INCOSE President-Elect, USA.
Title and abstract coming soon.
Olivier L. de Weck
Senior Vice President for Technology Planning and Roadmapping, Airbus, France.
Title and abstract coming soon.
James M. Tien
College of Engineering, University of Miami, USA.
Internet of Things, Real-Time Decision Making and Artificial Intelligence.
In several earlier papers, the author defined and detailed the concept of a servgood, which can be thought of as a physical good or product enveloped by a services-oriented layer that makes the good smarter or more adaptable and customizable for a particular use. Adding another layer of physical sensors could then enhance its smartness and intelligence, especially if it were to be connected with other servgoods – thus, constituting an Internet of Things (IoT) or servgoods. More importantly, real-time decision making (RTDM) is central to the Internet of Things; it is about decision informatics and embraces the advanced technologies of sensing (i.e., Big Data), processing (i.e., real-time analytics), reacting (i.e., real-time decision-making), and learning (i.e., deep learning). Indeed, RTDM is becoming an integral aspect of IoT and artificial intelligence (AI), including its improving abilities at voice and video recognition, speech and predictive synthesis, and language and social-media understanding. These three key and mutually supportive technologies – IoT, RTDM, and AI – are considered herein, including their progress to date.
Systems Engineering and Architecting Director within the Thales Technical Directorate, France.
Investigating in SoS Taxonomies to Improve Systems Engineering.
This talk provides an overview of the development of the ISO/IEC/IEEE 21841 Systems and software engineering — Taxonomies of systems of systems. This Standard aims at reflecting the latest thoughts on the way to analysing systems of systems, with their characteristics, development strategies and in-services operations. Outline of this presentation is:
- Taxonomies for systems of systems,
- Examples covering the proposed taxonomies,
- Systems Engineering approaches covering the proposed taxonomies,
- Why the systems of Systems approaches should concern each Systems Engineer,
Principal senior scientist in the MITRE Corporation Center for the MITRE Systems Enginering Technical Center and the Capability Action Team leader for Systems of Systems (SoS), USA.
Systems of Systems: Why Model Based Approaches to System of Systems Engineering?
Increasingly systems engineering is looking towards digitalization and model based approaches to address the challenges we face as our systems become more complex and we are seeking ways to improve the speed and accuracy of system development and delivery. Systems of systems (SoS) poses specific challenges based on the driving characteristics of systems of systems in today’s networked, interconnected world. The partnership between SoS engineering and model-based engineering enhances the value modeling brings by not only affording a rigorous approach to specifying complex SoS architectures and by providing a communications tool for integrating across the multiple domains often brought together in an SoS, but by also by providing a computational base for complex engineering analyses. This presentation looks at the challenges facing systems of systems engineering and how model based approaches are demonstrating their value in tackling to these challenges, highlighting the idea of heterogeneous toolchains, themselves SoS, as a promising strategy to enhance SoSE by bringing a heterogeneous set of analysis methods and tools coordinated with common data from the SoS architecture models to flexibly address SoSE challenges.
Director of Heudiasyc Lab., joint research unit formed by Université de technologie de Compiègne and CNRS, Compiègne, France.
Autonomous cars navigation: from standalone to cooperative systems.
The level of autonomy of driverless cars depends on the complexity of the navigation tasks and on the complexity of the environment, the ultimate goal being to ride on roads open to public traffic. For a long time, the main paradigm was to develop autonomous cars able to drive in a standalone mode by using information acquired by on-board sensors only. With the recent technological advances, some key functions rely on information coming from external sources that have to be fully mastered from a system point of view with measures of performance. The talk will first address this issue and will indicate the main drawbacks of such an approach. For some years now, cooperative systems are seen as a mean to improve the quality of the information needed by navigation systems. This can help to reduce the number and the cost of sensors embedded on cars. Several examples like lane changing on highways or roundabout crossing will be presented to illustrate this concept. This new cooperative paradigm raises many new questions, like the integrity of the information when there exists exchange cycles between the agents that are cooperating. Some cooperative localization problems studied at the Heudiasyc Lab will be outlined. Solutions currently under study will be presented with experimental results.
Digital Manufacturing Director, Plastic Omnium Auto Inergy, France.
Industrial Process Monitoring: from fault analysis to prediction.
Since more than 10 years Plastic Omnium Auto Inergy Division is collecting a lot of process data. It gives us the possibility to automatically control that the parts are produced within the tolerances and also to fill the traceability data base.
To move to the next step which is the creation of model for prediction, we are forced to break the “data silos” and integrate different systems like PES (plant execution system), Quality Data Acquisition System, and Process Data Acquisition System. To be able to predict a rejected part or a downtime, and correct the process before it appears, we need to catch all the events of the shop floor to sort stable mode from transitional modes for applying adequate models.
University of Reading, UK. School of Mathematical, Physical and Computational Sciences.
Framework Architecture for Multi-Modal Sensing and Situation Assessment of Human Gait Dynamics to Support Mobile Gait Rehabilitation.
Sensible (semi)autonomous and pro-active interaction of an assistive robot with its environment has to rely on self-awareness of emerging situations as a vital capability of such a cognitive system. In this keynote we examine the essential system-of-systems components for cooperation support in human-robot co-working or cooperative/empathic dialogue in companion/care robotics.
We identify the fundamental capabilities underpinning the above essential properties of a robotic co-worker/companion which are safety monitoring and control, and, seamless and coherent initiative- taking including emergency safety control, and, timely spoken/textual/physical interventions to support shared human-robot goals.
We explore the requisite functional and interactive layers of such a system and present a framework architecture for situation assessment and control. This notably includes a Human Sensory System, a Situation Assessment Blackboard, an FPGA-enabled Real-time Safety Controller and an Analytics Framework. We present some results and observations as collected from validation of the proposed framework architecture in two Demonstrator domains with a focus on Human Gait Rehabilitation, and, Companion Robotics.