Open Data

Augmented Intelligence

As a whole we have a gluttony of different technologies from common operational pictures, GIS maps, decade old form based applications and new comers like social media monitoring. We as an industry have not embraced technology and the lifesaving benefits that comes from well implemented solutions. Augmented intelligence, systems that enhance human capabilities can make the operator in the seat as smart as possible. Streamlining efforts to save lives and ease suffering.

 

Utopia Compression Corporation (UC) and the Kant Consulting Group has been pursuing a disaster rescue mission concept comprising of providing augmented intelligence to disaster responders. Very succinctly, the constituent components pertain to managing and predicting events in a wide scale, geographically distributed disaster in evolution (an earthquake, a hurricane, a tsunami, a terrorist attack…). Specifically, the components consist of:

 

1 – Dynamic (spacio-temporal) evolution of disastrous events – sporadic deaths, casualties; variety of destructions; power, water, communication … outage; destruction and disruption of traffic networks… Such an evolution of events requires in tandem an adaptive, soft and dynamic optimization methodology for resource matching and allocation in order to optimally save lives, attend to casualties and abort further destructions and disruptions.

 

2 – Probabilistic based optimization is a necessity in view of the inherent uncertainties pertaining to the (spacio-temporal) evolution of events. The nature of uncertainties relate to at least two computational fronts: (i) looking ahead, predicting the evolution and the spread of the disastrous events; and (ii) modeling the current standing of the losses, casualties, destructions, etc.

 

3 – Distributed (spacio-temporal) nature of the disastrous events in evolution also require in tandem a distributed computational methodology that optimally handles a spectrum of coarse/wide-area to fine/locally based resource matching and allocation.

 

4 – Hierarchical nature of the international, federal and local agencies, non-governmental services and other institutions involved in the rescue operation also requires a hierarchical computational model that: (i) minimizes the discord between different operational legacies residing within each institutions; (ii) generates the appropriate data abstraction vs. data instantiations/penetration pertaining to coarsity of data required by the institutions within the hierarchy; (iii) provides a dynamic bottom-up and top-down flow of information and executional instructions as the events and concomitant rescue missions unfold.

 

5 – Situated Cognitive Computational Agents residing at the nodes of the hierarchy conduct the operations of the aforementioned dynamic optimization. These cognitive agents consist of classes of expert systems collaborating together and coordinating their actions laterally at each level and transversally across levels of the hierarchy. Furthermore, these expert agents are endowed with learning capabilities to improve their executive/decision making and performative/execution tasks and share their lessons learnt with each other – in that sense there is a notion of knowledge transfer across the entire network membrane.

 

A – Research and develop a software system that will be infield operational when a disaster occurs and evolves.

 

B – Develop a simulator that incorporates the dynamic optimization concept for the purpose of training various echelons of decision makers and operational personnel involved in rescue mission operations.

 

C – Research and develop a Case Base Inference Engine that generates homogenous clusters of past instances of disasters over which species of probabilistic inference (such as multiple inheritance and abduction reasoning) can be conducted. Such an engine will have multiple usage including: (a) exploiting distributed (network) statistics and knowledge of past network activities to predict and possibly abort mal intentions and catastrophic outcomes; (b) lessons learnt so as to be applied for future disasters with improved performance.

Tags

Voting

1 vote
1 up votes
0 down votes
Active
Idea No. 61