Academia.eduAcademia.edu

Figure 6 – uploaded by Adang Ahmad

See full PDFdownloadDownload figure
Table 5.5b. Weather State at TR.

Table 5 5b. Weather State at TR.

Related Figures (21)

Figure 2.1. Human multi-source information fusion system [6].
Figure 2.1. Human multi-source information fusion system [6].
Figure 2.2. Illustration of an agent relation with its environment. In accomplishing the assigned tasks, the agent always senses his/her surrounding environment to get as much as information that can affect his/her activities. The gathered information is processed in his/her brain, combining it with the existing information, inferencing on the fused information, information inferencing fusion, and making the best decision for actions to be done in anticipating the environment dynamics. The relation of an agent with its surrounding environment as well as his/her information processing mechanism is simplv depicted in Figure 2.2.
Figure 2.2. Illustration of an agent relation with its environment. In accomplishing the assigned tasks, the agent always senses his/her surrounding environment to get as much as information that can affect his/her activities. The gathered information is processed in his/her brain, combining it with the existing information, inferencing on the fused information, information inferencing fusion, and making the best decision for actions to be done in anticipating the environment dynamics. The relation of an agent with its surrounding environment as well as his/her information processing mechanism is simplv depicted in Figure 2.2.
Figure 2.3. A multi-agent system architecture [18].
Figure 2.3. A multi-agent system architecture [18].
Figure 2.4. The scheme of information flow in C4ISR-based systems [11]. The main aspect in the C4ISR is information collected by the strategically-distributed sensors in Intelligence, Surveillance, and Reconnaissance (ISR) activities. The speed of information flow is depended on the Communications infrastructures supported by robust Computer systems. Meanwhile, good information product is extremely relied on the Information Fusion System supported by robust information fusion method. The scheme of information flow in C4ISR-based systems is depicted in Figure 2.4. The complete history of the development of the C4ISR framework can be read in [9].
Figure 2.4. The scheme of information flow in C4ISR-based systems [11]. The main aspect in the C4ISR is information collected by the strategically-distributed sensors in Intelligence, Surveillance, and Reconnaissance (ISR) activities. The speed of information flow is depended on the Communications infrastructures supported by robust Computer systems. Meanwhile, good information product is extremely relied on the Information Fusion System supported by robust information fusion method. The scheme of information flow in C4ISR-based systems is depicted in Figure 2.4. The complete history of the development of the C4ISR framework can be read in [9].
Table 1 Illustration of Bayes method with MAP technique. III.2 MSJP Method [5] technique is failed to accomplish its job. Table] illustrates Bayes method with MAP technique. The Maximum Score of the Total Sum of Joint Probabilities (MSJP) inference method is the enhancement of Bayes inference method with MAP after we observe its limitation. The MSJP method is presented in Equation (3.2).
Table 1 Illustration of Bayes method with MAP technique. III.2 MSJP Method [5] technique is failed to accomplish its job. Table] illustrates Bayes method with MAP technique. The Maximum Score of the Total Sum of Joint Probabilities (MSJP) inference method is the enhancement of Bayes inference method with MAP after we observe its limitation. The MSJP method is presented in Equation (3.2).
Table2__ [Illustration of the MSJP method. The MAP decision rule in Bayes method only takes the greatest value of a posteriori conditional probability from the available hypotheses as the best outcome. On the other hand, the MSJP method applies the total sum of joint (a posteriori conditional) probabilities from the available hypotheses and selects the greatest value as the best outcome. The use of P(A;|B,& ...& B;) notation represents fused (joint) probability of all a posteriori probability values obtained from the computation. Table II illustrates the mechanism of the MSJP method.
Table2__ [Illustration of the MSJP method. The MAP decision rule in Bayes method only takes the greatest value of a posteriori conditional probability from the available hypotheses as the best outcome. On the other hand, the MSJP method applies the total sum of joint (a posteriori conditional) probabilities from the available hypotheses and selects the greatest value as the best outcome. The use of P(A;|B,& ...& B;) notation represents fused (joint) probability of all a posteriori probability values obtained from the computation. Table II illustrates the mechanism of the MSJP method.
Figure 4.1. Block diagram of the MA-IFSDS. Before selecting the three important aspects in designing and implementing the MA-IFSDS, we have to have a big view of the system in form of block diagram. The system’s block diagram is depicted in Figure 4.1.
Figure 4.1. Block diagram of the MA-IFSDS. Before selecting the three important aspects in designing and implementing the MA-IFSDS, we have to have a big view of the system in form of block diagram. The system’s block diagram is depicted in Figure 4.1.
The original Dasarathy process model divides the types of input and output into three categories, namely data (DA), aspect (FE), and decision (DE) [20]. For our system, we define two new terms, namely Inference Input (IEI) and Inference Output (IEO) to replace Decision Input (DEI) and Decision Output (DEO). Information fusion product from each level is an inference for decision making support purpose. As the consequences, JDL’s Level 0 is for Data Input (DAI)-IEO fusion, Level 1 is for Feature Input (FEI)-IEO fusion, Level 2 is for FEI-Feature Output (FEO) fusion, and Level 3 is for IEI-IEO fusion.
The original Dasarathy process model divides the types of input and output into three categories, namely data (DA), aspect (FE), and decision (DE) [20]. For our system, we define two new terms, namely Inference Input (IEI) and Inference Output (IEO) to replace Decision Input (DEI) and Decision Output (DEO). Information fusion product from each level is an inference for decision making support purpose. As the consequences, JDL’s Level 0 is for Data Input (DAI)-IEO fusion, Level 1 is for Feature Input (FEI)-IEO fusion, Level 2 is for FEI-Feature Output (FEO) fusion, and Level 3 is for IEI-IEO fusion.
Referring to Figure 4.1, there will be three types of processed inputs, namely Data (DAI), Feature (FEI), and Inference (IEI), and two types of resulted outputs, namely Feature (FE) and Inference (IEO). We can create the MA-IFSDS architecture according to these inputs- outputs relation and combine it with the selected fusion method which in this case is the MSJP method. The complete architecture of the MA-IFSDS is depicted in Figure 4.2.
Referring to Figure 4.1, there will be three types of processed inputs, namely Data (DAI), Feature (FEI), and Inference (IEI), and two types of resulted outputs, namely Feature (FE) and Inference (IEO). We can create the MA-IFSDS architecture according to these inputs- outputs relation and combine it with the selected fusion method which in this case is the MSJP method. The complete architecture of the MA-IFSDS is depicted in Figure 4.2.
Figure 4.3. MA-IFSDS’ multi-agent configuration [18]. In order to automate the data/information processing, we replace the commandant’s supporting staffs witt agents according to each specific task. So, there will be five supporting agents, namely Intelligence A gen (IA), Operation Agent (OA), Personnel Agent (PA), Logistic Agent (LA) and Communication Electronic: (Comlec) Agent (CA), and one Main Agent. All agents have information fusion capability and use the MSJF method as their information fusion or inference method. The Information Fusion Agents (IFAs) are distributed on each level. Agents do information fusion according to tasks and follow rules that have beer assigned. The IFA configuration and the information fusion products flow is presented in Figure 4.3.
Figure 4.3. MA-IFSDS’ multi-agent configuration [18]. In order to automate the data/information processing, we replace the commandant’s supporting staffs witt agents according to each specific task. So, there will be five supporting agents, namely Intelligence A gen (IA), Operation Agent (OA), Personnel Agent (PA), Logistic Agent (LA) and Communication Electronic: (Comlec) Agent (CA), and one Main Agent. All agents have information fusion capability and use the MSJF method as their information fusion or inference method. The Information Fusion Agents (IFAs) are distributed on each level. Agents do information fusion according to tasks and follow rules that have beer assigned. The IFA configuration and the information fusion products flow is presented in Figure 4.3.
Table 5.1. Subaspect of the Weather State Feasibility.
Table 5.1. Subaspect of the Weather State Feasibility.
To ascertain the validity of MSJP method in carrying out information fusion at Level 0 as depicted in Figure 5.1, MA-IFSDS uses weather, field, and enemy data taken from a military training intelligence database [8,14] from three MO areas, namely DA (Area 1), TR (Area 2), and T (Area 3). In this paper, we only concentrate on the information fusion carried out in Level 0 to Level 2. Because MO is a “go-nogo” operation, we have to assume the conditions or states of the aspects’ subaspects for subsequent processes, so they can be determined as feasible (“go”) or not feasible (“nogo”) subaspect in supporting the MO. Table 5.1 to Table 5.4 lists the conditions for aspects’ subaspects feasibility. Figure 5.2. The divided MO domain and its MAS perspective.
To ascertain the validity of MSJP method in carrying out information fusion at Level 0 as depicted in Figure 5.1, MA-IFSDS uses weather, field, and enemy data taken from a military training intelligence database [8,14] from three MO areas, namely DA (Area 1), TR (Area 2), and T (Area 3). In this paper, we only concentrate on the information fusion carried out in Level 0 to Level 2. Because MO is a “go-nogo” operation, we have to assume the conditions or states of the aspects’ subaspects for subsequent processes, so they can be determined as feasible (“go”) or not feasible (“nogo”) subaspect in supporting the MO. Table 5.1 to Table 5.4 lists the conditions for aspects’ subaspects feasibility. Figure 5.2. The divided MO domain and its MAS perspective.
Table 5.2. Subaspect of the Field State Feasibility. Table 5.3. Subaspect of the Enemy Forces State.
Table 5.2. Subaspect of the Field State Feasibility. Table 5.3. Subaspect of the Enemy Forces State.
The weather, field, and enemy data are presented in Table 5.5, Table 5.6, and Table 5.7. The assumption taken in this case is the weather and field conditions at all DP and all TR are the same. Table 5.5a. Weather State at DA.
The weather, field, and enemy data are presented in Table 5.5, Table 5.6, and Table 5.7. The assumption taken in this case is the weather and field conditions at all DP and all TR are the same. Table 5.5a. Weather State at DA.
Table 5.5c. Weather State at T.
Table 5.5c. Weather State at T.
Table 5.7b. Enemy Composition State. V.3 Final Information Fusion Result
Table 5.7b. Enemy Composition State. V.3 Final Information Fusion Result
The final result of the information fusion is depicted in Figure 5.3 and Figure 5.4. The joint declaration of the weather-field-enemy state probability is indicated by color level of the graphic and the value is pointed out by its elevation from graphic base. From these integrated pictures, the IA will produce an inference as follow: “The execution of the MO will have better probability to be accomplished if it is done in Dry Season, by means of Air and or Sea Field while anticipating the enemy’s Navy and Air Force Engaged/Reinforcing join forces with Composition of Attack, Defense, and Supporting troops”
The final result of the information fusion is depicted in Figure 5.3 and Figure 5.4. The joint declaration of the weather-field-enemy state probability is indicated by color level of the graphic and the value is pointed out by its elevation from graphic base. From these integrated pictures, the IA will produce an inference as follow: “The execution of the MO will have better probability to be accomplished if it is done in Dry Season, by means of Air and or Sea Field while anticipating the enemy’s Navy and Air Force Engaged/Reinforcing join forces with Composition of Attack, Defense, and Supporting troops”
Figure 5.4. The information fusion product at Level 2 which fuses the joint declaration of weather, field, and enemy resulted from Level 1. This view is required in order to observe the unseen information that is not clearly shown in Figure 5.3.
Figure 5.4. The information fusion product at Level 2 which fuses the joint declaration of weather, field, and enemy resulted from Level 1. This view is required in order to observe the unseen information that is not clearly shown in Figure 5.3.
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved