We are pleased to introduce Galina Rogova, Ph.D. and Drs. Christopher Landauer, Kirstie Bellman, and Phyllis Nelson. Remember to register to attend one or both tutorials today!
Tutorial 1: Higher Level Fusion and decision making for crisis management
Critical situations can be caused by natural and/or man-made activities and usually result in dangerous social, political,economic, and large-scale environmental events as well as military operations. It is essential to monitor, recognize, and make sense of these activities as early as possible in order to either prevent a crisis or mitigate its outcome. Recognizing crises and effectively responding to them requires systematical gathering, analyzing, and fusing of a large amount of data and information. It is important to translate these data and information into knowledge that is a purpose of higher level fusion (situation and impact assessment). The higher level fusion processes infer, approximate and predict the critical characteristics of the environment in relation to specific goals, capabilities and policies of the decision makers. They utilize fused data about objects of interest, intelligence information, dynamic databases, maps, as well as expert knowledge and opinions for context processing to produce a coherent composite picture of the current and predicted situations. The dynamic current and predicted situational pictures are built by analyzing spatial and temporal relations of the situational entities considered at different levels of granularity and their behavior within the overall situational context. The process of building situational pictures is complicated by the fact that data and information obtained from observations, reports as well as information produced by human and automatic processes are heterogeneous and often unreliable, of low fidelity, contradictory, and redundant. Threat can be unknown, unconventional or even unimaginable. Thus the higher level fusion processing has to be adaptive to resource and time constraints, new and uncertain environments and reactive to heterogeneous inputs of variable quality. An important component of situation and threat assessment is detection of suspicious events based on abnormal characteristics and behavior of situational items along with discovery of underlying causes of such characteristics and behavior.
The tutorial will discuss the role of higher level fusion in designing crisis management systems, the architecture and processes required for building dynamic current and future situational pictures, challenges of, and computational approaches to designing such processes. Specific scenarios in the domains of homeland security and natural disasters will be also considered to show how the processes discussed can be used for the solution of real world problems.
Tutorial 2: Self-Modeling for Adaptive Situation Awareness
This is a tutorial about how to build systems with enough self-information to decide when and how to adapt to new situations, and how to construct, analyze, and communicate models of their operational environments with their human handlers. Our goal is to make autonomous computing systems that can be trusted to be appropriately situation aware, so they can act as our information partners for complex tasks or in complex environments, including hazardous, distributed, remote, and / or incompletely knowable settings.
The purpose of this tutorial is to show how to build systems that can make and assess their own models (and also evaluate and improve the preliminary models we may provide them) of their operational environment, of their history of interaction with that environment, and of their own behavior and internal decision processes. These systems will explore their environment, using active experimentation to assess hypotheses and adjust their models.
If we are going to build this kind of information partners, we will expect them to communicate among themselves and with us. We will expect them at least to interpret models suggested by us and communicate to us the models that they construct. That means that we need mechanisms of mutual model interpretation (we are specifically not trying to reach mutual understanding at this stage of development). To that end, we draw principles from theoretical biology and show how to use them in our computational processes. We use the Wrapping integration infrastructure that implements this style of reflective computing, with all computational resources implemented as limited-scope functions, explicit descriptions of all of these functions and when it is appropriate to use them, and powerful Knowledge-Based integration support processes, all of which are themselves computational resources with explicit descriptions.
We have shown that the Wrapping approach is ideal for adaptive and autonomous systems, including Self-Modeling Systems, in many previous papers, and in full day tutorials in the last two SASO conferences (see the references). We have observed that many techniques for inference of information from data have been developed, and we believe that they are ready to be used by autonomous embedded systems to help us help them behave appropriately in complex environments. We describe these methods and the kinds of models they can produce, what assessment methods they require, and how to implement them within our computational system. We will present and discuss examples in developing situation awareness capabilities using a testbed for embedded real-time systems, called CARS (Computational Architectures for Reflective Systems).