A Multi-Sensor Information Fusion Method Based on Trust Index Divergence

Abstract

This invention provides a multi-sensor information fusion method based on trust index divergence. When determining evidence weights, it considers both the degree of conflict between pieces of evidence and the inherent uncertainty of the evidence itself. This overcomes the problem that previous studies mostly relied on the degree of conflict between pieces of evidence when determining evidence weights, rarely considering the characteristics of the evidence itself. It can fully utilize existing evidence information to reasonably measure and correct the degree of conflict, reducing the impact of conflicting evidence and avoiding unreasonable results. Information fusion of multi-sensor decision-making layer data overcomes the shortcomings of previous methods, such as high computational cost and the influence of subjective factors in manually assigned weight coefficients. It reduces the impact of conflicting evidence and accelerates the speed and accuracy of multi-sensor information fusion. The multi-sensor information fusion method proposed in this invention is suitable for target recognition scenarios with high conflict information from multiple sensors in engineering applications, and can effectively solve the problem of fusing high-conflict information from multiple sensors.

Description

Technical Field

[0001] This invention relates to the field of information fusion, and in particular to a fusion method of a multi-sensor information fusion algorithm and DS evidence theory. Background Technology

[0002] DS evidence theory is a widely used method for uncertainty modeling and reasoning. It characterizes uncertainty by constructing confidence intervals through the definition of trust functions and likelihood functions. Its defined combination rules satisfy the commutative and associative laws. Evidence theory can handle uncertainty problems caused by randomness and fuzziness well without prior probability and conditional probability density. It effectively solves the problem of multi-sensor information fusion and has been widely used in information fusion, target recognition, fault diagnosis, and image processing in recent years.

[0003] Multi-sensor information fusion involves automatically analyzing, synthesizing, and optimizing multi-level, multi-angle data from multiple homogeneous or heterogeneous sensors distributed at different locations, according to certain criteria, to achieve the required decision-making and estimation, and arrive at a final judgment. Multi-sensor information fusion can reduce potential redundancy and contradictions between multi-sensor information, enhance information complementarity, reduce uncertainty, and obtain a consistent interpretation of things or the environment. DS evidence theory provides strong theoretical support for the expression and synthesis of uncertain information and has become an important theory and method in the field of multi-source information fusion. However, in practical applications, due to electromagnetic interference and the potential instability of the sensors themselves (such as sudden failures of a few sensors), the information collected from different sensors has significant uncertainty and high conflict. When using DS evidence theory to fuse highly conflicting evidence, counterintuitive results often occur, which limits the application of DS evidence theory.

[0004] Currently, in research on evidence theory, important research directions include how to more accurately determine the degree of conflict between pieces of evidence and its measurement methods, how to identify indicators characterizing the strength of conflict, and how to rationally utilize the information inherent in the evidence itself when determining its weight. Currently, many research methods determine the weight of evidence primarily based on the degree of conflict between pieces of evidence, rarely considering the characteristics of the evidence itself. While these methods have improved the synthesis results of highly conflicting evidence to varying degrees, overall, these methods are not yet perfect, and the synthesis effect is not ideal. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides a multi-sensor information fusion method based on trust index divergence.

[0006] The steps of the technical solution adopted by this invention to solve its technical problem are as follows:

[0007] (1) All possible n single-type targets θ i A target type set Θ is formed, denoted as Θ = {θ1, θ2, ... θ}. i ,…,θ n}, i = 1, 2, ..., n, define A j Candidate target types include one or more single target types, i.e. If sensors j = 1, 2, ..., M, N detect and identify the same target, and each sensor assigns a basic probability value corresponding to a candidate target type, then the l-th sensor is said to provide the candidate target type A. j The corresponding basic probability is assigned the value m. l (A j The list contains evidence number e. l ,Right now

[0008] E l :m l (A1),m l (A2),…,m l (A j ),…,m l (A M ), (l=1,2,…,N), where m l Evidence E l The basic probability assignment function, m l (A j ) represents candidate target type A j The corresponding basic probability assignment,

[0009] (2) N sensors provide N pieces of evidence; for evidence E l Performing a Pignistic probability transformation yields the corresponding new evidence E. l ′, l=1,2,…,N, and simultaneously new evidence E is obtained. l The basic probability assignment function m' l ′:

[0010] (3) Calculate E for any two of the N pieces of evidence provided by the N sensors. u E v Trust index divergence BED(E) u E v ), u=1, 2,...,N, v=1, 2,...,N;

[0011] Calculate the trust index divergence between each pair of N pieces of evidence provided by N sensors, and construct an N×N trust divergence matrix (DMM).

[0012] Calculation of evidence E u Average Trust Index Divergence

[0013]

[0014] (4) Determine the validity of all other evidence against evidence E. u Support Sup u ,

[0015]

[0016] Determining Evidence u Credibility u ,

[0017]

[0018] (5) Calculate evidence E u credibility entropy Ed u ,

[0019]

[0020] Calculation of evidence E u Information content I u ,

[0021]

[0022] (6) Determine Evidence E u weight w u ,

[0023]

[0024] (7) Weighted average evidence is obtained by weighting and correcting the N pieces of evidence provided by N sensors. Its basic probability assignment function is Represented as:

[0025]

[0026] Among them, w u Evidence E u The weight, Weighted average evidence Candidate Target Type A j Basic probability assignment, m u (A j Evidence E u Candidate Target Type A j The basic probability assignment;

[0027] (8) Using the DS evidence fusion rule to analyze weighted average evidence Perform N-1 synthesis The fusion result E is obtained;

[0028] The fusion result E is:

[0029] E:m(θ1),m(θ2),…,m(θ i ),…,m(θ n (i = 1, 2, ..., n)

[0030] The target type with the highest basic probability in the fusion result E is the final result of multi-sensor target recognition.

[0031] The basic probability assignment function m l ′:

[0032]

[0033] in Let m be the Pignistic probability transformation function. l ′(θ i E is new evidence. l 'Target Single Type θ i The basic probability assignment, the Pignistic probability transformation function is defined as follows:

[0034] Where |A j | Indicates candidate target type A j The number of target single types included.

[0035] The trust index divergence BED(E) u E v )for:

[0036]

[0037] Where, m u ′(θ i Evidence E u The corresponding new evidence E after Pignistic probability transformation u 'Target Single Type θ i Basic probability assignment, m v ′(θ i Evidence E v The corresponding new evidence E after Pignistic probability transformation v 'Target Single Type θ i The basic probability is assigned.

[0038] The trust divergence matrix (DMM) is:

[0039]

[0040] The DS evidence fusion rules are as follows:

[0041]

[0042] Where m(θ) i ) represents the target single-type θ in the fusion result E. i Basic probability assignment, A s A t (s = 1, 2, ..., M, t = 1, 2, ..., M) represents the weighted average evidence. Candidate target types, Weighted average evidence Candidate Target Type A s A t The basic probability is assigned.

[0043] An electronic device includes one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to perform the methods described above.

[0044] A computer-readable storage medium storing program code that can be invoked by a processor to perform the method described above.

[0045] The beneficial effects of this invention lie in its consideration of both the degree of conflict between pieces of evidence and the uncertainty of the evidence itself when determining evidence weights. This overcomes the problem that previous studies mostly relied on the degree of conflict between pieces of evidence when determining evidence weights, rarely considering the characteristics of the evidence itself. It can fully utilize existing evidence information to reasonably measure and correct the degree of conflict, reducing the impact of conflicting evidence and avoiding unreasonable results. Using the above method to fuse multi-sensor decision-making layer data overcomes the shortcomings of previous methods, such as high computational load and the influence of subjective factors in manually assigned weight coefficients. It reduces the impact of conflicting evidence and accelerates the speed and accuracy of multi-sensor information fusion. The multi-sensor information fusion method proposed in this invention is suitable for target recognition scenarios with high conflict information from multiple sensors in engineering applications, and can effectively solve the problem of fusing high-conflict information from multiple sensors. Attached Figure Description

[0046] Figure 1 This is a flowchart of the multi-sensor information fusion method based on trust index divergence proposed in this invention. Detailed Implementation

[0047] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0048] Taking "aerial target recognition," a function in multi-sensor information fusion, as an example, we will illustrate the specific implementation process of the above-mentioned multi-sensor information fusion method based on evidence similarity, that is, using the proposed method to determine the target type.

[0049] (1) A multi-sensor system consisting of six sensors is used to observe and identify an aerial target. During the identification process, there are three possible target types: θ1: tanker aircraft, θ2: early warning aircraft, and θ3: bomber aircraft. Therefore, the target type set is Θ = {T: tanker aircraft, W: early warning aircraft, R: bomber}. When the six sensors observe and identify the aerial target, they assign basic probability values ​​to the candidate target types, i.e., the six pieces of evidence under the target type set Θ are (where the candidate target types are [A1:T; A2:W; A3:R; A4:T,W; A5:T,R; A6:Θ]):

[0050] E1: m1(T)=0.42, m1(W)=0.28, m1(R)=0.30, m1(T,W)=0, m1(T,R)=0, m1(Θ)=0;

[0051] E2: m2(T)=0.01, m2(W)=0.90, m2(R)=0.08, m2(T,W)=0.01, m2(T,R)=0, m2(Θ)=0;

[0052] E3: m3(T)=0.79, m3(W)=0.16, m3(R)=0.02, m3(T,W)=0, m3(T,R)=0.03, m3(Θ)=0;

[0053] E4: m4(T)=0.52, m4(W)=0, m4(R)=0.18, m4(T,W)=0, m4(T,R)=0, m4(Θ)=0.30;

[0054] E5: m5(T)=0.65, m5(W)=0.09, m5(R)=0.11, m5(T,W)=0.15, m5(T,R)=0, m5(Θ)=0;

[0055] E6: m6(T)=0.75, m6(W)=0.10, m6(R)=0, m6(T,W)=0, m6(T,R)=0, m6(Θ)=0.15.

[0056] (2) The six pieces of evidence provided by the above six sensors were subjected to Pignistic probability transformation, resulting in the following six pieces of evidence:

[0057] E1′: m1′(T)=0.4200, m1′(W)=0.2800, m1′(R)=0.3000;

[0058] E2′: m2′(T)=0.0150, m2′(W)=0.9050, m2′(R)=0.0800;

[0059] E3′: m3′(T)=0.8050, m3′(W)=0.1600, m3′(R)=0.0350;

[0060] E4′: m4′(T)=0.620, m4′(W)=0.1000, m4′(R)=0.2800;

[0061] E5′: m5′(T)=0.7250, m5′(W)=0.1650, m5′(R)=0.1100;

[0062] E6′: m6′(T)=0.8000, m6′(W)=0.1500, m6′(R)=0.0500.

[0063] (3) Calculate the confidence index divergence between each pair of the 6 pieces of evidence, resulting in a 6×6 divergence. Trust Divergence Matrix (DMM):

[0064]

[0065] Calculate the average confidence index divergence for each piece of evidence:

[0066] BED1=0.2765; BED2=0.7040; BED3=0.2423; BED4=0.2508; BED5=0.2052; BED6=0.2353.

[0067] (4) Determine the support level of each piece of evidence:

[0068] Sup1=3.6162; Sup2=1.4204; Sup3=4.1263; Sup4=3.9866; Sup5=4.8726; Sup6=4.2502.

[0069] Further determine the credibility of each piece of evidence:

[0070] Cred1=0.1624; Cred2=0.0638; Cred3=0.1853; Cred4=0.1790; Cred5=0.2188; Cred6=0.1908.

[0071] (5) Calculate the reliability entropy of each piece of evidence:

[0072] Ed1=1.5610; Ed2=0.5770; Ed3=1.0039; Ed4=2.3795; Ed5=1.7152; Ed6=1.4751.

[0073] The information content of the evidence was calculated, and the results are as follows:

[0074] I1=4.7634; I2=1.7808; I3=2.7288; I4=10.7995; I5=5.5578; I6=4.3716.

[0075] (6) Determine the weight of each piece of evidence:

[0076] w1=0.1439; w2=0.0211; w3=0.0940; w4=0.3596; w5=0.2262; w6=0.1552.

[0077] The six pieces of evidence from the six sensors were weighted and adjusted to obtain a weighted average evidence. for:

[0078]

[0079] (7) Evidence of weighted average After five synthesis processes, the final fusion result E is:

[0080] E: m(T)=0.9991, m(W)=0.0002, m(R)=0.0007.

[0081] In the final fusion result E, the basic probability of target type T is the highest, at 0.9991. The synthesis results show that when observing aerial targets using multiple sensors, the multi-sensor information fusion method based on the trust index divergence is used for aerial target identification, and the true target is identified as T: a refueling aircraft, with a basic probability of 0.9991 for target T.

Claims

1. A multi-sensor information fusion method based on trust index divergence, characterized in that... Includes the following steps: (1) All possible n target single types Form a target type set , represented as , ,definition Candidate target types include one or more single target types, i.e. , , If each of the following sensors detects and identifies the same target, and each sensor assigns a basic probability value corresponding to the candidate target type, then the first sensor is called the third sensor. Each sensor provides candidate target types Corresponding basic probability assignment The list is number 1 evidence ,Right now , ,in As evidence The basic probability assignment function, Candidate target type The corresponding basic probability assignment, ; (2) Each sensor provides One piece of evidence; on the evidence The corresponding new evidence was obtained by performing a piggyback probability transformation. , At the same time, new evidence was obtained. Basic probability assignment function : (3) Calculation Provided by each sensor Any two pieces of evidence , Trust index divergence , , ; calculate Provided by each sensor Construct a trust index divergence between each pair of pieces of evidence. Trust divergence matrix ; Calculation evidence Average Trust Index Divergence : ； (4) Determine the validity of all other evidence against the evidence. support , ; Determine evidence Credibility , ； (5) Calculation evidence Credibility entropy , ； Calculation evidence Information content , ； (6) Determine evidence weight , ； (7) Provided by each sensor The evidence was weighted and adjusted to obtain a weighted average evidence. Its basic probability assignment function is , Represented as: : ； in, As evidence The weight, Weighted average evidence Candidate target types Basic probability assignment, As evidence Candidate target types The basic probability assignment; (8) Using the DS evidence fusion rule to analyze weighted average evidence Perform N-1 synthesis The fusion result was obtained. ; Fusion results for: ( ) Fusion results The target type with the highest basic probability assignment is the final result of multi-sensor target recognition.

2. The multi-sensor information fusion method based on trust index divergence according to claim 1, characterized in that: The basic probability assignment function : ； in This is the Pignistic probability transformation function. For new evidence Target Single Type The basic probability assignment, the Pignistic probability transformation function is defined as follows: ( , ),in Indicates the type of candidate target Number of target single types included.

3. The multi-sensor information fusion method based on trust index divergence according to claim 2, characterized in that: The trust index divergence for: ； in, As evidence New evidence corresponding to the Pignistic probability transformation Target Single Type Basic probability assignment, As evidence New evidence corresponding to the Pignistic probability transformation Target Single Type The basic probability is assigned.

4. The multi-sensor information fusion method based on trust index divergence according to claim 3, characterized in that: The trust divergence matrix for: 。 5. The multi-sensor information fusion method based on trust index divergence according to claim 1, characterized in that: The DS evidence fusion rules are as follows: ； in, For the fusion result Target Single Type Basic probability assignment, , ( , (This refers to weighted average evidence) Candidate target types, , Weighted average evidence Candidate target types , The basic probability is assigned.

6. An electronic device, characterized in that, include: One or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the method as described in any one of claims 1-5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code that can be invoked by a processor to execute the method as described in any one of claims 1-5.

Images