Information fusion for threat analysis in the domain of ground situation awareness

Project summary

Superior situation awareness and decision superiority are important concepts in tomorrows network based defense. To achieve high-level situation awareness, the military decision-makers have to be able to project the current situation into the future, also known as impact assessment. In order for this projection to be sound, it demands processing of all available data. Since military decision-makers often are overwhelmed by the amount and complexity of data, they need decision support systems to help with the computationally hard inference-making task.

This project addresses the high-level information fusion problem of threat analysis, which mainly consists of threat evaluation and weapons allocation. Threat evaluation is an area with a quite small amount of publicly available research results. Weapons allocation has been studied more thoroughly, especially within the field of operations research. However, initial work on threat evaluation shows that in order to compare different threat evaluation algorithms, the following weapons allocation process is of importance.

Research question

To achieve situation awareness and to make the right decisions are anything but trivial tasks for the human decision-maker. Military situations are complex in nature and the available information will always be uncertain to a high degree. Often the objective is to predict the state of entities or relations between entities and the environment. To infer these states and relations is hard, since many possible hypotheses regarding the future may be consistent with given data and information. Moreover, since new data and information arrive in realtime, new hypotheses must continuously be formed and tested.

Hence, there is a need for dynamic decision support systems, which, in real-time, help the decision-maker with acquiring important information from various sources, combine the uncertain information pieces, form plausible hypotheses, make inferences regarding the present and future situation and to create a situation picture that the decision-maker can use to achieve improved situation awareness and to make the right decisions.

Relevance to information fusion

In information fusion, one tries to combine data from multiple sources in order to make inferences that may not be possible to do from a single source alone. Low-level fusion typically consists of fusion of multi-sensor data to determine characteristics of an entity. High-level information fusion include situation assessment, i.e., automated reasoning to refine our estimate of a situation, and impact assessment, i.e., projection of the current situation into the future to define alternative hypotheses regarding possible threats or future conditions. Low-level fusion has been researched for a long time and is quite well understood, but when it comes to high-level information fusion, the research is more immature, with numerous prototypes but only a few operational systems.

Threat evaluation is a part of threat analysis, which in an information fusion context is a central part of impact assessment in the well-known Joint Directories of Laboratories (JDL) data fusion model. Even though threat evaluation obviously is important, few papers have been written on the topic, especially when it comes to systems for automatic or semi-automatic threat evaluation. Threat evaluation is a prerequisite for weapons allocation, a process in which the decision-maker decides on which weapon system that should be assigned to a certain target. Threat evaluation can also be used to support intelligent sensor management, by allocating more sensor resources to targets with high threat values.

Approach

In our work in threat analysis, we started out with a literature study on what parameters that have been suggested for threat evaluation, and what kind of algorithms that have been suggested. From this study, we developed a precise description of the threat evaluation problem. The study also resulted in implementations of two threat evaluation systems, based on a Bayesian network and fuzzy inference rules, respectively. Initial comparisons between the two approaches have been made, both on a theoretical and empirical basis. However, we have concluded that in order to make better comparisons, implementation of a weapons allocation system is needed. Study and implementation of the weapons allocation process will be the next step in our work.

Another idea is to make a hybrid of the two approaches to threat evaluation implemented so far, i.e., Bayesian networks and fuzzy logic. One way to achieve such a hybrid is to create an ensemble of the two approaches, while another is to try to incorporate the use of fuzzy sets in the Bayesian network approach. Reasons for such an approach are to make the construction of the conditional probability tables easier, as well as making the output from the Bayesian network smoother.

Members

Göran Falkman, PhD (Project Leader).

Former members

Fredrik Johansson (now at the Swedish Defence Research Agency).

Publications

Johansson, F. (2010) Evaluating the performance of TEWA systems. Doctoral Dissertation. Örebro Studies in Technology, 40. ISBN 978-91-7668-761-1.  http://oru.diva-portal.org/smash/record.jsf?pid=diva2:381336&searchId=null

Johansson, F. and Falkman, G. (2010) A Suite of Metaheuristic Algorithms for Static Weapon-Target Allocation. In Arabnia, H.R., Hashemi, R.R., and Solo, A.M. (Eds.) GEM 2010. Proceedings of the 2010 International Conference on Genetic and Evolutionary Methods, July 12–15, 2010, Las Vegas, USA, pp 132–138. CSREA Press.

Johansson, F. and Falkman, G. (2010) Real-time Allocation of Defensive Resources To Rockets, Artillery, and Mortars. In Proceedings of the 13th International Conference on Information Fusion (FUSION 2010), 26–29 July 2010, Edinburgh, UK.

Johansson, F. and Falkman, G. (2010) SWARD: System for Weapon Allocation Research & Development. In Proceedings of the 13th International Conference on Information Fusion (FUSION 2010), 26–29 July 2010, Edinburgh, UK.

Johansson, F. and Falkman, G. (2009) Performance evaluation of TEWA systems for improved decision support. Accepted for publication and presentation at The 6th International Conference on Modeling Decisions for Artificial Intelligence (MDAI 2009), November 30–December 2, 2009, Awaji Island, Japan. Springer-Verlag.

Johansson, F. and Falkman, G. (2009) A testbed based on survivability for comparing threat evaluation algorithms. In Mott, S., Buford, J.F., Jakobson, G., and Mendenhall, M.J. (Eds.) Intelligent Sensing, Situation Management, Impact Assessment, and Cyber-Sensing. Proceedings of SPIE Defense, Security, and Sensing 2009, 13–17 April 2009, Orlando, FL, USA. SPIE Volume 7352, 73520C. DOI: 10.1117/12.816819.

Johansson, F. and Falkman, G. (2008) A Survivability-based Testbed for Comparing Threat Evaluation Algorithms. Paper presented at SWIFT 2008 – Skövde Workshop on Information Fusion Topics, 4–6 November, 2008.

Johansson, F. and Falkman, G. (2008) A comparison between two approaches to threat evaluation in an air defense scenario. In Torra, V. and Narukawa, Y. (Eds.) Modeling Decisions for Artificial Intelligence. Proceedings of the 5th International Conference, MDAI 2008, Sabadell, Spain, October 30–31, 2008, LNAI 5285, pp 110–121. Springer-Verlag.

Johansson, F. and Falkman, G. (2008) A Bayesian network approach to threat evaluation with application to an air defense scenario. In Proceedings of the 11th IEEE International Conference on Information Fusion (FUSION 2008), Cologne, Gemany, June 30–July 3, 2008. IEEE.

Niklasson, L., Riveiro, M., Johansson, F., Dahlbom, A., Falkman, G., Ziemke, T., Brax, C., Kronhamn, T., Smedberg, M., Warston, H., and Gustavsson, P.M. (2008) Extending the Scope of Situation Analysis. In Proceedings of the 11th IEEE International Conference on Information Fusion (FUSION 2008), Cologne, Gemany, June 30–July 3, 2008, pp 454–461. IEEE.

Riveiro, M., Johansson, F., Falkman, G., and Ziemke, T. (2008) Supporting Maritime Situation Awareness Using Self Organizing Maps and Gaussian Mixture Models. In Holst, A., Kreuger, P., and Funk, P. (Eds.) Tenth Scandinavian Conference on Artificial Intelligence. Proceedings of SCAI 2008. Frontiers in Artificial Intelligence and Applications 173, pp 84–91. IOS Press.

Johansson, F. and Falkman, G. (2007) Detection of vessel anomalies – a Bayesian network approach. In Proceedings of the 3rd IEEE International Conference on Intelligent Sensors, Sensor Networks and Information (ISSNIP 2007), Melbourne, Australia, 3–6 Dec 2007, pp 395–400. IEEE.

Niklasson, L., Riveiro, M., Johansson, F., Dahlbom, A., Falkman, G., Ziemke, T., Brax, C., Kronhamn, T., Smedberg, M., Warston, H., and Gustavsson, P.M. (2007) A Unified Situation Analysis Model for Human and Machine Situation Awareness. In Koschke, R., Herzog, O., Rödiger, K.-H., and Ronthaler, M. (Eds.) Trends, Solutions, Applications. Proceedings of SDF 2007. LNI P-109, pp 105–110. Köllen Druck & Verlag.

Johansson, F. and Falkman, G. (2006) Implementation and integration of a Bayesian Network for prediction of tactical intention into a ground target simulator. In Proceedings of the 9th IEEE International Conference on Information Fusion (FUSION 2006), Florence, Italy, July 10–13, 2006. pp 1–7. IEEE.