Target deterrence effect evaluation method based on quadruped robot

Abstract

The invention provides a target deterrence effect evaluation method based on a quadruped robot, and belongs to the technical field of quadruped robots, and the method specifically comprises the steps: obtaining a monitoring image of the quadruped robot, and determining that a dangerous target with a preset dangerous behavior exists in the monitoring image according to an analysis result of the monitoring image, when it is determined that the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets, the dangerous targets are deterred through the calling function of the quadruped robot, and the similar situation of the different dangerous targets and the specified dangerous behaviors is determined based on the image of the quadruped robot; when it is determined that no risk exists through the similar situation, the recognition results of the target states of different dangerous targets are determined based on the analysis results of the images, the action strategy of the quadruped robot is determined through the recognition results, and the accuracy of assessment of the target deterrence effect is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of quadruped robots, and particularly relates to a method for evaluating the target deterrence effect based on a quadruped robot. Background Art

[0002] Quadruped robots can be used for patrolling and protecting target areas, reducing the need for patrol personnel. Specifically, in the invention patent CN201610017541.4 "A Security Robot", a security robot is set to replace manual all-weather patrolling on ships, realizing the integration of mobile detection, alarm and expulsion, and maintaining the navigation safety of ships. However, at the same time, how to evaluate its deterrence effect can help determine its deterrence level against potential intruders, so as to guide the formulation of security strategies and the allocation of resources, which has become an urgent technical problem to be solved.

[0003] In view of the above technical problems, the present invention provides a method for evaluating the target deterrence effect based on a quadruped robot. Summary of the Invention

[0004] To achieve the object of the present invention, the present invention adopts the following technical solutions:

[0005] According to one aspect of the present invention, there is provided a method for evaluating the target deterrence effect based on a quadruped robot.

[0006] A method for evaluating the target deterrence effect based on a quadruped robot, characterized in that:

[0007] S1 Obtain the monitoring image of the quadruped robot, and when it is determined that there is a dangerous target with a preset dangerous behavior in the monitoring image according to the analysis result of the monitoring image, and when the risk is determined to be controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets, proceed to the next step;

[0008] S2 Deter the dangerous target through the voice function of the quadruped robot, and based on the image of the quadruped robot, determine the similarity between different dangerous targets and the specified dangerous behavior. When it is determined that there is no risk through the similarity situation, proceed to the next step;

[0009] S3 Determine the recognition result of the target state of different dangerous targets based on the analysis result of the image, and determine the action strategy of the quadruped robot through the recognition result.

[0010] The beneficial effects of the present invention are as follows:

[0011] 1. Determine whether the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets, thereby realizing the assessment of the overall risk of dangerous targets from the perspective of the matching results of different preset dangerous behaviors of dangerous targets, and avoiding the risk of risk expansion caused by the failure to issue early warning information in a timely manner due to uncontrollable risks.

[0012] 2. Based on the analysis results of the image, identify the target states of different dangerous targets and the action strategies of the quadruped robot, avoiding the technical problems of being unable to further achieve deterrence and issue early warning information in a timely manner caused by adopting fixed action strategies, and ensuring the reliability of the setting of action strategies.

[0013] A further technical solution is that the image and the monitoring image are determined according to the extraction results of the camera of the quadruped robot.

[0014] A further technical solution is that the preset dangerous behaviors include holding dangerous instruments and different types of aggressive actions.

[0015] A further technical solution is that when there is no dangerous target with preset dangerous behaviors, there is no need to perform threat processing through the quadruped robot.

[0016] A further technical solution is that the identification results of the target states include successful deterrence and failure to threaten successfully but no specified dangerous behavior.

[0017] A further technical solution is that the action strategy of the quadruped robot is determined through the identification result, specifically including:

[0018] Divide the dangerous target into different types through the identification result, and determine the target recognition risk of different types of dangerous targets according to the number of different types of dangerous targets;

[0019] Determine the comprehensive recognition risk value based on the target recognition risks of different types of dangerous targets, and determine the action strategy of the quadruped robot through the comprehensive recognition risk value.

[0020] A further technical solution is that the action strategy of the quadruped robot is determined through the comprehensive recognition risk value, specifically including:

[0021] Determine the matching action strategy of the quadruped robot based on the comprehensive recognition risk value, and determine the action strategy of the quadruped robot based on the matching action strategy.

[0022] Other features and advantages will be described in the subsequent specification. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification and the drawings.

[0023] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following specifically gives preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. Description of the Drawings

[0024] By referring to the accompanying drawings and describing its exemplary embodiments in detail, the above and other features and advantages of the present invention will become more obvious.

[0025] Figure 1 is a flowchart of a method for evaluating the target deterrence effect based on a quadruped robot;

[0026] Figure 2 is a flowchart of a method for determining risk controllability based on the matching situation of different preset dangerous behaviors of different dangerous targets;

[0027] Figure 3 is a flowchart of a method for determining no risk through similar situations;

[0028] Figure 4 is a flowchart of a method for determining the action strategy of the quadruped robot based on the recognition result. Detailed Embodiments

[0029] To enable those skilled in the art to better understand the technical solutions in this specification, the following will, in conjunction with the accompanying drawings in the embodiments of this specification, clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0030] Embodiment 1

[0031] To solve the above problems, according to one aspect of the present invention, as Figure 1 shown, a method for evaluating the target deterrence effect based on a quadruped robot is provided, which is characterized in that:

[0032] S1 Obtain the monitoring image of the quadruped robot, and when it is determined based on the analysis result of the monitoring image that there is a dangerous target with a preset dangerous behavior in the monitoring image, and when it is determined that the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets, proceed to the next step;

[0033] Furthermore, the image and the monitoring image are determined according to the extraction result of the camera of the quadruped robot.

[0034] It should be noted that the preset dangerous behaviors include holding dangerous instruments and different types of aggressive actions.

[0035] Further, when there is no dangerous target with a preset dangerous behavior, there is no need to perform threat processing through the quadruped robot.

[0036] It should be noted that, as Figure 2 shown, determining that the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets specifically includes:

[0037] Determining the matching results of different dangerous targets and different types of preset dangerous behaviors through the matching situation with different preset dangerous behaviors, and determining the risk coefficients of different dangerous targets based on the matching results;

[0038] Determining the risk weight values of different dangerous targets according to the distance situation between different dangerous targets and other dangerous targets and the distance situation between different dangerous targets and the quadruped robot, and using the risk weight values to correct the risk coefficients to obtain the corrected risk coefficients of different dangerous targets;

[0039] Obtaining the number of the dangerous targets, and determining the comprehensive risk coefficient in combination with the corrected risk coefficients of different dangerous targets, and determining whether the risk is controllable based on the comprehensive risk coefficient.

[0040] Further, determining whether the risk is controllable based on the comprehensive risk coefficient specifically includes:

[0041] When the comprehensive risk coefficient meets the requirements, it is determined that the risk is controllable.

[0042] In another possible embodiment, determining that the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets specifically includes:

[0043] Judging whether the number of the dangerous targets is greater than the preset target number. If so, it is determined that the risk is uncontrollable. If not, proceed to the next step;

[0044] Determining the risk weight values of different dangerous targets according to the distance situation between different dangerous targets and other dangerous targets and the distance situation between different dangerous targets and the quadruped robot, and judging whether the sum of the risk weight values of the dangerous targets meets the requirements. If not, it is determined that the risk is uncontrollable. If so, proceed to the next step;

[0045] Determining the matching results of different dangerous targets and different types of preset dangerous behaviors through the matching situation with different preset dangerous behaviors, and determining the risk coefficients of different dangerous targets based on the matching results, and judging whether the number of dangerous targets with risk coefficients not meeting the requirements meets the requirements. If so, proceed to the next step. If not, it is determined that the risk is uncontrollable;

[0046] Use the risk weight value to correct the risk coefficient to obtain the corrected risk coefficients of different dangerous targets, and determine whether the number of dangerous targets whose corrected risk coefficients do not meet the requirements meets the requirements. If so, proceed to the next step; if not, determine that the risk is uncontrollable.

[0047] Obtain the number of the dangerous targets, and combine the corrected risk coefficients of different dangerous targets to determine the comprehensive risk coefficient, and determine whether the risk is controllable based on the comprehensive risk coefficient.

[0048] In another possible embodiment, it is determined that the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets, specifically including:

[0049] S11 Determine whether the number of the dangerous targets is within the preset target number range. If so, proceed to the next step; if not, proceed to step S14.

[0050] S12 Determine the matching results of different dangerous targets and different types of preset dangerous behaviors through the matching situation with different preset dangerous behaviors, and determine the risk coefficients of different dangerous targets based on the matching results. Determine whether there are dangerous targets whose risk coefficients do not meet the requirements. If so, proceed to step S14; if not, proceed to the next step.

[0051] S13 Calculate the risk weight values of different dangerous targets according to the distance between different dangerous targets and other dangerous targets and the distance between different dangerous targets and the quadruped robot. Use the risk weight value to correct the risk coefficient to obtain the corrected risk coefficients of different dangerous targets, and determine whether there are dangerous targets whose corrected risk coefficients do not meet the requirements. If so, proceed to the next step; if not, determine that the risk is controllable.

[0052] S14 Divide the dangerous targets into different groups according to the distance situation of different dangerous targets, and determine the group risk coefficients of different groups according to the number of dangerous targets in different groups and the corrected risk coefficients of different dangerous targets. Determine whether there are groups whose group risk coefficients do not meet the requirements. If so, proceed to the next step; if not, determine that the risk is controllable.

[0053] S15 Obtain the distances between different groups, and determine the comprehensive risk coefficient by combining the group risk coefficients of different groups, and determine whether the risk is controllable based on the comprehensive risk coefficient.

[0054] S2 Deter the dangerous targets through the voice function of the quadruped robot, and based on the image of the quadruped robot, determine the similarity between different dangerous targets and the specified dangerous behavior. When it is determined that there is no risk through the similarity situation, proceed to the next step.

[0055] Further, deter the dangerous target through the voice function of the quadruped robot, specifically including:

[0056] Determine the voice command of the quadruped robot according to the voice command sent down by the control platform and the preset voice command, and deter the dangerous target based on the voice command.

[0057] It should be noted that the specified dangerous behaviors include potential dangerous movement actions and different types of aggressive actions.

[0058] Specifically, as Figure 3 shown, determine that there is no risk through the similarity situation, specifically including:

[0059] Determine the similarity coefficient between the dangerous target and different specified dangerous behaviors according to the action recognition results of different dangerous targets, and identify the suspected dangerous behaviors of the dangerous target based on the similarity coefficient. Take the dangerous target with suspected dangerous behaviors as the screened dangerous target;

[0060] Determine the risk assessment quantity of different screened dangerous targets through the types of suspected dangerous behaviors of the screened dangerous targets and the similarity coefficients with different suspected dangerous behaviors;

[0061] Determine the comprehensive risk assessment quantity based on the risk assessment quantities of different screened targets, and determine whether there is a risk based on the comprehensive risk assessment quantity.

[0062] Further, determine whether there is a risk based on the comprehensive risk assessment quantity, specifically including:

[0063] When the comprehensive risk assessment quantity is not within the preset range, it is determined that there is a risk.

[0064] It can be understood that when it is determined that there is a risk through the similarity situation, directly send an alarm message and construct the action strategy of the quadruped robot using a preset strategy.

[0065] In another possible embodiment, determine that there is no risk through the similarity situation, specifically including:

[0066] Determine the similarity coefficient between the dangerous target and different specified dangerous behaviors according to the action recognition results of different dangerous targets, and identify the suspected dangerous behaviors of the dangerous target based on the similarity coefficient. Take the dangerous target with suspected dangerous behaviors as the screened dangerous target, and judge whether the number of the screened dangerous targets meets the requirements. If so, enter the next step. If not, it is determined that there is a risk;

[0067] Based on the recognition results of the suspected dangerous behaviors for screening dangerous targets, determine the quantities of different suspected dangerous behaviors. When it is determined that the quantity of suspected dangerous behaviors does not meet the requirements during the screening of dangerous targets, if so, proceed to the next step; if not, determine that there is a risk.

[0068] Determine the risk assessment quantities of different screened dangerous targets by means of the types of the suspected dangerous behaviors of the screened dangerous targets and the similarity coefficients with different suspected dangerous behaviors. When it is determined that there are screened dangerous targets with risk assessment quantities not meeting the requirements, if so, proceed to the next step; if not, determine that there is a risk.

[0069] Determine the comprehensive risk assessment quantity based on the risk assessment quantities of different screening targets, and determine whether there is a risk based on the comprehensive risk assessment quantity.

[0070] In another possible embodiment, it is determined that there is no risk through the similarity situation, which specifically includes:

[0071] Determine the similarity coefficients between the dangerous targets and different specified dangerous behaviors according to the action recognition results of different dangerous targets, and identify the suspected dangerous behaviors of the dangerous targets based on the similarity coefficients. Take the dangerous targets with suspected dangerous behaviors as the screened dangerous targets.

[0072] Based on the recognition results of the suspected dangerous behaviors of the screened dangerous targets, determine the types of the suspected dangerous behaviors of the screened dangerous targets, and determine the weight values of the suspected dangerous behaviors through the types. When it is determined that there are screened dangerous targets with the sum of the weight values of the suspected dangerous behaviors not meeting the requirements, if so, proceed to the next step; if not, determine that there is a risk.

[0073] Determine the risk assessment quantities of different screened dangerous targets by means of the types of the suspected dangerous behaviors of the screened dangerous targets and the similarity coefficients with different suspected dangerous behaviors, and determine whether the sum of the quantities of the risk assessment quantities of the screened dangerous targets meets the requirements. If so, proceed to the next step; if not, determine that there is a risk.

[0074] Screen the risk targets based on the risk assessment quantities, and determine whether the quantity of the risk targets meets the requirements. If so, proceed to the next step; if not, determine that there is a risk.

[0075] Determine the comprehensive risk assessment quantity based on the risk assessment quantities of different screening targets, and determine whether there is a risk based on the comprehensive risk assessment quantity.

[0076] S3 Based on the analysis results of the image, determine the recognition results of the target states of different dangerous targets, and determine the action strategy of the quadruped robot through the recognition results.

[0077] Further, the recognition result of the target state includes successful deterrence and no threat success but no specified dangerous behavior.

[0078] Specifically, as Figure 4 shown, the action strategy of the quadruped robot is determined through the recognition result, which specifically includes:

[0079] The dangerous targets are classified into different types through the recognition result, and the target recognition risks of different types of dangerous targets are determined according to the number of different types of dangerous targets;

[0080] Based on the target recognition risks of different types of dangerous targets, the comprehensive recognition risk value is determined, and the action strategy of the quadruped robot is determined through the comprehensive recognition risk value.

[0081] Further, determining the action strategy of the quadruped robot through the comprehensive recognition risk value specifically includes:

[0082] Based on the comprehensive recognition risk value, the matching action strategy of the quadruped robot is determined, and the action strategy of the quadruped robot is determined based on the matching action strategy.

[0083] Through the above embodiments, the present invention has the following beneficial effects:

[0084] 1. Determine whether the risk is controllable based on the matching situation of different preset dangerous behaviors of different dangerous targets, thereby realizing the overall risk assessment of dangerous targets from the perspective of the matching results of different preset dangerous behaviors of dangerous targets, and avoiding the risk of risk expansion caused by the failure to issue early warning information in time due to uncontrollable risks.

[0085] 2. Based on the analysis result of the image, the recognition result of the target state of different dangerous targets and the action strategy of the quadruped robot are obtained, avoiding the technical problems of the inability to further achieve deterrence and timely issue early warning information caused by adopting a fixed action strategy, and ensuring the reliability of the setting of the action strategy.

[0086] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0087] The above description has been made of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The above is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. A target deterrence effect evaluation method based on a quadruped robot, characterized in that: Acquire a monitoring image of the quadruped robot, and when it is determined that there is a dangerous target with a preset dangerous behavior in the monitoring image according to the analysis result of the monitoring image, and when it is determined that the risk is controllable based on the matching of different preset dangerous behaviors of different dangerous targets, proceed to the next step; Deterring dangerous targets through the shouting function of the quadruped robot, and performing similarities between different dangerous targets and designated dangerous behaviors based on the image of the quadruped robot, and when it is determined that there is no risk through the similarities, proceeding to the next step; Based on the analysis result of the image, the recognition result of the target state of different dangerous targets is determined, and the action strategy of the quadruped robot is determined by the recognition result.

2. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The image and the monitoring image are determined based on the extraction results of the camera of the quadruped robot.

3. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The preset dangerous behaviors include holding dangerous instruments and different types of aggressive actions.

4. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: When there is no dangerous target with preset dangerous behavior, there is no need for threat handling by the quadruped robot.

5. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The risk is controllable based on the matching of different preset dangerous behaviors with different dangerous targets, including: Determining the matching results between different dangerous targets and different types of preset dangerous behaviors by matching with different preset dangerous behaviors, and determining the risk coefficients of different dangerous targets based on the matching results; According to the distance between different dangerous targets and other dangerous targets and the distance between different dangerous targets and the quadruped robot, different risk weight values ​​of dangerous targets are calculated, and the risk coefficient is corrected by using the risk weight value to obtain the corrected risk coefficient of different dangerous targets; The number of the dangerous targets is obtained, and a comprehensive risk coefficient is determined in combination with the modified risk coefficients of different dangerous targets, and whether the risk is controllable is determined based on the comprehensive risk coefficient.

6. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The risk is controllable based on the matching of different preset dangerous behaviors with different dangerous targets, including: Determine whether the number of the dangerous targets is greater than the preset target number, if so, determine that the risk is uncontrollable, if not, proceed to the next step; According to the distance between different dangerous targets and other dangerous targets and the distance between different dangerous targets and the quadruped robot, different risk weight values ​​of dangerous targets are calculated, and it is judged whether the sum of the risk weight values ​​of the dangerous targets meets the requirements. If not, it is determined that the risk is uncontrollable. If so, proceed to the next step; The matching results of different dangerous targets and different types of preset dangerous behaviors are determined by matching different preset dangerous behaviors, and the risk coefficients of different dangerous targets are determined based on the matching results, and it is judged whether the number of dangerous targets that do not meet the requirements of the risk coefficients meets the requirements. If so, proceed to the next step, if not, it is determined that the risk is uncontrollable; The risk coefficient is modified by using the risk weight value to obtain modified risk coefficients of different dangerous targets, and it is determined whether the number of dangerous targets whose modified risk coefficients do not meet the requirements meets the requirements. If so, proceed to the next step; if not, it is determined that the risk is uncontrollable; The number of the dangerous targets is obtained, and a comprehensive risk coefficient is determined in combination with the modified risk coefficients of different dangerous targets, and whether the risk is controllable is determined based on the comprehensive risk coefficient.

7. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The quadruped robot's shouting function is used to deter dangerous targets, including: The shouting command of the quadruped robot is determined according to the voice command sent down by the control platform and the preset voice command, and the dangerous target is deterred based on the shouting command.

8. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The designated dangerous behaviors include potentially dangerous changing actions and different types of aggressive actions.

9. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: The absence of risk is determined through similar situations, including: Determine the similarity coefficient between the dangerous target and different designated dangerous behaviors according to the action recognition results of different dangerous targets, and identify the suspected dangerous behaviors of the dangerous target based on the similarity coefficient, use the dangerous targets with suspected dangerous behaviors as the screened dangerous targets, and judge whether the number of the screened dangerous targets meets the requirements. If yes, proceed to the next step, if not, determine that there is a risk; Determine the number of different suspected dangerous behaviors based on the identification result of the suspected dangerous behaviors of the screening dangerous targets, and judge whether there is a screening dangerous target whose number of suspected dangerous behaviors does not meet the requirements. If so, proceed to the next step; if not, determine that there is a risk; By screening the types of suspected dangerous behaviors of dangerous targets and the similarity coefficients with different suspected dangerous behaviors, different risk assessment amounts of the dangerous targets to be screened are determined, and it is judged whether there are dangerous targets whose risk assessment amounts do not meet the requirements. If so, proceed to the next step; if not, it is determined that there is a risk; A comprehensive risk assessment amount is determined based on the risk assessment amounts of different screening targets, and whether there is a risk is determined based on the comprehensive risk assessment amount.

10. The target deterrence effect evaluation method based on a quadruped robot according to claim 1, characterized in that: Determining the action strategy of the quadruped robot according to the recognition result specifically includes: Classifying the dangerous targets into different types according to the identification results, and determining the target identification risks of different types of dangerous targets according to the number of different types of dangerous targets; A comprehensive identification risk value is determined based on target identification risks of different types of dangerous targets, and an action strategy of the quadruped robot is determined by the comprehensive identification risk value.

Images