Simulation-based defense planning
2013-2015, project manager: Johan Schubert


The project develops a multi-criteria simulation-based approach for defense planning, where the concern is the best use of available resources in military ground warfare operations. With this method a decision maker can analyse alternative scenarios, where the concern is the best use of available resources in military operations. As the number of possible combination of parameter values is extremely large we use a data farming approach. The scenario is simulated using the FLAMES simulation framework. We develop a scenario model that represent various military units and contain basic logic for various components of each unit, such as platforms, sensors, cognition models governing the behaviour of different units, etc. We analyse large amounts of output data from the simulations with different values of the input data in order to find the best parameter values. With this methodology it is possible to find which combination of parameter ranges that leads to overall blue success.




Real-time simulation supporting effects-based planning
2008-2012, project manager: Johan Schubert


The project will develop and evaluate simulation based decision support for those planning processes that occur in Effects-based Planning (EBP) of an Effects-Based Approach to Operations (EBAO). The project cannot only focus on EBP however, but must also consider other parts of the EBAO process. It will particularly take into consideration the feed-back to EBP that can be done from Effects-based Assessment (EBA) during plan execution, EBE. This makes possible dynamic-replanning in EBP taking into account the collected information about the operation's status. The aim is to evaluate the plan, among other things to detect its weaknesses, and to understand its consequences. Important conditions for a good planning work are to find and use appropriate indicators and to make good Priority Intelligence Requirements.


Important issues: 1. Will the plan function well for a sufficient number of alternative chain of events? 2. Which resource conflicts can occur? 3. How large is the risk associated with different alternative developments? 4. How can one simulate group behavior, for example as emergent phenomena from a few properties of small groups?


The project works with three complementary directions: 1. Our main task is to develop decision support that supports planning and replanning within EBP using simulations. 2. We develop an analysis tool for morphological and statistical analysis of plans. 3. We model and simulate our own GOP and EBAO process in order to see if it is sufficiently fast. That is important information in replanning within EBP.


The project includes a doctoral student program that is carried out in collaboration with the Royal Institute of Technology's Department of Electronic, Computer, and Software Systems. The project will finance a doctoral student within the research group of Distributed Computer Systems.




Tools for information management and analysis

2010-2012, project managers: Pontus Svenson (2010), Christian Mårtenson (2011-2012)


The project will develop and test concept tools, methods and technology that support intelligence analysis and decision making in the Swedish Armed Forces. The tools are to handle and fuse both structured and unstructured information from human intelligence sources, open sources and from different kinds of sensors. Research questions to be addressed include: How should information be structured and tagged to allow efficient information sharing and retrieval? How should knowledge be represented to ensure source traceability and consistent uncertainty management? How can computer tools improve structured analytical techniques? How should information be visualized and interacted with in order to optimize analysis support?




Situation and threat analysis for BG 2011

2007-2009, project manager: Pontus Svenson


The project develops concept tools for user supported information management, putting together a situation picture and threat analysis aimed at asymmetric warfare and BG 2011 needs. Capability for information fusion is a condition for development of effective and user friendly decision support in future command and control system. Those new environments, antagonists and types of commissions, that the BG may face puts new requirements on the information fusion methods that are used in decision support systems for situation and threat analysis. Important research questions in the project are how information from sensors, people (HUMINT) and prior information will be put together, labeled as to quality and fused, and how computer tools that combine good properties of humans with the computer's (mixed-initiative interaction) can be used in order to create a better situation and threat analysis. How information fusion will be used in order to support naturalistic decision-making and the dynamic OODA-loop are other important questions. The project develops Impactorium, which is an information fusion tool tailored for operations other than war (OOTW).




Highway to security: Interoperability for situation awareness and crisis management (HiTS/ISAC); EU PASR project
2006-2008, project coordinator: Jan Larsson (Saab Systems), FOI project manager: Per Svensson


The vision of HiTS/ISAC is a more secure Europe through prevention of terrorism and organized crime. Superior situation awareness and cross-border interoperability are key enablers, leading to new technical and operational methods to work, train and co-operate across Europe.


The objective of HiTS/ISAC is to enable information analysis and fusion from many different sources, through secure cross-border on-line group cooperation between authorities, in order to detect and provide early warnings for suspicious activities, be it communication between suspected criminals, or anomalous movement of persons, goods or money, etc.


HiTS/ISAC will develop a Problem Solving Environment and demonstrate it in a Virtual Operations Room which can be established anywhere, at any time. Tools and processes will be developed and implemented, and demonstrated using realistic scenarios.




Simulation-based decision support in real-time
2006-2007, project manager: Johan Schubert


The project develops a decision support system for management of riots. The decision support uses fuzzy matching between a current situation and simulated situations with associated riot control strategies. The idea is to find a simulated situation that best matches the current situation and use the best riot control strategy for this situation as decision support in the current case. The learning is performed by studying a number of different situations. For each situation various control strategies are evaluated using simulation. As the number of alternative strategies is very large and it is impossible to study them all, we use machine learning (genetic algorithms) in order to evolutionarily create a smaller amount of new strategies for evaluation on the basis of relatively successful strategies. All strategies are assessed and those that succeed well are saved along with the current situation.



Technology, methodology and demonstration system for information fusion
2004-2006 project manager: Pontus Hörling


The project studies information management and information fusion on higher levels than sensor fusion. Information that is continuously collected from sensors and from humans, as well as information that can be searched for in data bases in order to be associated to that which has been gathered is included in this process. As of 2005 the project has been aimed toward urban scenarios and MOOTW (Military operations Other Than War). We work with identifying important information management sequences that may be important in urban scenarios, and to find information fusion functions that can make these more effective. In urban scenarios we naturally find a large population, making the management of information from humans increasingly important. A lot of information that is handled will contain descriptions of the urban environment's terrain and infrastructure, and the people that live there. Data bases with such information become vital components in the information management.


We have identified two important scenarios for international operations that Swedish troops may force; riots and attempted assignation situations. The project studies which information management that is most important in order to manage such situations, i.e. to anticipate, to monitor and prevent them.




Decision support in international operations
2003-2006 project manager: Göran Neider


The project is aimed towards international operations within the framework of Military Operations Other Than War (MOOTW). The project focuses on the needs of battalion command, primarily on local levels (within one's own area of responsibility AOR). The project studies theories and knowledge regarding collective behavior, the decision-maker's models at the battalion level, and has described the environment of Military Operations Other Than War and its lack of current relevant doctrines. Thus, we have selected an experience-based decision making approach according to Klein's analogy based decision support model in order to replace the lack of doctrines around the collective 's behavior, for example in riots.


As computer science method, we have selected an AI-method called Case Based Reasoning (CBR) in order to store and reuse experiences of experienced cases.  We will develop a CBR-prototype in order to try the methodology in this application, where for example temporal aspects and event chains seem to be unresearched within CBR. The cases description relies on indicators and countermeasures. These indicators are caught from the observations that are gathered and classified in a structure. Those indicator descriptions we found often come from international levels and must be expanded with those local perspectives that are requested. The riot in Kosovo in March 2004 gives us both reasons and information to the project.




Information fusion in the new defense's command and control systems
2001-2003 project manager: Per Svensson


The project studies and develops methods for co-exploitation and combination of reconnaissance and intelligence information on tactical and operational levels and gives proposals on how such methods evolutionary will be integrated in a distributed, network based future command and control system. This is done in collaboration with the Swedish Armed Forces, the Swedish National Defence College, the Swedish Defence Materiel Administration and the defense industry.


A first hand objective is that during the years 2001-2003 develop one simple demonstrator for tactic information fusion for operational command and control, with accentuation at land warfare. The demonstrator has two central components: an analysis system pursues information fusion and a supporting simulator for generation of sensor- and intelligence reports.


The project builds on knowledge and research results from the previous projects Tactical Information Fusion, Sensor Management and System Architecture. We assess that a project with this scope after three years should demonstrate examples on relevant and operationally interesting information fusion methods. On the other hand, many years' work remains national and internationally before an effective decision support with sought after broad functionality, based on information fusion, is available in the command and control system.




Strategic research core information fusion
2001-2003 project manager: Per Svensson


The research project aims to develop methodology that will make possible information superiority in the Swedish Armed Force's future command and control system. The intention is to put the basis to practical applications in the command and control system in the time perspective 2005-2010 and beyond.


The project means knowledge build-up and method development within the areas management of uncertainty, control of information collection and fusion architectures, with applications within situation assessment, threat assessment and resource control processes in an information fusion system for command and control.


Results of the research are expected to become increased ability that with well-known information quality to make propositions on position, identity, scope, activity, threat potential and course of action alternatives for complex, movable set objects (primarily military units). The propositions are based on methodology for force aggregation, correlation, activity identification etc. in a distributed system of fusion nodes. Input data to the information fusion process is output data delivered from sensor fusion processes and environmental and other prior information stored in databases.


The project includes a doctoral student program that is carried out in collaboration with the Royal Institute of Technology's Department of Numerical Analysis and Computer Science. The project will finance a doctoral student within each research group of Theoretical Computer Science and Centre for Autonomous Systems.