Method and device for controlling a patrol robot, electronic equipment and storage medium

Abstract

A control method and device of a patrol robot, an electronic device and a storage medium are disclosed. The method comprises: when the patrol robot responds to a patrol task, acquiring a real-time video stream collected by the patrol robot; performing abnormal event detection according to the real-time video stream; if there is an abnormal event in the real-time video stream, generating corresponding alarm information according to the abnormal event, and controlling the patrol robot to broadcast the alarm information at the occurrence place of the abnormal event. Thus, it is not necessary for city management personnel to personally go to the scene to enforce the law, the supervision of urban areas can timely discover and manage, and the human resources and time are reduced, and the management efficiency is higher.

Description

Technical Field

[0001] This invention relates to the field of robotics, and more specifically to a control method, device, electronic equipment, and storage medium for a patrol robot. Background Technology

[0002] The existing urban governance methods mainly rely on citizens' phone calls and urban governance personnel to view real-time surveillance videos to obtain road information in various areas of the city. For example, abnormal events such as street vendors occupying roads in areas where street vending is prohibited or cars illegally parking on roads are detected. When urban governance personnel confirm that abnormal events have occurred in certain areas of the city, they usually need to go to the scene in person to enforce the law, which consumes manpower and time and has low governance efficiency. Summary of the Invention

[0003] Firstly, the main objective of this invention is to provide a control method for a patrol robot, comprising: When the patrol robot responds to a patrol mission, the real-time video stream collected by the patrol robot is acquired; Anomaly detection is performed based on the real-time video stream; If an abnormal event occurs in the real-time video stream, a corresponding alarm message is generated based on the abnormal event, and the patrol robot is controlled to broadcast the alarm message at the location where the abnormal event occurred.

[0004] Optionally, the abnormal event detection based on the real-time video stream includes: A corresponding bounding box is generated for the target object in each frame of the real-time video stream; the real-time video stream is the video stream of the region to be detected. The detection distance between the patrol robot and the target object is determined based on the target bounding box, and the presence of abnormal events in the area to be detected is determined based on the detection distance.

[0005] Optionally, determining the detection distance between the patrol robot and the target object based on the target bounding box, and determining whether there is an abnormal event in the area to be detected based on the detection distance, includes: Determine the selection range of the target box in each frame of the image; If the target box reaches the maximum selection range, calculate the detection distance between the patrol robot and the target object; The intersection point between the patrol robot and the area to be detected is determined based on the detection distance, so as to determine whether there is an abnormal event in the area to be detected based on the intersection point.

[0006] Optionally, determining the intersection point between the patrol robot and the area to be detected based on the detection distance, and then determining whether there is an abnormal event in the area to be detected based on the intersection point, includes: Based on the detection distance, a corresponding detection line is assigned between the target object and the patrol robot; Determine whether there is an intersection point between the detection line and the area to be detected; If there is an intersection point between the detection line and the area to be detected, it is determined that there is an abnormal event in the area to be detected.

[0007] Optionally, if an abnormal event exists in the real-time video stream, generating corresponding alarm information based on the abnormal event and controlling the patrol robot to broadcast the alarm information at the location of the abnormal event includes: If abnormal events exist in the real-time video stream, determine the number of abnormal events; The alarm level is determined based on the number of abnormal events, and alarm information is determined based on the alarm level, so as to control the patrol robot to broadcast the alarm information to the location where the abnormal event occurred.

[0008] Optionally, the step of generating corresponding alarm information based on the abnormal event and controlling the patrol robot to broadcast the alarm information at the location of the abnormal event if an abnormal event exists in the real-time video stream includes: When the patrol robot is detected to have re-entered the location of the abnormal event, a real-time video stream of the patrol robot re-entering the location of the abnormal event is acquired. Based on the real-time video stream of the patrol robot re-entering the location of the abnormal event, the location of the abnormal event is re-detected for abnormal events.

[0009] Optionally, after performing anomaly detection again on the location of the anomaly based on the real-time video stream of the patrol robot re-entering the location of the anomaly, the process includes: If an abnormal event occurs at the location where the abnormal event occurred, the alarm information is broadcast to the location where the abnormal event occurred again. If no abnormal event occurs at the location where the abnormal event occurred, the event at that location is determined to be normal.

[0010] Secondly, embodiments of the present invention provide a control device for a patrol robot, comprising: The acquisition module acquires the real-time video stream collected by the patrol robot when the patrol robot responds to the patrol task; The detection module is used to detect abnormal events based on the real-time video stream; The broadcasting module is used to generate corresponding alarm information based on the abnormal event if there is an abnormal event in the real-time video stream, and to control the patrol robot to broadcast the alarm information at the location where the abnormal event occurs.

[0011] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the patrol robot control method described above.

[0012] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the patrol robot control method described above.

[0013] The above-described solution of the present invention has at least the following beneficial effects: The patrol robot control method provided by this invention acquires real-time video streams collected by the patrol robot when responding to patrol tasks; detects abnormal events based on the real-time video streams; if abnormal events are found in the real-time video streams, generates corresponding alarm information based on the abnormal events, and controls the patrol robot to broadcast the alarm information at the location of the abnormal event. This eliminates the need for urban management personnel to personally visit the site for enforcement, enabling timely detection and management of urban areas, reducing manpower and time, and improving management efficiency. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0015] Figure 1 This is a network architecture diagram of the control method for a patrol robot provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the overall process of the control method for the patrol robot provided in an embodiment of the present invention; Figure 3 Example diagram of the control method for the patrol robot provided in the embodiment of the present invention; Figure 4 Another example diagram of the control method for the patrol robot provided in the embodiments of the present invention; Figure 5 This is a structural block diagram of the control device for a patrol robot provided in an embodiment of the present invention; Figure 6 This is a structural block diagram of an electronic device provided in an embodiment of the present invention.

[0016] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] The terms "first," "second," and "third," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects and not to describe a specific order. Furthermore, the term "comprising," and any variations thereof, is intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0019] First, let's take a look at the relevant accompanying drawings to illustrate the scheme and network architecture of the embodiments of this application.

[0020] like Figure 1 As shown, the network architecture involves servers, a display platform, and patrol robots deployed in various urban areas. The server can be understood as being used for route planning and target detection of the patrol robots. The patrol robots can collect patrol images or video streams from various urban areas through their cameras. The display platform receives and displays the patrol information uploaded by the server, facilitating the analysis and display of this information for regulatory personnel. This eliminates the need for city management personnel to physically visit enforcement sites, enabling timely detection and control of urban issues, reducing manpower and time, and increasing efficiency.

[0021] like Figure 2 As shown, a specific embodiment of the present invention provides a control method for a patrol robot, including: S10. When the patrol robot responds to the patrol task, acquire the real-time video stream collected by the patrol robot.

[0022] In this embodiment, the patrol task can be a pre-set task, which includes a patrol map. The patrol map can be a map obtained by the patrol robot during a historical patrol period. The patrol map contains multiple path turning points, which can form the patrol robot's cyclic patrol trajectory, allowing the patrol robot to perform cyclic patrols based on the patrol map. Therefore, when the patrol robot responds to a patrol task, it can move to the corresponding patrol map and patrol according to the patrol trajectory. When it reaches the corresponding path turning point, the patrol robot can turn and continue patrolling. Optionally, the patrol robot can acquire images or video streams through a panoramic camera. Therefore, during the patrol robot's patrol at various locations, the corresponding video streams can be acquired in real time and uploaded to the server.

[0023] Understandably, patrol robots can perform real-time positioning during movement, thus determining their location within the real-time video stream. Therefore, the patrol robot can determine its location information via a GPS module. Consequently, when responding to a patrol task sent by the server, the patrol robot can send its current location information to the server. The server then determines whether the current location information includes a pre-defined detection area. If the current location information does not include the detection area, it means the patrol robot has not entered the detection area; if it does, it means the patrol robot has entered the detection area. For example, such as... Figure 3 As shown, Figure 3 Points a, b, and c represent the patrol trajectory of the patrol robot. The path from a to b includes the area to be detected, while the path from b to c does not.

[0024] S20. Detect abnormal events based on the real-time video stream.

[0025] In this embodiment, abnormal event detection can be performed on any area, or on a pre-defined area to be detected. For example, the pre-defined area to be detected could be a no-stall area, a non-business area outside a store, or a no-parking area. Abnormal events can include whether stalls are set up in no-stall areas, whether businesses are operating in non-business areas outside stores, or whether parking occurs in no-parking areas. Therefore, when the patrol robot reaches the area to be detected, the real-time video stream collected by the patrol robot can perform target detection and tracking on any object within its shooting range, and then determine whether there are abnormal events in the area to be detected based on the target detection and tracking results. It is understood that the patrol robot can collect video streams from different locations within the area to be detected in real time during its movement, thereby determining whether there are abnormal events in the area to be detected. For example... Figure 2In the path trajectory from a to b, when the patrol robot moves from a to b, the real-time video stream collected is the video stream corresponding to the path from a to b. Therefore, abnormal events can be detected in the area to be detected in the path from a to b.

[0026] Specifically, the above-mentioned abnormal event detection based on real-time video stream includes: generating corresponding target bounding boxes for target objects in each frame of the real-time video stream; the real-time video stream being the video stream of the area to be detected; determining the detection distance between the patrol robot and the target object based on the target bounding boxes; and determining whether there are abnormal events in the area to be detected based on the detection distance.

[0027] In this context, each frame of the real-time video stream represents a continuous image frame. The real-time video stream can contain different objects, and pre-defined target objects can be detected. Target objects can be objects existing in the area to be detected, such as stalls in areas where street vending is prohibited, fruit racks in non-commercial areas outside shops, and cars in areas where parking is prohibited. The bounding boxes can be obtained by detecting target objects using a pre-trained target tracking model, which can be a MOT (Multiple Object Tracking) model. The detection distance represents the minimum distance between the patrol robot and the target object. Since the patrol robot is in a moving state during patrol, it can detect and track target objects. It can be understood that when the patrol robot captures a target object in the area to be detected, each frame of the video stream can generate a corresponding bounding box. When the detection distance between the patrol robot and the target object is minimized, the presence of an abnormal event can be determined based on the detection distance. For example, in... Figure 4 In the diagram, the detection range of the patrol robot is the triangular area. When the patrol robot moves to the middle of the detection area, it can capture images of the target object (small square). Thus, when the patrol robot is in the middle of the detection area, it can start detecting and tracking the target object. When it reaches a position perpendicular to the target object, the detection distance can be calculated.

[0028] Furthermore, the above-mentioned determination of the detection distance between the patrol robot and the target object based on the target bounding box, and determination of whether there is an abnormal event in the area to be detected based on the detection distance, includes: determining the selection range of the target bounding box in each frame of the image; calculating the detection distance between the patrol robot and the target object when the target bounding box reaches the maximum selection range; determining the intersection point between the patrol robot and the area to be detected based on the detection distance, so as to determine whether there is an abnormal event in the area to be detected based on the intersection point.

[0029] In this embodiment, the target box generated by tracking and detecting the target object is at its smallest when the target object appears within the shooting range of the patrol robot. The target box is at its largest when the patrol robot moves to a position perpendicular to the target object. Therefore, when the target box is at its largest, the detection distance between the patrol robot and the target object can be calculated using infrared detection or a camera. This detection distance determines whether there is an intersection point between the patrol robot and the area to be detected. If an intersection point exists, it indicates an abnormal event in the area to be detected, meaning the target object is within the area. If no intersection point exists, it indicates no abnormal event in the area to be detected, meaning the target object is outside the area. For example, when a stall appears in a prohibited area, the detection distance between the stall and the patrol robot can be detected. When the patrol robot and the stall form a perpendicular relationship, the detection distance between them is minimized. Therefore, the intersection point between the patrol robot and the target area can be determined when the detection distance is minimized.

[0030] Furthermore, the above-mentioned determination of the intersection point between the patrol robot and the area to be detected based on the detection distance, so as to determine whether there is an abnormal event in the area to be detected based on the intersection point, includes: assigning a corresponding detection line between the target object and the patrol robot based on the detection distance; determining whether there is an intersection point between the detection line and the area to be detected; if there is an intersection point between the detection line and the area to be detected, then determining that there is an abnormal event in the area to be detected.

[0031] The detection line represents the vertical line between the target object and the patrol robot. When the target box reaches the maximum selection range, the detection distance is minimized and the corresponding detection line is generated. Therefore, if there is an intersection between the detection line and the area to be detected, it means that the target object is within the area to be detected, and thus it can be determined that there is an abnormal event in the area to be detected, making the patrol robot's detection of whether there is an abnormal event in the area to be detected more accurate. It can be understood that after determining that there is an abnormal event in the area to be detected, the area to be detected can be determined as the location of the abnormal event.

[0032] In an optional embodiment, when there is an intersection between the detection line and the area to be detected, the abnormal event can be judged based on the current patrol time. If the current patrol time is within a preset time period, an alarm can be issued for the abnormal event; if the current patrol time is not within the preset time period, no alarm can be issued for the abnormal event. Optionally, the preset time period may include a start time and an end time. If the current patrol time is within the preset time period, the difference between the current patrol time and the end time of the preset time period can be calculated. If the difference between the current patrol time and the end time of the preset time period is less than a predetermined difference, no alarm can be issued for the abnormal event. If the difference between the current patrol time and the end time of the preset time period is greater than a predetermined difference, an alarm can be triggered for abnormal events. For example, if the preset time period is set from 9:00 AM to 8:00 PM, an alarm can be triggered for abnormal events when the current patrol time is 4:00 PM; no alarm can be triggered for abnormal events when the current patrol time is 8:00 PM; if the current patrol time is 7:30 PM, the difference between the current patrol time and the end time of the preset time period can be determined to be 30 minutes. If the predetermined difference is set to 40 minutes, no alarm can be triggered for abnormal events. This ensures that alarms are dynamically triggered for different scenarios during the patrol process.

[0033] S30. If there is an abnormal event in the real-time video stream, generate corresponding alarm information based on the abnormal event and control the patrol robot to broadcast the alarm information at the location where the abnormal event occurred.

[0034] In this embodiment, the alarm information can be pre-set, and different alarm information can be triggered according to different abnormal events. Therefore, after determining that the area to be detected is the location of the abnormal event, the corresponding alarm information can be triggered according to the number of abnormal events, and the corresponding alarm information can be broadcast to the location of the abnormal event. Optionally, the alarm information can include voice broadcast, warning music, etc. By sending the alarm information to the patrol robot, the patrol robot can issue an alarm in the area to be detected, and then communicate and warn the parties involved in the area to be detected. This makes the supervision more timely and the efficiency of urban governance higher.

[0035] Specifically, if an abnormal event exists in the real-time video stream, a corresponding alarm message is generated based on the abnormal event, and the patrol robot is controlled to broadcast the alarm message at the location where the abnormal event occurred. This includes: if an abnormal event exists in the real-time video stream, determining the number of abnormal events; determining the corresponding alarm level based on the number of abnormal events; and determining the alarm message based on the alarm level, so as to control the patrol robot to broadcast the alarm message at the location where the abnormal event occurred.

[0036] Specifically, when an abnormal event is determined to exist in the area to be detected, the frequency of the abnormal event can be determined. When the frequency of the abnormal event is the first value, the alarm level is the lowest level, and the alarm information is the alarm information corresponding to the lowest level. When the frequency of the abnormal event is the second value, the alarm level is the medium level, and the alarm information is the alarm information corresponding to the medium level. When the frequency of the abnormal event is the third value, the alarm level is the highest level, and the alarm information is the alarm information corresponding to the highest level. Optionally, the first, second, and third values ​​can be pre-defined. The first value can be 1 time, the second value can be 2 times, and the third value can be 3 times. For example, when patrolling a no-stall area in a loop, if a stall is found during the first patrol, it can be determined that the corresponding abnormal event in the no-stall area is 1 time, and the alarm level is the lowest level. If a stall is found during the second patrol, it can be determined that the corresponding abnormal event in the no-stall area is 2 times, and the alarm level is the medium level. If a stall is found during the third patrol, it can be determined that the corresponding abnormal event in the no-stall area is 3 times, and the alarm level is the highest level.

[0037] Optionally, when the alarm level is the highest level, a warning message can be output to the display platform to inform the supervisors, so that the supervisors can control the patrol robot to make remote manual alarms, or obtain the contact information of the parties in the area to be detected based on the location information of the area to be detected and communicate by phone, thereby ensuring that abnormal events can be detected and managed in a timely manner in various areas of the city to reduce human resources.

[0038] In a preferred embodiment, after generating corresponding alarm information based on the abnormal event and controlling the patrol robot to broadcast the alarm information at the location of the abnormal event if an abnormal event exists in the real-time video stream, the process includes: when it is detected that the patrol robot has re-entered the location of the abnormal event, acquiring the real-time video stream of the patrol robot re-entering the location of the abnormal event; and performing abnormal event detection again at the location of the abnormal event based on the real-time video stream of the patrol robot re-entering the location of the abnormal event.

[0039] In this embodiment, the patrol task may include a number of cyclic patrols, which may be preset, for example, 2 times. When the patrol robot completes the first patrol task, it may re-enter the area to be detected when performing the second patrol task. Therefore, when the patrol robot re-enters the area to be detected, abnormal events may be detected again in the area to be detected, and it may determine whether there are abnormal events according to the method provided in the above embodiment. If there are abnormal events, the number of times the previous patrol task was accumulated to determine the number of abnormal events in the area to be detected, and then different alarm information is executed for the area to be detected.

[0040] Furthermore, the above-mentioned actions, based on the real-time video stream of the patrol robot re-entering the location of the abnormal event and performing abnormal event detection again, include: if an abnormal event exists at the location of the abnormal event, broadcasting an alarm message to the location of the abnormal event again; if no abnormal event exists at the location of the abnormal event, determining that the event at the location of the abnormal event has ended.

[0041] When broadcasting alarm information, a predetermined broadcast frequency can be used. By using a predetermined broadcast frequency, the parties involved at the location of the abnormal event can be reminded, ensuring that they can react quickly and end the abnormal event. It is understood that the predetermined broadcast frequency can be broadcast at certain intervals, such as once every 2 seconds, thus ensuring more efficient reminders to the parties involved. In a preferred embodiment, the continuous violation time can be generated based on the time when the patrol robot last broadcasted alarm information at the location of the abnormal event and the current time. The video streams captured at the location of the abnormal event at the last and current times are then reported together to the relevant departments for processing.

[0042] In an optional embodiment, when the patrol robot re-patrols the area to be detected, if a target object is detected again in the area, the bounding box corresponding to the target object can be compared with the bounding box corresponding to the first patrol task. If the similarity between the bounding box and the bounding box corresponding to the first patrol task is greater than or equal to 50%, the alarm level can be set to the highest level; if the similarity between the bounding box and the bounding box corresponding to the first patrol task is less than 50%, the alarm level can be set to the medium level. This ensures higher governance efficiency for the area to be detected.

[0043] The patrol robot control method provided by this invention acquires real-time video streams collected by the patrol robot when it responds to patrol tasks; detects abnormal events based on the real-time video streams; if abnormal events are found in the real-time video streams, corresponding alarm information is generated based on the abnormal events, and the patrol robot is controlled to broadcast the alarm information at the location where the abnormal events occurred. This eliminates the need for urban management personnel to personally visit the site for enforcement, enabling timely detection and management of urban areas, reducing manpower and time, and improving management efficiency.

[0044] like Figure 5 As shown, an embodiment of the present invention provides a control device for a patrol robot, comprising: The acquisition module acquires the real-time video stream collected by the patrol robot when it responds to a patrol task. The detection module is used to detect abnormal events based on the real-time video stream; The broadcast module is used to generate corresponding alarm information based on abnormal events in the real-time video stream, and control the patrol robot to broadcast the alarm information at the location where the abnormal event occurred.

[0045] The control device for the patrol robot provided by this invention acquires real-time video streams collected by the patrol robot when it responds to patrol tasks; detects abnormal events based on the real-time video streams; if abnormal events are found in the real-time video streams, it generates corresponding alarm information based on the abnormal events and controls the patrol robot to broadcast the alarm information at the location where the abnormal events occurred. This eliminates the need for urban management personnel to personally visit the site for enforcement, enabling timely detection and management of urban areas, reducing manpower and time, and improving management efficiency.

[0046] It should be noted that the control device 10 of the patrol robot provided in the specific embodiment of the present invention is a device corresponding to the control method of the patrol robot described above. All embodiments of the control method of the patrol robot are applicable to the control device 10 of the patrol robot. Each embodiment of the control device 10 of the patrol robot has a corresponding module corresponding to the steps in the control method of the patrol robot, which can achieve the same or similar beneficial effects. In order to avoid excessive repetition, each module in the control device 2 of the patrol robot will not be described in detail here.

[0047] like Figure 6 As shown, a specific embodiment of the present invention also provides an electronic device 20, including a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201. When the processor 201 executes the computer program, it implements the steps of the above-described patrol robot control method.

[0048] Specifically, processor 201 calls the computer program stored in memory 202 and performs the following steps: When the patrol robot responds to a patrol mission, acquire the real-time video stream collected by the patrol robot; Detect abnormal events based on real-time video streams; If an abnormal event is detected in the real-time video stream, a corresponding alarm message is generated based on the abnormal event, and the patrol robot is controlled to broadcast the alarm message at the location where the abnormal event occurred.

[0049] Optionally, the abnormal event detection performed by the processor 201 based on the real-time video stream includes: Generate corresponding bounding boxes for target objects in each frame of the real-time video stream; the real-time video stream is the video stream of the region to be detected. The detection distance between the patrol robot and the target object is determined based on the target bounding box, and the presence of abnormal events in the area to be detected is determined based on the detection distance.

[0050] Optionally, the processor 201 performs the following: determining the detection distance between the patrol robot and the target object based on the target bounding box, and determining whether there are abnormal events in the area to be detected based on the detection distance, including: Determine the selection range of the target box in each frame of the image; If the target box reaches the maximum selection range, calculate the detection distance between the patrol robot and the target object; The intersection point between the patrol robot and the area to be inspected is determined based on the detection distance, and the presence of abnormal events in the area to be inspected is determined based on the intersection point.

[0051] Optionally, the processor 201 performs the following: determining the intersection point between the patrol robot and the area to be detected based on the detection distance, so as to determine whether there is an abnormal event in the area to be detected based on the intersection point, including: Based on the detection distance, assign a corresponding detection line between the target object and the patrol robot; Determine whether there is an intersection between the detection line and the area to be detected; If there is an intersection between the detection line and the area to be detected, it is determined that there is an abnormal event in the area to be detected.

[0052] Optionally, if an abnormal event exists in the real-time video stream, the processor 201 generates corresponding alarm information based on the abnormal event and controls the patrol robot to broadcast the alarm information at the location where the abnormal event occurred, including: If abnormal events exist in the real-time video stream, determine the number of abnormal events; The alarm level is determined based on the number of abnormal events, and the alarm information is determined based on the alarm level to control the patrol robot to broadcast the alarm information to the location where the abnormal event occurred.

[0053] Optionally, if an abnormal event exists in the real-time video stream, the processor 201 generates corresponding alarm information based on the abnormal event and controls the patrol robot to broadcast the alarm information at the location where the abnormal event occurred, including: When the patrol robot is detected to have re-entered the location of the abnormal event, acquire a real-time video stream of the patrol robot re-entering the location of the abnormal event; Based on the real-time video stream of the patrol robot re-entering the location of the abnormal event, the location of the abnormal event is re-detected for abnormal events.

[0054] Optionally, after the processor 201 performs real-time video streaming on the location of the abnormal event again based on the patrol robot re-entering the location of the abnormal event, and then performs abnormal event detection again on the location of the abnormal event, the following steps are included: If an abnormal event occurs at the location where the abnormal event occurred, broadcast the alarm information to the location where the abnormal event occurred again; If no abnormal event occurs at the location where the abnormal event occurred, the event that determines the location of the abnormal event ends.

[0055] That is, in a specific embodiment of the present invention, when the processor 201 of the electronic device 20 executes the computer program, it implements the steps of the above-mentioned patrol robot control method, thereby eliminating the need for urban management personnel to personally go to the scene to enforce the law, enabling timely detection and management of urban areas, reducing human resources and time, and improving management efficiency.

[0056] It should be noted that since the processor 201 of the electronic device 20 implements the steps of the above-described patrol robot control method when executing the computer program, all embodiments of the above-described patrol robot control method are applicable to the electronic device 20 and can achieve the same or similar beneficial effects.

[0057] The computer-readable storage medium provided in this embodiment of the invention stores a computer program. When the computer program is executed by a processor, it implements the various processes of the control method for the patrol robot or the control method for the application-side patrol robot provided in this embodiment of the invention, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0058] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0059] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0060] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made under the concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A control method for a patrol robot, characterized in that, include: When the patrol robot responds to a patrol mission, the real-time video stream collected by the patrol robot is acquired; Anomaly detection is performed based on the real-time video stream; If an abnormal event occurs in the real-time video stream, a corresponding alarm message is generated based on the abnormal event, and the patrol robot is controlled to broadcast the alarm message at the location where the abnormal event occurred. The abnormal event detection based on the real-time video stream includes: A corresponding bounding box is generated for the target object in each frame of the real-time video stream; the real-time video stream is the video stream of the region to be detected. The detection distance between the patrol robot and the target object is determined based on the target bounding box, and the presence of abnormal events in the area to be detected is determined based on the detection distance. The step of determining the detection distance between the patrol robot and the target object based on the target bounding box, and determining whether there are abnormal events in the area to be detected based on the detection distance, includes: Determine the selection range of the target box in each frame of the image; If the target box reaches the maximum selection range, calculate the detection distance between the patrol robot and the target object; The intersection point between the patrol robot and the area to be detected is determined based on the detection distance, so as to determine whether there is an abnormal event in the area to be detected based on the intersection point.

2. The control method for the patrol robot according to claim 1, characterized in that, The step of determining the intersection point between the patrol robot and the area to be detected based on the detection distance, and then determining whether there is an abnormal event in the area to be detected based on the intersection point, includes: Based on the detection distance, a corresponding detection line is assigned between the target object and the patrol robot; Determine whether there is an intersection point between the detection line and the area to be detected; If there is an intersection point between the detection line and the area to be detected, it is determined that there is an abnormal event in the area to be detected.

3. The control method for the patrol robot according to claim 1, characterized in that, If an abnormal event exists in the real-time video stream, a corresponding alarm message is generated based on the abnormal event, and the patrol robot is controlled to broadcast the alarm message at the location where the abnormal event occurred, including: If abnormal events exist in the real-time video stream, determine the number of abnormal events; The alarm level is determined based on the number of abnormal events, and alarm information is determined based on the alarm level, so as to control the patrol robot to broadcast the alarm information to the location where the abnormal event occurred.

4. The control method for the patrol robot according to claim 1, characterized in that, If an abnormal event exists in the real-time video stream, then generating corresponding alarm information based on the abnormal event and controlling the patrol robot to broadcast the alarm information at the location of the abnormal event includes: When the patrol robot is detected to have re-entered the location of the abnormal event, a real-time video stream of the patrol robot re-entering the location of the abnormal event is acquired. Based on the real-time video stream of the patrol robot re-entering the location of the abnormal event, the location of the abnormal event is re-detected for abnormal events.

5. The control method for the patrol robot according to claim 4, characterized in that, The step of re-detecting the abnormal event at the location of the abnormal event based on the real-time video stream of the patrol robot re-entering the location of the abnormal event includes: If an abnormal event occurs at the location where the abnormal event occurred, the alarm information is broadcast to the location where the abnormal event occurred again. If no abnormal event occurs at the location where the abnormal event occurred, the event at the location where the abnormal event occurred ends.

6. A control device for a patrol robot, characterized in that, include: The acquisition module acquires the real-time video stream collected by the patrol robot when the patrol robot responds to the patrol task; The detection module is used to detect abnormal events based on the real-time video stream; The broadcasting module is used to generate corresponding alarm information based on the abnormal event if there is an abnormal event in the real-time video stream, and to control the patrol robot to broadcast the alarm information at the location where the abnormal event occurs. The abnormal event detection based on the real-time video stream includes: A corresponding bounding box is generated for the target object in each frame of the real-time video stream; the real-time video stream is the video stream of the region to be detected. The detection distance between the patrol robot and the target object is determined based on the target bounding box, and the presence of abnormal events in the area to be detected is determined based on the detection distance. The step of determining the detection distance between the patrol robot and the target object based on the target bounding box, and determining whether there are abnormal events in the area to be detected based on the detection distance, includes: Determine the selection range of the target box in each frame of the image; If the target box reaches the maximum selection range, calculate the detection distance between the patrol robot and the target object; The intersection point between the patrol robot and the area to be detected is determined based on the detection distance, so as to determine whether there is an abnormal event in the area to be detected based on the intersection point.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the control method for the patrol robot as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the control method for the patrol robot as described in any one of claims 1 to 5.

Images