Unmanned aerial vehicle safety improvement method and device

Abstract

The invention provides an unmanned aerial vehicle safety improvement method and device. The method comprises the following steps: S101, determining a first feature weight of an unmanned aerial vehicle criterion dimension and the credibility of an original weight based on the original weight of the preset unmanned aerial vehicle criterion dimension; s102, if the credibility is within a preset threshold range, determining a danger level of a criterion dimension of the unmanned aerial vehicle based on the first feature weight and a preset risk weight; and S103, the behavior of the unmanned aerial vehicle is adjusted based on the danger level so as to improve the safety of the unmanned aerial vehicle. According to the invention, the influence of data of different dimensions on the unmanned aerial vehicle can be analyzed, so that the safety of the unmanned aerial vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the field of drone safety technology, and in particular to a drone safety enhancement method and device. Background Art

[0002] The drone safety improvement method is a method for conducting drone safety research and control. On the drone, the main controller analyzes the data collected by various sensors and issues corresponding commands to the power system based on the analysis results to maintain the normal flight of the drone. If the abnormal behavior of the drone cannot be discovered in time, the flight safety of the drone will be affected or even crash.

[0003] Data of different dimensions have different impacts on the safety of drones. Existing drone safety improvement methods cannot accurately determine the impact of different data on drones, which can easily lead to misjudgment and affect the safe flight of drones. Summary of the invention

[0004] Based on this, the present invention provides a method and device for improving the safety of a drone, so as to analyze the impact of data of different dimensions on the drone, thereby improving the safety of the drone.

[0005] In a first aspect, an embodiment of the present invention provides a method for improving the safety of a drone, the method comprising: S101: determining a first feature weight of the drone criterion dimension and the credibility of the original weight based on a preset original weight of the drone criterion dimension; S102: if the credibility is within a preset threshold range, determining a danger level of the drone criterion dimension based on the first feature weight and a preset risk weight; S103: adjusting the behavior of the drone based on the danger level to improve the safety of the drone.

[0006] Furthermore, the drone criteria dimensions include: drone ID, WIFI status, network traffic, access IP, power supply power, data transmission volume, interconnection status, time information, location and load status.

[0007] Further, S101 includes: S101-1: processing the original weight of the drone criterion dimension into a first weight matrix; S101-2: determining a first feature weight based on the first weight matrix; S101-3: determining the credibility of the original weight based on the first feature weight and a pre-acquired consistency index database.

[0008] Further, S102 includes: S102-1: based on the preset danger level classification, each of the preset risk weights is processed into a second weight matrix, wherein each of the drone criterion dimensions corresponds to a second weight matrix; S102-2: according to the second weight matrix, the corresponding second feature weight is determined, wherein each second feature weight includes a low risk weight, a medium risk weight and a high risk weight; S102-3: according to the second feature weight and the first feature weight, a low risk degree value, a medium risk degree value and a high risk degree value are determined; S102-4: based on the numerical values ​​of the low risk degree value, the medium risk degree value and the high risk degree value, the danger level of the drone criterion dimension is determined.

[0009] Furthermore, S102 also includes: if the credibility is not within a preset threshold range, adjusting the original weight of the drone criterion dimension.

[0010] In a second aspect, an embodiment of the present invention provides a drone safety improvement device, which includes: a first drone safety improvement module, used to determine a first feature weight of the drone criterion dimension and the credibility of the original weight based on a preset original weight of the drone criterion dimension; a second drone safety improvement module, used to determine the danger level of the drone criterion dimension based on the first feature weight and a preset risk weight if the credibility is within a preset threshold range; and a third drone safety improvement module, used to adjust the behavior of the drone based on the danger level to improve the safety of the drone.

[0011] Furthermore, in the first drone safety enhancement module, the drone criteria dimensions include: drone ID, WIFI status, network traffic, access IP, power supply power, data transmission volume, interconnection status, time information, location and load status.

[0012] Furthermore, the first drone safety improvement module is also used to: process the original weight of the drone criterion dimension into a first weight matrix; determine the first feature weight based on the first weight matrix; determine the credibility of the original weight based on the first feature weight and the acquired consistency index database.

[0013] Furthermore, the second UAV safety enhancement module is also used to: based on the preset danger level classification, process each of the preset risk weights into a second weight matrix, wherein each of the UAV criterion dimensions corresponds to a second weight matrix; determine the corresponding second feature weights according to the second weight matrix, wherein each second feature weight includes a low risk weight, a medium risk weight and a high risk weight; determine a low risk degree value, a medium risk degree value and a high risk degree value according to the second feature weight and the first feature weight; determine the danger level of the UAV criterion dimension based on the numerical values ​​of the low risk degree value, the medium risk degree value and the high risk degree value.

[0014] Furthermore, the second drone safety enhancement module is also used to: if the credibility is not within a preset threshold range, adjust the original weight of the drone criterion dimension.

[0015] The beneficial effects of the embodiments of the present invention are as follows:

[0016] The embodiment of the present invention provides a method and device for improving the safety of drones, the method comprising: S101: determining the first feature weight of the drone criterion dimension and the credibility of the original weight based on the preset original weight of the drone criterion dimension; S102: if the credibility is within a preset threshold range, determining the danger level of the drone criterion dimension based on the first feature weight and the preset risk weight; S103: adjusting the behavior of the drone based on the danger level to improve the safety of the drone. Through this application, the impact of data of different dimensions on the drone can be analyzed, thereby improving the safety of the drone. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a drone safety enhancement method provided for the implementation of the present invention;

[0018] Figure 2 A problem processing hierarchy diagram is provided for the implementation of the present invention. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below through the accompanying drawings and specific embodiments.

[0020] Embodiment 1

[0021] like Figure 1 As shown, an embodiment of the present invention provides a method for safely improving a drone. The method includes:

[0022] S101: Determine a first feature weight of the drone criterion dimension and the credibility of the original weight based on a preset original weight of the drone criterion dimension.

[0023] Before S101, from a macro-architecture perspective, drone safety issues need to be divided into Figure 2 The three layers shown are: target layer, criterion layer and solution layer, where the criterion layer contains data of different dimensions, including but not limited to drone ID, WIFI status, network traffic, access IP, power, data transmission volume, interconnection status, time information, location and load status. That is, the drone criterion dimensions include: drone ID, WIFI status, network traffic, access IP, power, data transmission volume, interconnection status, time information, location and load status.

[0024] S101 includes:

[0025] S101-1: Process the original weights of the drone criterion dimension into a first weight matrix.

[0026] Specifically, the first weight matrix is ​​an n×n two-dimensional matrix, where n represents the number of drone criteria dimensions selected during the study. The mutual influence relationship between any two drone criteria dimensions is the matrix value, i.e., the original weight. The original weight is determined based on the consistent matrix method.

[0027] More specifically, the original weight is preset based on experience. Specifically, this embodiment adopts a 1-5 scaling method, that is, for example, factor A (the factor is the dimension of the drone criteria) is very important relative to factor B, and the original weight is 5, then factor B is 1 / 5 relative to factor A; factor A is relatively important relative to factor B, and the weight is 3; factor A is equally important relative to factor B, and the weight is 1 at this time. Note: The wifi name is more important than the drone, so the weight is 3; on the contrary, the wifi is relative to the drone ID, so the weight is 1 / 3, the network traffic size is more important than the drone ID, which is 2, and the access IP is more important than the drone ID, so it is set to 2, and the rest are set similarly.

[0028] S101 - 2 : Determine a first feature weight according to the first weight matrix.

[0029] Here, S101-2 includes:

[0030] S101-2-1: Normalize the original weight of each drone criterion dimension, and then sum them up to obtain the first eigenvector, where each drone criterion dimension corresponds to one first eigenvector.

[0031] S101-2-1: Sum all first eigenvectors (called the first value). The weight of each drone criterion dimension relative to the first value is the first feature weight of the drone criterion dimension. Each drone criterion dimension corresponds to 1 first feature weight.

[0032] S101 - 3 : Determine the credibility CR of the original weight according to the first feature weight and a pre-acquired consistency indicator database.

[0033] Here, S101-3 includes:

[0034] S101-3-1: Based on the calculation method of AW in AHP analysis, calculate the AW value corresponding to each drone criterion dimension.

[0035] S101-3-2: Compare and sum the AW value with the corresponding first feature weight, and then average them based on the number of drone criterion dimensions to determine the maximum feature root λ.

[0036] S101-3-3: Determine the consistency index CI based on the number of drone criterion dimensions and the maximum characteristic root λ.

[0037] S101-3-4: Determine the consistency data RI based on the pre-acquired consistency index database corresponding to the number of drone criterion dimensions.

[0038] Table 1

[0039]

[0040] Table 1 is the pre-acquired consistency index database, such as Figure 2 , that is, 4 drone criterion dimensions were selected, that is, n=4.

[0041] S101-3-5: The ratio of the consistency data RI to the consistency index CI is used as the credibility CR of the original weight.

[0042] S102: If the credibility is within a preset threshold range, the danger level of the drone criterion dimension is determined based on the first feature weight and the preset risk weight; if the credibility is not within the preset threshold range, the original weight of the drone criterion dimension is adjusted.

[0043] Specifically, generally speaking, if the credibility CR is within [0-0.1], the original weight is considered to be credible, otherwise it is necessary to return to adjust the parameter.

[0044] S102 includes:

[0045] S102-1:

[0046] Based on the preset danger level classification, each preset risk weight is processed into a second weight matrix, wherein each drone criterion dimension corresponds to a second weight matrix.

[0047] Specifically, each risk weight is pre-set, and the setting idea is the same as the original weight idea, that is, if the mutual influence between the corresponding two data is relatively large, the weight is high, otherwise, the weight is low. The second weight matrix is ​​an m×m two-dimensional matrix, where m is the number of preset risk level categories, such as Figure 2 There are three preset categories: low risk, medium risk and high risk. The risk weight is determined based on the consistent matrix method.

[0048] S102-2: Determine corresponding second feature weights according to the second weight matrix, wherein each second feature weight includes a low risk weight, a medium risk weight, and a high risk weight.

[0049] Specifically, the principle and calculation formula of this step are the same as S101-2, and will not be repeated here. At this point, each drone criterion dimension corresponds to m second feature weights.

[0050] S102-3: Determine a low risk level value, a medium risk level value, and a high risk level value according to the second feature weight and the first feature weight.

[0051] Specifically, this step is to calculate the degree value of each hazard level classification, specifically, to sum the product of the first feature weight and the second feature weight under the hazard level classification to obtain the corresponding degree value.

[0052] S102-4: Determine the danger level of the drone criterion dimension based on the numerical values ​​of the low danger level value, the medium danger level value, and the high danger level value.

[0053] S103: Adjusting the behavior of the drone based on the danger level to improve the safety of the drone.

[0054] Specifically, if the low-risk value is greater than the values ​​of other levels, the drone is considered safe and does not need to change its flight status; if the medium-risk value is greater than the values ​​of other levels, the drone's flight status needs to be fine-tuned; if the high-risk value is greater than the values ​​of other levels, it means that the drone is in a relatively dangerous state and needs to be forced to return or make an emergency landing. The above adjustment methods can be set as needed.

[0055] The present invention provides a method and device for improving the safety of drones, the method comprising: S101: determining the first feature weight of the drone criterion dimension and the credibility of the original weight based on the preset original weight of the drone criterion dimension; S102: if the credibility is within a preset threshold range, determining the danger level of the drone criterion dimension based on the first feature weight and the preset risk weight; S103: adjusting the behavior of the drone based on the danger level to improve the safety of the drone. Through this application, the impact of data of different dimensions on the drone can be analyzed, thereby improving the safety of the drone.

[0056] Embodiment 2

[0057] Compared with the first embodiment, this embodiment provides a more specific method for improving the safety of a drone.

[0058] S101: Determine the first feature weight of the drone criterion dimension and the credibility of the original weight based on the preset drone criterion dimension. See Table 2 for specific parameters.

[0059] Table 2

[0060]

[0061] In this embodiment, four dimensions, namely, drone ID, WIFI status, network traffic, and access IP, are illustrated.

[0062] S101 includes:

[0063] S101-1: Processing the original weights of the drone criterion dimension into a first weight matrix.

[0064] Specifically, the original weights of the drone criterion dimensions are determined empirically, and the original weights are processed into a 4×4 first weight matrix, i.e., the contents of the 2nd row and 2nd column to the 5th row and 5th column of Table 2.

[0065] S101 - 2 : Determine a first feature weight according to a first weight matrix.

[0066] After constructing the first weight matrix, the above judgment matrix is ​​normalized, that is, each cell is normalized vertically, and the calculation method is: drone ID1 / (1+3+2+2)=0.125; WIFI name: 3 / (1+3+2+2)=0.375, and the rest is not repeated.

[0067] Determine the first eigenvector, for example: drone ID: 0.125+0.143+0.125+0.091=0.484; network traffic: 0.25+0.214+0.25+0.364=1.078, and the rest will not be repeated.

[0068] Determine the weight of the first feature: Drone ID: 0.484 / (0.484+1.667+1.078+0.771)=12.09%; Wi-Fi name: 1.667 / (0.484+1.667+1.078+0.771)=41.68%, and the rest will not be repeated.

[0069] S101 - 3 : Determine the credibility of the original weight according to the first feature weight and a pre-acquired consistency index database.

[0070] Determine the AW value: For the drone ID, its AW value is: 1*12.094%+0.333*41.68%+0.5*26.948%+0.5*19.278%=0.49083, and the rest will not be repeated.

[0071] Determine the largest eigenvalue λ:

[0072]

[0073] Where n is the number of drone criterion dimensions. In this embodiment, n=4, W i is the weight value of the first feature.

[0074] In this embodiment, λ=1 / 4×(0.49 / 0.12094+1.704 / 0.4168+1.105 / 0.26948+0.778 / 0.19278)=4.071.

[0075] Determine the CI value: CI = (λ-n) / (n-1) = 0.024. The consistency index is represented by CI, and the smaller the CI, the greater the consistency.

[0076] Determine the credibility CR: CR = CI / RI = 0.027, RI is determined by the query consistency index database, and since n = 4, RI = 0.89.

[0077] S102: If the credibility is within a preset threshold range, determine the danger level of the drone criterion dimension based on the first feature weight and a preset risk weight. If the credibility is not within the preset threshold range, adjust the original weight of the drone criterion dimension.

[0078] Specifically, if CR is within the range of [0-0.1], the judgment matrix is ​​considered to pass the consistency test, otherwise it does not have satisfactory consistency. Since CR = 0.027 is within [0-0.1], the original weight is considered to be credible.

[0079] Table 3

[0080]

[0081]

[0082] Table 4

[0083]

[0084] Table 5

[0085]

[0086] Table 6

[0087]

[0088] S102-1: Based on the preset danger level classification, each preset risk weight is processed into a second weight matrix, wherein each drone criterion dimension corresponds to a second weight matrix.

[0089] Specifically, the second row and second column to the fourth row and fourth column of Table 3 to Table 6 are an m×m second weight matrix. In the present application, the values ​​in the second weight matrix are determined by the degree of mutual influence between risks, which is based on experience.

[0090] S102-2: Determine corresponding second feature weights according to the second weight matrix, wherein each second feature weight includes a low-risk weight, a medium-risk weight, and a high-risk weight.

[0091] Specifically, the process of normalizing and calculating the second feature weight is the same as the process of calculating the first weight matrix, which will not be repeated again.

[0092] S102-3: Determine a low risk level value, a medium risk level value, and a high risk level value according to the second feature weight and the first feature weight.

[0093] Specifically, the calculation method is:

[0094] Low risk value: 0.5954×0.12094+0.1061×0.4168+0.4286×0.26948+0.6328×0.19278=0.3537;

[0095] Moderate risk value: 0.2764×0.12094+0.2605×0.4168+0.4286×0.26948+0.1924×0.19278=0.2946;

[0096] High risk level value: 0.1282×0.12094+0.6334×0.4168+0.1428×0.26948+0.1748×0.19278=0.3516.

[0097] S102-4: Determine the danger level of the drone criterion dimension based on the numerical values ​​of the low danger level value, the medium danger level value, and the high danger level value.

[0098] Since the low risk level value > high risk level value > medium risk level value, the four drone criterion dimensions selected in this embodiment are low risk levels.

[0099] S103: Adjusting the behavior of the drone based on the danger level to improve the safety of the drone.

[0100] Specifically, since the four drone criterion dimensions selected in this embodiment are of low hazard level, it is determined that the current drone status is good and no intervention is required.

[0101] This technology can more accurately respond to changes in parameters of different dimensions of the drone, thereby improving the safety of the drone.

[0102] Embodiment 3

[0103] Corresponding to the drone safety lifting method in Embodiment 1 and Embodiment 2, an embodiment of the present invention provides a drone safety lifting device, which includes:

[0104] The first drone safety enhancement module is used to determine the first feature weight of the drone criterion dimension and the credibility of the original weight based on the preset original weight of the drone criterion dimension; the second drone safety enhancement module is used to determine the danger level of the drone criterion dimension based on the first feature weight and the preset risk weight if the credibility is within a preset threshold range; the third drone safety enhancement module is used to adjust the behavior of the drone based on the danger level to improve the safety of the drone.

[0105] In the first drone safety improvement module, the drone criteria dimensions include: drone ID, WIFI status, network traffic, access IP, power supply power, data transmission volume, interconnection status, time information, location and load status.

[0106] The first drone safety enhancement module is also used to: process the original weight of the drone criterion dimension into a first weight matrix; determine the first feature weight according to the first weight matrix; and determine the credibility of the original weight according to the first feature weight and a pre-acquired consistency indicator database.

[0107] The second drone safety enhancement module is also used to: based on the preset danger level classification, process each preset risk weight into a second weight matrix, wherein each drone criterion dimension corresponds to a second weight matrix; determine the corresponding second feature weight according to the second weight matrix, wherein each second feature weight includes a low risk weight, a medium risk weight and a high risk weight; determine a low risk degree value, a medium risk degree value and a high risk degree value according to the second feature weight and the first feature weight; determine the danger level of the drone criterion dimension based on the numerical values ​​of the low risk degree value, the medium risk degree value and the high risk degree value.

[0108] The second drone safety enhancement module is further used to: if the credibility is not within a preset threshold range, adjust the original weight of the drone criterion dimension.

[0109] The drone safety lifting device provided in this embodiment has the same implementation principle and technical effects as those in the aforementioned embodiments. For the sake of brief description, for matters not mentioned in the drone safety lifting device embodiment, reference may be made to the corresponding contents in the aforementioned drone safety lifting method embodiment, and this embodiment will not repeat them any more.

[0110] The above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for safely lifting an unmanned aerial vehicle, characterized in that: The method comprises: S101: Determine a first feature weight of the drone criterion dimension and the credibility of the original weight based on a preset original weight of the drone criterion dimension; S102: If the credibility is within a preset threshold range, determining the danger level of the drone criterion dimension based on the first feature weight and a preset risk weight; S103: Adjusting the behavior of the drone based on the danger level to improve the safety of the drone.

2. The drone safety lifting method according to claim 1, characterized in that: The drone criteria dimensions include: drone ID, WIFI status, network traffic, access IP, power supply, data transmission volume, interconnection status, time information, location and load status.

3. The drone safety lifting method according to claim 1, characterized in that: S101 includes: S101-1: Processing the original weights of the drone criterion dimension into a first weight matrix; S101-2: Determine a first feature weight according to the first weight matrix; S101 - 3 : Determine the credibility of the original weight according to the first feature weight and a pre-acquired consistency index database.

4. The drone safety lifting method according to claim 3, characterized in that: S102 includes: S102-1: Based on the preset danger level classification, each of the preset risk weights is processed into a second weight matrix, wherein each of the drone criterion dimensions corresponds to a second weight matrix; S102-2: Determine corresponding second feature weights according to the second weight matrix, wherein each second feature weight includes a low risk weight, a medium risk weight, and a high risk weight; S102-3: Determine a low risk level value, a medium risk level value, and a high risk level value according to the second feature weight and the first feature weight; S102-4: Determine the danger level of the drone criterion dimension based on the numerical values ​​of the low danger level value, the medium danger level value, and the high danger level value.

5. The drone safety lifting method according to claim 4, characterized in that: S102 also includes: if the credibility is not within a preset threshold range, adjusting the original weight of the drone criterion dimension.

6. A drone safety lifting device, characterized in that: The device comprises: A first drone safety enhancement module, configured to determine a first feature weight of the drone criterion dimension and a credibility of the original weight based on a preset original weight of the drone criterion dimension; a second drone safety enhancement module, configured to determine the danger level of the drone criterion dimension based on the first feature weight and a preset risk weight if the credibility is within a preset threshold range; The third drone safety enhancement module is used to adjust the behavior of the drone based on the danger level to improve the safety of the drone.

7. The drone safety lifting device according to claim 6, characterized in that: In the first drone safety enhancement module, the drone criteria dimensions include: drone ID, WIFI status, network traffic, access IP, power supply power, data transmission volume, interconnection status, time information, location and load status.

8. The drone safety lifting device according to claim 6, characterized in that: The first drone safety enhancement module is also used for: Processing the original weights of the drone criterion dimensions into a first weight matrix; Determine a first feature weight according to the first weight matrix; The credibility of the original weight is determined according to the first feature weight and a pre-acquired consistency indicator database.

9. The drone safety lifting device according to claim 8, characterized in that: The second drone safety enhancement module is further used for: Based on the preset danger level classification, each of the preset risk weights is processed into a second weight matrix, wherein each of the drone criterion dimensions corresponds to a second weight matrix; Determining corresponding second feature weights according to the second weight matrix, wherein each second feature weight includes a low risk weight, a medium risk weight, and a high risk weight; Determine a low risk level value, a medium risk level value, and a high risk level value according to the second feature weight and the first feature weight; Based on the numerical values ​​of the low risk level value, the medium risk level value and the high risk level value, the risk level of the drone criterion dimension is determined.

10. The drone safety lifting device according to claim 9, characterized in that: The second drone safety enhancement module is further used to: if the credibility is not within a preset threshold range, adjust the original weight of the drone criterion dimension.

Images