeled separately. The elementary models are then linked together to obtain the full architecture. The generic components of the application are then monitored by an agent-based expert system which can then perform dynamic changes in reconfiguration, adaptation, and evolution at the architectural level. For validation purposes, the proposed multiagent architectures and their dynamic reconfiguration are applied to practical examples, including a W3C application.

A context-sensitive incremental learning paradigm of an ubiquitous multimodal multimedia computing system

In this paper, we proposed a context-sensitive incremental learning paradigm of an ubiquitous multimodal multimedia computing system. An ubiquitous computing environment supports a busy and mobile users need of being able to work on his task anytime and anywhere he wants. Along with users data (his profile, task, and application registry) the machine-acquired intelligence needs to be transported as well in order that the user could continue working on an intelligent environment. Machine intelligence is acquired through incremental learning. In a context-sensitive environment that has a rich selection of modalities and media for data input and output, an intelligent computing system could determine the I/O devices appropriate for the users setting after considering the users location, the noise level in the environment, and the presence or absence of other people in the vicinity. Every new setting (pre-condition scenario) produces a new I/O devices configuration (post-condition scenario) suited for the setting; each new scenario knowledge gets stored onto knowledge database. Overtime, the machine would have enough knowledge to deal with whatever context scenario that comes up.

A Refined Functional Size Measurement Procedure for Real-Time Embedded Software Requirements Expressed Using the Simulink Model

COSMIC-based functional size measurement (FSM) procedures should all produce the same measurement results, no matter what design decisions are made when modeling the functional requirements of the software to be measured. When the FSM procedure proposed in [14] is applied to different models of the same functional requirements, the measurement results vary. This problem is addressed, and a refined FSM procedure for real-time embedded software is proposed. The revised rules of this refined FSM procedure are based on an analysis of the issues detected in [14], and a solution to the variance issue is offered as a result.

Verifying the Accuracy of Automation Tools for the Measurement of Software with COSMIC -- ISO 19761 Including an AUTOSAR-Based Example and a Case Study

Automating functional size measurement (FSM) is important for organizations needing to measure a large number of projects within a short timeframe, provided, of course, that the results automatically generated are accurate, particularly when such measurement is based on an international measurement standard. A literature review has shown that very little work has been conducted on verifying measurement results produced by FSM automation. This paper presents a verification protocol designed to provide evidence of the accuracy of an automated FSM tool using the COSMIC ISO 19761 measurement standard. Also included are: an example of its use for the verification of an AUTOSAR-based FSM automation prototype tool developed at ESTACA, and a case study on the application of the proposed verification protocol on another prototype tool developed at Renault S.A.

Multi levels semantic architecture for multimodal interaction

This paper presents a semantic architecture for solving multimodal interaction. Our architecture is based on multi agent systems where agents are purely semantic using ontologies and inference system. Multi levels concepts and behavioural models are taken into account to bring a fast high level reasoning on a big amount of percepts and low level actions. We apply this architecture to make a system aware of different situations in a network like tracking object behaviours of the environment. As a proof of concept, we apply our architecture to an assistant robot helping blind or disabled people to cross a road in a virtual reality environment.

A Multi-Agent based Multimodal System Adaptive to the User’s Interaction Context

Communication is an important aspect of human life; it is with communication that helps human beings connect with each other as individuals and as independent groups. In informatics, the very purpose of the existence of computer is information dissemination – to be able to send and receive information. Humans are quite successful in conveying ideas with one another and reacting appropriately because we share the richness of our language, have a common understanding of how things work and have an implicit understanding of everyday situations. When human communicate with human, they comprehend the information that is apparent to the current situation, or context, hence increasing the conversational bandwidth. This ability to convey ideas, however, does not transfer when human interacts with computer. On its own, computers do not understand our language, do not understand how the world works and cannot sense information about the current situation. In a typical impoverished computing set-up where providing computer with information is through the use of mouse, keyboard and screen, the result is we explicitly provide information to computers, producing an effect that is contrary to the promise of transparency and calm technology in Marc Weiser’s vision of ubiquitous computing (Weiser 1991; Weiser 1993; Weiser and Brown 1996). To reverse this, it is imperative that methodologies are developed that will enable computers to have access to context. It is through context-awareness that we can increase the richness of communication in humancomputer interaction, through which we can reap the most likely benefit of more useful computational services. Context (Dey and Abowd 1999; Gwizdka 2000; Dey 2001; Coutaz, Crowley et al. 2005) is a subjective idea and its interpretation is personal. Context evolves and the acquisition of contextual information is essential. However, we believe that the one with the final word on whether the envisioned context is correctly captured/acquired or not is the end user. Current research works indicate that some contextual information are already predefined by their systems from the very beginning – this is correct if the application domain is fixed but is incorrect if we infer that a typical user does different computing tasks in different occasions. With the aim of coming up with more conclusive and inclusive design, we conjure that the contextual information that is important to the user should be left to the judgment of the end

Agents paradigm to solve a complex optimization problem

Agents are computational systems interacting with dynamic, and non-entirely predictable environments. They decide for themselves, on the basis of their individual beliefs, goals, etc., how to respond to the environment. An agent is usually motivated to achieve a feasible objective by means of the collaboration of other agents. For example, lifting a heavy table may be impossible without help. However, it can be done by simply asking others for help. The process of gaining collaboration can take many forms. However, some tasks need more detailed communication to generate an explicit mutually acceptable agreement through negotiation. This paper presents a new method to solve the inverse kinematic problem of a redundant robot using the agents paradigm. This method guarantees rapid convergence of the robot to desired position, with substantially good accuracy. The originality of the proposed approach is its ability to overcome some of the standard problems such as the computation of the inverse or pseudoinverse Jacobian matrix, problem of singularity, redundancy, and considerably increased computational complexity, etc.

Multimodal Fusion, Fission and Virtual Reality Simulation for an Ambient Robotic Intelligence☆

Abstract In this paper, we present an architecture that demonstrates multimodal fusion and fission using semantic agents and web services that interacts with worldwide-web computers, services and users. This solution extracts the meaning of a situation using the semantic memory of agents to manage the interaction process involved. The fusion of values from different sensors produces an event that needs implementation. The fission process suggests a detailed set of actions that are for implementation. Before such actions are implemented by actuators, these actions are first evaluated in a virtual environment which mimics the real-world environment. If no danger arises from such virtual evaluation, then implementation is feasible. Otherwise, there might be a need to add one or more smaller actions to render the action safe and free from danger. Our work presents the following contributions: (i) a design of agent memory and a model of the world environment using a knowledge representation language that is compatible with the current standards, (ii) creation of a pervasive architecture with several scenarios of composition and adaptation, (iii) presentation of how agents and services interact to provide support in a real-world, and (iv) simulation of an event in a virtual environment to assess the feasibility of the events implementation.

A paradigm of a pervasive multimodal multimedia computing system for the visually-impaired users

Incorporating multimodality in a computing system makes computing more accessible to a wide range of users, including those with impairments. This work presents a paradigm of a multimodal multimedia computing system to make informatics accessible to visually-impaired users. The systems infrastructure determines the suitable applications to be used. The users context and user data type are considered in determining the types of applications, media and modalities that are appropriate to use. The system design is pervasive, fault-tolerant and capable of self-adaptation under varying conditions (e.g. missing or defective components). It uses machine learning so that the system would behave in a pre-defined manner given a pre-conceived scenario. Incremental learning is adapted for added machine knowledge acquisition. A simulation of systems behaviour, using a test case scenario, is presented in this paper. This work is our original contribution to an ongoing research to make informatics more accessible to handicapped users.

Exploring cognitive approach through the neural network paradigm: "trajectory planning application"

In recent years, artificial neural networks have received a great deal of attention for their ability to perform nonlinear mappings. In trajectory control of robotic devices, neural networks provide a fast method of autonomously learning the relation between a set of output states and a set of input states. In this paper, we will apply the cognitive approach to solve the problems related to the position controller using the inverse geometrical model. In order to control a robot manipulator to accomplish a task, trajectory planning is required in advance or in real time. The desired trajectory is usually described in Cartesian coordinates and needs to be converted to joint space for the purpose of analyzing and controlling the system behavior. In this paper, we use the memory neural network MNN to solve the optimization problem concerning the inverse of the direct geometrical model of the redundant manipulator subject to some constraints. Our approach offers substantially better accuracy, avoids the computation of the inverse or pseudoinverse Jacobian matrix and do not produce problems such as singularity, redundancy, and considerably increased computational complexity, etc.