sp03-edited

Information about sp03-edited

Published on August 5, 2014

Author: arpithaydv

Source: authorstream.com

Content

THE ROLE OF ARTIFICIAL INTELLIGENCE TECHNOLOGIES IN CRISIS RESPONSE: THE ROLE OF ARTIFICIAL INTELLIGENCE TECHNOLOGIES IN CRISIS RESPONSE IN PRESENTED BY: Sushma Praveen Kumar M Arpitha A PowerPoint Presentation: No of events: 431 No of people killed: 143,039 Average killed per year: 4,614 No of people affected: 1,521,726,127 Average affected per year: 49,087,940 Economic Damage (US$ X 1,000): 48,063,830 Economic Damage per year (US$ X 1,000): 1,550,446 India - Disaster Statistics Data related to human and economic losses from disasters that have occurred between 1980 and 2010.  Natural Disasters from 1980 - 2010 WHAT ARE WE DOING HERE ?: WHAT ARE WE DOING HERE ? WHAT IS AI ? HISTORY OF AI FIELDS WHERE AI IS USED CRISIS RESPONSE AI TECHNOLOGIES IN CRISIS RESPONSE ROBOTICS ONTOLOGY AND SEMANTIC WEB MULTI-AGENT SYSTEMS FUTURE OF AI CONCLUSIONS REFERENCES WHAT IS ARTIFICIAL INTELLIGENCE ?: W HAT IS A RTIFICIAL I NTELLIGENCE ? John McCarthy, who coined the term ”Artificial Intelligence”, defines AI as the science and engineering of making INTELLIGENT MACHINES, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence. This definition is appropriate as it contains three main features of AI. Emulation of intelligence with the computer Creation of mechanical articrafts AI does not always restrict itself to human intelligence HISTORY OF AI: H ISTORY OF AI Evidence of AI folklore can be traced back to ancient Egypt. With the development of the electronic computer in 1941 The technology finally became available to create machine intelligence The term AI was coined in 1956, at the Dartmouth conference, From its birth 4 decades ago, there have been a variety of AI programs, and they have impacted other technological advancements . FIELDS WHERE AI IS USED … : F IELDS W HERE AI IS U SED … Finance Heavy Industry Toys and Games Robotics Cognitive Data Mining Biologically inspired computing Online and telephone customer service Hospitals and Medicine Aviation CRISIS RESPONSE: CRISIS RESPONSE It is the immediate protection of property and life during the Crisis events to reduce deaths, losses and injuries. Includes actions taken before the actual crisis event(e.g., hurricane warning is received), in response to the immediate impact of the crisis and as sustained effort during the course of the crisis. Depending upon the magnitude and complexity of the crisis , response may be a large scale operation involving many layers of authorities, commercial entities, volunteer organizations, median and the public(Virtual Organization). The success of crisis response largely depends on Gathering information, integrating and making decisions AI TECHNOLOGIES IN CRISIS RESPONSE: AI T ECHNOLOGIES IN C RISIS R ESPONSE AI TECHNOLOGY Tries to improve the efficiency of the management process during crisis response via ROBOTICS: It helps in gathering information from distributed resources ONTOLOGY AND SEMANTIC WEB: are adopted to solve data integrating problems MULTI-AGENT SYSTEM: provides the decisive solution to all problems related to interaction and co-ordination of response teams 1. ROBOTICS: 1. ROBOTICS The diverse structure of crisis area, rescuers’ safety and the necessity of quick and reliable examination of targeted regions forces agencies to use Multi-Robot solutions in Urban search and rescue Robots can provide variety of functions in the crisis context Area exploration Mapping using Radio Frequency Identification (RFID) tags Expediting the search for victims They can navigate through tightly confined spaces which people cannot access within the crisis site. ROBOTICS: ROBOTICS During the search for victims, they can deposit radio transmitters to be able to communicate with victims. They use small probes to check victims heart rate and body temperature. They even help to supply heat source and small amounts of food and medication to sustain the survivors in crisis site. One of the first uses of robots in search and rescue operation was during the 9/11, in NY using Micro-VGTV (Variable Geometry Tracked Vehicle) by INUKTUN robots. ONTOLOGY AND SEMANTIC WEB: ONTOLOGY AND SEMANTIC WEB Ontology: A description of the concepts and relationships that can exist for an agent or a community of agents they are used in integrating heterogeneous information sources There is a need to maintain a common core of information that can be shared about the crisis as we have huge diverse set of information sources and information users This common core set of information can be represented by ontology Different types of ontologies: Ontologies for over all crisis response Robot ontology for urban search and rescue Decision making ontologies Steps in Crisis response Ontology: Steps in Crisis response Ontology According to W3C definition, crisis response ontology must describe the following critical steps PowerPoint Presentation: 1.Sharing of data describing response and resource characteristics when crisis is anticipated 2.Gathering data on its scope and emerging effects 3,Gathering data on its outages and missing links 4.Sharing relief requests 5.Gathering and integrating data from all responders to build a common map of the situation 6.Characterizing problem states as chaotic, complex or manageable 10.Passing of all data gathered in the disaster to the appropriate authority after the crisis passes, updating databases. 8.Calling for expert review 7.Rapidly deploying compatible information and communication systems to local authorities 9.Comparing predictive to measured effects of crisis Semantic web services: Semantic web services Based on the deployed ontology semantic web services are prototyped It is used to provide data to crisis actors Emergency management application(EMA) is an example of deployed semantic web services The EMA has been designed to enable data and functionalities provided by existing legacy systems to be exposed as web services This system involves number of ontologies to gather information from different sources Based on the embedded ontologies, emergency officer can retrieve, process, display and interact with relevant information quickly and accurately PowerPoint Presentation: ONTOLOGY IN SEMANTIC WEB MULTI-AGENT SYSTEMS (MAS): MULTI-AGENT SYSTEMS (MAS) Crisis response problems cannot be solved by a single responder, a heterogeneous team is needed This team needs planning and coordination capabilities to complete the mission successfully MAS is a system composed of multiple interacting software agents. MAS can be used to solve problems which are difficult for an individual agent to solve such as crisis response etc. The original idea comes out from agent characteristics Autonomy Local view of environment Capability of learning, palnning,coordination and decision making MAS: MAS If we can imagine that an agent can represent a crisis responder We can build a crisis response system based on agents’ interaction and coordination. Agents can help crisis responders in doing their planning, and coordination task They can even replace a human in information gathering and specific decision making tasks Agent-based modeling and simulation are used for responders training and system testing E.g. of MAS DrillSim, DEFACTO, WIPER PowerPoint Presentation: Criteria Objectives Architecture Application Domain Features DrillSim Test-bed for IT Solutions Multi-Agent simulation ( Fire) Floor Evacuation Simulation Micro level- every agent simulates and interacts with a real person. -Agent Learning using recurrent artificial neural networks. DEFACTO - Address limitations of: i) human situational awareness ii) the agent team’s rigid interaction strategies . -Software proxy architecture ( Machinetta ) and 3D visualization system . -Fire RoboCop rescue - Omnipresent viewer -Proxy Framework -Flexible interaction -Adjustable autonomy WIPER - Evacuate potential plans of action using a series of GIS enabled Agent-Based simulations. -Web service and service oriented Architecture + Multi-Agent System design . - Building large-scale ad hoc sensor network based on the existing cellular telephone network. -The Simulation and Prediction System will initially be used to predict simple movement and traffic patterns. - Data stream will be monitored by an anomaly detection algorithm flagging potential crisis events. -Agent-based simulations will attempt to predict the course of events and suggest potential mitigation plans. FUTURE OF AI : F UTURE OF AI Since the start of the 21st century, there's no question that mankind has made tremendous strides into the field of  AI While modern robots can now replicate the movements actions of humans, the next challenge lies in teaching robots to think for themselves and react to changing conditions.  The field of artificial intelligence promises to give machines the ability to think analytically, using concepts and advances in computer science, robotics and mathematics .   While scientists have yet to realize the full potential of artificial intelligence, this technology will likely have far-reaching effects on human life in the years to come CONCLUSION   : CONCLUSION    AI techniques offer potentially powerful tools for the development of crisis response and management systems The technologies of robotics, ontology and semantic web, and multi-agent systems can be useful to solve the problems Robotics can be useful in urban search and rescue to bypass challenges faced by rescue workers and to expedite the search operations of crisis response Ontologies concepts and semantic web offers as many advantages in solving systems integration and interoperability problems . Ontologies are also used to collect, integrate and reason on heterogeneous information sources CONCLUSION: CONCLUSION Multi-agent systems features and methodologies is the core of crisis response systems. Crisis response systems take advantage of coordination and planning capabilities of multi-agent systems to handle response teams’ coordination and interaction problems . PowerPoint Presentation: QUESTIONS ??? PowerPoint Presentation: THANK YOU 

Related presentations


Other presentations created by arpithaydv

COGNITIVE MODELLING PPT
05. 08. 2014
0 views

COGNITIVE MODELLING PPT

AP(003)
05. 08. 2014
0 views

AP(003)