Patrolling security robot

Abstract

To provide a patrolling security robot capable of performing appropriate security actions according to the security state of an adjacent area.SOLUTION: A patrolling security robot is a patrolling security robot that moves around a predetermined patrol area, and includes: an acquisition unit that acquires a security state of an adjacent area adjacent to the patrol area; a determination unit that determines a security action to be performed while moving in the patrol area based on the security state; and an execution unit that executes the determined security action.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a patrol security robot. [Background technology]

[0002] There are known robots that patrol a predetermined area and perform security operations. Patent Document 1 describes a security robot that, when it detects a person, determines whether the area it is patrolling is in a security state or a security-free state, and if it is in a security state, determines that the person is an intruder. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 4057324 Summary of the Invention [Problem to be solved by the invention]

[0004] There is a need to take appropriate security measures by considering the security status of areas adjacent to the area being patrolled. For example, there is a need to strengthen surveillance of the patrolled area when the security status of the adjacent area is in alert mode (monitoring).

[0005] An object of the present invention is to provide a patrol security robot that can perform appropriate security operations depending on the security situation in the adjacent area. [Means for solving the problem]

[0006] A patrol security robot according to an embodiment of the present invention is a patrol security robot that moves within a predetermined patrol area, and has an acquisition unit that acquires the security status of adjacent areas adjacent to the patrol area, a decision unit that determines security actions to be performed while moving within the patrol area based on the security status, and an execution unit that executes the determined security actions.

[0007] Preferably, the determination unit determines the detection range for detecting an abnormality or an object during security operation based on the security status of an adjacent area.

[0008] In addition, the security operation is an operation to detect abnormalities or objects in the patrol area, and it is preferable that the determination unit determines at least one of the detection target for abnormalities or objects in the security operation, the detection conditions for the abnormalities or objects, and the destination to report when an abnormality or object is detected, based on the security status of an adjacent area.

[0009] In addition, it is preferable that the patrol security robot further has an output unit that outputs an audio warning while moving through the patrol area, and the determination unit determines at least one of the content and volume of the warning during security operation based on the security status of adjacent areas.

[0010] Preferably, the determination unit further determines the security operation based on a scheduled time when the security status of the adjacent area will be changed.

[0011] Preferably, the acquisition unit acquires, as the security state, either an alert mode in which an alarm is output when a person is present in the adjacent area, or a release mode in which an alarm is not output. [Effects of the Invention]

[0012] The patrol security robot according to the present invention can perform security operations taking into account the security status of adjacent areas. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a schematic configuration diagram of a security system 1. FIG. [Figure 2] FIG. 2 is a functional block diagram of the patrol security robot 2. [Figure 3] FIG. 10 is a diagram showing the data structure of a security operation table T1. [Figure 4] FIG. 10 is a flowchart showing the flow of security processing. [Figure 5]FIG. 10 is a flowchart showing the flow of audio output processing. DETAILED DESCRIPTION OF THE INVENTION

[0014] Various embodiments of the present invention will be described below with reference to the drawings. Please note that the technical scope of the present invention is not limited to these embodiments, but extends to the inventions set forth in the claims and their equivalents.

[0015] FIG. 1 is a schematic diagram of a security system 1 according to an embodiment of the present invention. The security system 1 includes a patrol security robot 2, a security device 3, a security guard terminal 4, and a monitoring terminal 5. The patrol security robot 2 is placed in a patrol area A1 of a facility. The security device 3 is placed in an adjacent area A2 adjacent to the patrol area A1 via an entrance / exit E. The security guard terminal 4 is carried by a security guard in the patrol area A1, the adjacent area A2, or a security room (not shown). The monitoring terminal 5 is placed in a monitoring center. The monitoring center may be located inside or outside the facility. The patrol security robot 2, the security device 3, the security guard terminal 4, and the monitoring terminal 5 communicate with each other via a network N, which may be the Internet or an intranet.

[0016] The adjacent area A2 being adjacent to the patrolled area A1 means that there are no areas guarded by other security devices between the patrolled area A1 and the adjacent area A2. Therefore, the patrolled area A1 and the adjacent area A2 do not have to be spatially connected. In the example shown in FIG. 1, the patrolled area A1 and the adjacent area A2 are spatially connected via an entrance / exit E provided in the wall W, but the entrance / exit E does not have to be provided in the wall W. For example, the patrolled area A1 is a common area of ​​a facility, and the adjacent area A2 is a private area of ​​the facility. The patrolled area A1 may be outdoors, and the adjacent area A2 may be indoors. Note that in FIG. 1, there are no physical restrictions by doors or windows on entry and exit through the entrance / exit E, but physical restrictions by doors or windows may also be provided.

[0017] The patrol security robot 2 patrols the patrol area A1 and performs security operations, which are operations for detecting abnormalities in the patrol area.

[0018] The security device 3 issues an alarm based on the presence or absence of a person in the adjacent area A2. The security device 3 is set to either an alert mode or a deactivation mode by operation of a user in the adjacent area A2. If a person is present in the adjacent area A2 while the security device 3 is set to the alert mode, the security device 3 issues an alarm to the monitoring terminal 5. Whether a person is present in the adjacent area A2 may be determined by a sensor (not shown) arranged in the adjacent area A2.

[0019] The security guard terminal 4 is an information processing terminal equipped with a memory and a processor. For example, the security guard terminal 4 is a smartphone, mobile phone, tablet PC (Personal Computer), etc. carried by the security guard. If the patrol security robot 2 detects an abnormality, the security guard terminal 4 notifies the security guard of the abnormality by sound or image.

[0020] The monitoring terminal 5 is an information processing terminal equipped with a memory and a processor. For example, the monitoring terminal 5 is a server, PC, smartphone, mobile phone, tablet PC, etc. When the patrol security robot 2 detects an abnormality, the monitoring terminal 5 notifies the supervisor at the monitoring center of the abnormality by sound or image.

[0021] 2 is a functional block diagram of the patrol security robot 2. The patrol security robot 2 has a memory 21, a communication interface 22, a sensor 23, tires 24, a tire drive unit 25, a speaker 26, and a processor 27.

[0022] The memory 21 is an example of a storage unit and stores data and programs. The memory 21 is a semiconductor memory, a magnetic disk, or the like. The memory 21 stores an operating system program, a driver program, an application program, data, and the like used for processing by the processor 27. Programs may be installed into the memory 21 from a computer-readable, non-transitory, portable storage medium such as a CD-ROM (Compact Disc Read Only Memory).

[0023] The memory 21 stores a security operation table T1, which will be described later, as data.

[0024] The communication interface 22 enables the patrol security robot 2 to communicate with other devices. For example, the communication interface 22 is an interface for wireless communication such as mobile communication or wireless LAN (Local Area Network). The communication interface 22 transmits data supplied from the processor 27 to other devices, and supplies data received from other devices to the processor 27.

[0025] The sensor 23 generates information about the environment in the vicinity of the patrol security robot 2. For example, the sensor 23 may include a camera that captures an image of the vicinity of the patrol security robot 2. The sensor 23 may include a thermal imaging camera that generates information about heat sources in the vicinity of the patrol security robot 2. The sensor 23 may include an acoustic sensor such as a microphone that generates information about sounds in the vicinity of the patrol security robot 2. The sensor 23 may also include other sensors that generate information about the temperature, humidity, vibration, illuminance, etc. in the vicinity of the patrol security robot 2. The sensor 23 supplies the generated information about the environment in the vicinity of the patrol security robot 2 to the processor 27.

[0026] The tires 24 are disposed on the bottom surface of the patrol security robot 2 and rotate to allow the patrol security robot 2 to move.

[0027] The tire driving unit 25 drives the tire 24. The tire driving unit 25 includes a motor and a control circuit. The motor is connected to the tire 24. The control circuit controls the motor in accordance with instructions from the processor 27 to rotate the tire 24.

[0028] The speaker 26 outputs sound based on the audio signal supplied from the processor 27 .

[0029] The processor 27 comprehensively controls the operation of the patrol security robot 2. For example, the processor 27 is at least one CPU (Central Processing Unit). The processor 27 may be an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), or the like. The processor 27 executes processing based on a program stored in the memory 21.

[0030] The processor 27 has an acquisition unit 271, a determination unit 272, an execution unit 273, and an output unit 274. Each of these units is a functional module realized by the processor 27 executing a program. Each of these units may be implemented in the patrol security robot 2 as a dedicated processing circuit.

[0031] FIG. 3 is a diagram showing the data structure of a security operation table T1 stored in the memory 21. The security operation table T1 stores information about multiple security operations performed by the patrol security robot 2. The multiple security operations are operations for detecting specific abnormalities in the patrol area and identifying specific objects present in the patrol area. In the example shown in FIG. 3, the multiple security operations include detecting abnormalities such as detecting loitering individuals, detecting abnormal behavior, detecting abnormal sounds, detecting heat sources, identifying abandoned objects, and identifying specific individuals. Hereinafter, "detection" and "identification" will be collectively referred to as "detection." Information about each security operation includes the detection target (the type of abnormality or object to be detected), the detection conditions, and the notification destination when an abnormality is detected. For the multiple security operations, at least one of the detection target, the detection conditions, and the notification destination when an abnormality is detected differs depending on the security status of the adjacent area. The (type of) abnormalities or objects to be detected may be set so that more types of abnormalities or objects are detected when the adjacent area is in alert mode than when the adjacent area is in deactivated mode. Furthermore, the detection conditions may be set so that a notification is more likely to be sent when the adjacent area is in alert mode than when the adjacent area is in deactivation mode. Furthermore, the notification destinations upon detection may be set so that when the adjacent area is in alert mode, the notification is sent to a notification destination with a supervisor with higher security command authority than when the adjacent area is in deactivation mode. In this embodiment, the supervisor of the monitoring terminal 5 in the monitoring center has higher security command authority than the security guard who owns the security guard terminal 4. Furthermore, the notification destinations upon detection may be set so that when the adjacent area is in alert mode, the notification is sent to more notification destinations than when the adjacent area is in deactivation mode. For example, when the adjacent area is in deactivation mode, the notification may be sent to the security guard terminal 4 carried by the security guard near the patrol security robot 2, whereas when the adjacent area is in alert mode, the notification may be sent to all security guard terminals 4, or to both the security guard terminal 4 and the monitoring terminal 5.

[0032] In the example shown in FIG. 3, the patrol security robot 2 detects a person staying in the same location in the facility as part of its security operation. A person staying in the same location in the facility for a predetermined period of time or more is a person who has been staying in the same location for a predetermined period of time or more. The detection conditions for detecting a person staying in the same location differ depending on the security status of the adjacent area. In the example shown in FIG. 3, the detection condition when the adjacent area is in alert mode is that the person has been staying in the same location for 10 minutes or more, and when the adjacent area is in de-altered mode, the detection condition is that the person has been staying in the same location for 20 minutes or more. In addition, the notification destination when a person staying in the same location is detected differs depending on the security status of the adjacent area. In the example shown in FIG. 3, the notification destination when the adjacent area is in alert mode is the security guard terminal 4, and the notification destination when the adjacent area is in alert mode is the monitoring terminal 5.

[0033] In the example shown in FIG. 3, the patrol security robot 2 detects abnormal behavior as a security operation. The abnormal behavior is a behavior that is specified in advance. The abnormal behavior to be detected differs depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the abnormal behavior is staying, and when the adjacent area is in release mode, the abnormal behavior is people falling and fighting. In addition, when abnormal behavior is detected, the notification destination differs depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the notification destination is the security guard terminal 4, and when the adjacent area is in alert mode, the notification destination is the monitoring terminal 5.

[0034] Furthermore, the patrol security robot 2 detects abnormal sounds as part of its security operation. The abnormal sounds are pre-specified sounds. The abnormal sounds to be detected differ depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the abnormal sounds are screams and conversations, and when the adjacent area is in de-alerted mode, the abnormal sounds are screams. Furthermore, the detection conditions for abnormal sounds differ depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the detection condition is that the volume of the sound is 70 dB or more, and when the adjacent area is in de-alerted mode, the detection condition is that the volume of the sound is 90 dB or more. Furthermore, when abnormal behavior is detected, the notification destination differs depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the notification destination is the security guard terminal 4, and when the adjacent area is in alert mode, the notification destination is the monitoring terminal 5.

[0035] Furthermore, the patrol security robot 2 detects heat sources as part of its security operations. A heat source is an object left in the facility whose surface temperature is equal to or higher than a predetermined value. When a heat source is detected, the notification destination differs depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the notification destination is the security guard terminal 4, and when the adjacent area is in alert mode, the notification destination is the monitoring terminal 5.

[0036] The patrol security robot 2 also detects abandoned objects as part of its security operations. An abandoned object is an object that has been left in the same location in the facility for a predetermined period of time or more. The abandoned objects that are the target of detection vary depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the abandoned objects are bags and garbage bags, and when the adjacent area is in de-altered mode, the abandoned objects are bags. The condition for detecting an abandoned object is that the object has been left there for 10 minutes or more. In addition, the destination of notification when abnormal behavior is detected varies depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the destination of notification is the security guard terminal 4, and when the adjacent area is in alert mode, the destination of notification is the monitoring terminal 5.

[0037] Furthermore, the patrol security robot 2 detects a specific person as part of its security operation. A specific person is a person whose facial recognition results satisfy a predetermined condition. The specific person to be detected differs depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the specific person is a person other than a facility user, and when the adjacent area is in deactivation mode, the specific person is a specific person who has been registered in advance as a suspicious person (so-called a blacklisted person). The detection condition differs depending on the security status of the adjacent area. In the example shown in FIG. 3, when the adjacent area is in alert mode, the detection condition is that the reliability is 0.7 or higher, and when the adjacent area is in deactivation mode, the detection condition is that the reliability is 0.9 or higher. The reliability is a value indicating the likelihood that a person is a specific person, and is calculated by applying facial recognition technology to a facial image. When a specific person is detected, the security guard terminal 4 is notified.

[0038] The data in the monitoring behavior table T1 is set in advance.

[0039] 4 is a flow diagram showing an example of the flow of security processing executed by the patrol security robot 2. The security processing is executed regularly or irregularly while the patrol security robot 2 patrols its patrol area. The security processing is realized by the processor 27 executing a program and cooperating with other components of the patrol security robot 2.

[0040] First, the acquisition unit 271 acquires the security status of the adjacent area (step S11). The acquisition unit 271 receives information from the security device 3 indicating whether the adjacent area is in the security mode or the de-security mode.

[0041] Next, the determination unit 272 determines a security operation to be performed while the patrol security robot 2 moves through the patrol area based on the security status of the adjacent area (step S12). The determination unit 272 determines the security operation to be performed by obtaining the detection target, detection conditions, and notification destination associated with the obtained security status from the security operation table T1.

[0042] Next, the execution unit 273 executes the determined security operation (step S13). The execution unit 273 acquires information about the environment near the patrol security robot 2 from the sensor 23. The execution unit 273 may acquire information about the environment within a predetermined range by controlling the tire driving unit 25 to control the orientation of the patrol security robot 2 or by controlling the sensor 23. The execution unit 273 refers to the acquired information and detects the target determined in step S12 in accordance with the conditions determined in step S12.

[0043] The execution unit 273 detects a staying person from images generated by a camera included in the sensor 23. For example, the execution unit 273 acquires multiple images generated by the camera capturing images of the vicinity of the patrol security robot 2 at different times. The execution unit 273 detects a person from each of the multiple images. The person may be detected using a trained model such as a neural network or a support vector machine (SVM), or by comparing features extracted from the image with features of a pre-stored person model. If the same person is detected from multiple images and the amount of change in the position of the detected person is less than or equal to a threshold, the execution unit 273 calculates the difference in the capture times of the multiple images as the person's staying time. Whether the person is the same or not may be determined by comparing the feature values ​​of the person. The position of the person may be calculated based on the position of the patrol security robot 2 when the image was captured, the shooting direction of the camera, the size and position of the person depicted in the image, etc. If the calculated staying time satisfies the detection condition determined in step S12, the execution unit 273 detects a staying person.

[0044] The execution unit 273 detects abnormal behavior from an image generated by a camera included in the sensor 23. For example, the execution unit 273 acquires an image generated by a camera capturing an image of the vicinity of the patrol security robot 2. The execution unit 273 estimates the posture of a person depicted in the image. The posture may be estimated using a model-based method or a trained model. The execution unit 273 detects abnormal behavior when the similarity between the estimated posture and a posture previously set for each abnormal behavior to be detected, determined in step S12, is equal to or greater than a threshold.

[0045] The execution unit 273 detects an abnormal sound from an audio signal generated by a microphone included in the sensor 23. For example, the execution unit 273 determines whether the similarity between the waveform or spectrum of the sound in the vicinity of the patrol security robot 2 indicated by the audio signal and the waveform or spectrum set for each abnormal sound to be detected as determined in step S12 is equal to or greater than a threshold. If the similarity is equal to or greater than the threshold, the execution unit 273 determines whether the volume of the sound in the vicinity of the patrol security robot 2 indicated by the audio signal satisfies the detection condition determined in step S12. If the volume satisfies the condition, the execution unit 273 detects an abnormal sound.

[0046] The execution unit 273 detects a heat source from a thermal image generated by a thermal imaging camera included in the sensor 23. For example, the execution unit 273 detects a heat source from a thermal image based on the difference between the temperature corresponding to each pixel constituting the thermal image and the temperature corresponding to an adjacent pixel. If the temperature of the detected heat source is equal to or higher than a threshold, the execution unit 273 detects the heat source as abnormal. The execution unit 273 may detect the heat source as abnormal based on the shape of the detected heat source, whether the heat source is moving, or the like.

[0047] The execution unit 273 detects abandoned objects from images generated by a camera included in the sensor 23. For example, the execution unit 273 acquires multiple images generated by a camera capturing images of the vicinity of the patrol security robot 2 at different times. The execution unit 273 detects objects from each of the multiple images. The objects may be detected using a trained model or may be detected using a background subtraction method or the like. When the same object is detected from the multiple images and the amount of change in the position of the detected object is equal to or less than a threshold, the execution unit 273 calculates the difference in the capture times of the multiple images as the abandoned time of the object. When the type of object is the detection target acquired in step S12 and the abandoned time satisfies the detection condition determined in step S12, the execution unit 273 detects an abandoned object.

[0048] The execution unit 273 detects a specific person from an image generated by a camera included in the sensor 23. For example, the execution unit 273 acquires an image generated by the camera capturing an image of the vicinity of the patrol security robot 2. The execution unit 273 detects a person's face from the image and acquires a reliability that the detected face is the target person acquired in step S12. The reliability may be calculated using a trained model that is a face recognition model, or may be calculated by comparing the face detected from the image with a pre-stored face image of the person. The execution unit 273 detects a specific person when the acquired reliability satisfies the detection condition determined in step S12.

[0049] When an abnormality or an object is detected, the execution unit 273 notifies the notification destination acquired in step S12. This completes the security processing.

[0050] 5 is a flow diagram showing an example of the flow of audio output processing executed by the patrol security robot 2. The audio output processing is executed in parallel with security processing, either regularly or irregularly, while the patrol security robot 2 is patrolling the patrol area. The audio output processing is realized by the processor 27 executing a program and cooperating with other components of the patrol security robot 2.

[0051] First, the acquisition unit 271 acquires the security status of the adjacent area (step S21). The acquisition unit 271 receives information from the security device 3 indicating whether the adjacent area is in the security mode or the de-security mode.

[0052] Next, the determination unit 272 determines at least one of the content and volume of the warning to be output as audio by the patrol security robot 2 based on the security status of the adjacent area (step S22). When the adjacent area is in the release mode, the determination unit 272 may determine the content or volume of the warning so that the user is more likely to recognize surveillance by the patrol security robot 2 than when the adjacent area is in the security mode. For example, when the adjacent area is in the release mode, the determination unit 272 determines an audio message such as "Patrol security in progress" as the content of the warning, and when the adjacent area is in the security mode, the determination unit 272 determines music or the like notifying the presence of the patrol security robot 2 as the content of the warning. When the adjacent area is in the release mode, the determination unit 272 may set the volume to be higher than when the adjacent area is in the alert mode. Furthermore, when the adjacent area is in the release mode, the determination unit 272 may turn off music or the like notifying the presence of the patrol security robot 2 so that abnormal sounds in the surrounding area can be more easily detected.

[0053] Next, the output unit 274 outputs a warning by voice (step S22). This completes the voice output process.

[0054] As described above, the patrol security robot 2 has the acquisition unit 271 that acquires the security status of adjacent areas, and the determination unit 272 that determines the security operation to be performed while moving within the patrol area based on the security status. This allows the patrol security robot 2 to perform security operations that take into account the security status of adjacent areas.

[0055] In the above-described embodiment, in step S12 of the security processing, the determination unit 272 determines the security operation based on the security status of the adjacent area. The determination unit 272 may further determine the range for acquiring environmental information in step S13 based on the security status of the adjacent area. For example, the determination unit 272 may set a wider detection range for detecting abnormalities and objects when the adjacent area is in alert mode than when the adjacent area is in deactivated mode. This reduces false alarms when the adjacent area is in alert mode. The detection range may be set based on an angle range relative to the traveling direction of the patrol security robot 2 or a distance from the patrol security robot 2.

[0056] Furthermore, in step S12, the determination unit 272 may determine a security operation based on the security status of the adjacent area and the scheduled time for changing the security status of the adjacent area. For example, the scheduled time for changing the adjacent area from the alert mode to the deactivation mode is the opening time of the facility, and the scheduled time for changing the adjacent area from the deactivation mode to the alert mode is the closing time of the facility. The determination unit 272 may determine a security operation different from the cases when the adjacent area is in the alert mode and the deactivation mode when the scheduled time for changing the adjacent area to the alert mode has passed but the adjacent area is in the deactivation mode. For example, when the scheduled time for changing the adjacent area to the alert mode has passed but the adjacent area is in the deactivation mode, the determination unit 272 determines the detection target, detection conditions, and notification destination for the security operation so that a notification is less likely to be issued than when the adjacent area is in the alert mode and more likely to be issued than when the adjacent area is in the deactivation mode. This reduces missed notifications even if a user in the adjacent area forgets to change the mode to the alert mode. Similarly, when the adjacent area is in the alert mode even though the scheduled time for the adjacent area to be set to the deactivation mode has passed, the determination unit 272 determines the detection target, detection conditions, and notification destination for the security operation so that a notification is less likely to be issued than when the adjacent area is in the alert mode and more likely to be issued than when the adjacent area is in the deactivation mode. This reduces false alarms even if a user in the adjacent area forgets to change to the deactivation mode.

[0057] Some of the functions of the patrol security robot 2 described above may be implemented by the security guard terminal 4, the monitoring terminal 5, or another information processing device. For example, in step S13 of the security processing, the execution unit 273 may transmit information acquired from the sensor 23 to a server (not shown) and acquire information indicating whether an abnormality has been detected from the server.

[0058] It should be understood that those skilled in the art can make various changes, substitutions, and alterations to the present invention without departing from the spirit and scope of the present invention. The above-described embodiments and modifications may be appropriately combined and implemented within the scope of the present invention.

[0059] The autonomous mobile robot according to one embodiment of the present invention can contribute to solving social issues such as a declining labor force and long working hours. In addition, the autonomous mobile robot according to one embodiment of the present invention can contribute to achieving Goal 9 of the Sustainable Development Goals (SDGs) adopted by the United Nations, "Build resilient infrastructure, promote inclusive and sustainable industrialization, and promote innovation and resilience." [Explanation of symbols]

[0060] 2 Patrol Security Robot 271 Acquisition Department 272 Decision Section 273 Executive Department 274 Output Section

Claims

1. A patrol security robot that moves in a predetermined patrol area, an acquisition unit that acquires a security status of an adjacent area adjacent to the patrol area; a determination unit that determines a security operation to be performed while the robot moves through the patrol area based on the security status; an execution unit that executes the determined security operation; A patrol security robot with

2. The determination unit determines a detection range for detecting an abnormality or an object during the security operation based on the security status of the adjacent area. The patrol security robot according to claim 1 .

3. the security operation is an operation for detecting an abnormality or an object in the patrol area, The determination unit determines at least one of a detection target for the abnormality or the object in the security operation, a detection condition for the abnormality or the object, and a notification destination when the abnormality or the object is detected, based on the security status of the adjacent area. The patrol security robot according to claim 1 .

4. an output unit that outputs a voice warning while the robot moves through the patrol area; the determination unit determines at least one of the content and volume of the warning in the security operation based on the security status of the adjacent area. The patrol security robot according to claim 1 .

5. the determination unit further determines the security operation based on a scheduled time when the security status of the adjacent area will be changed. The patrol security robot according to claim 1 .

6. the acquisition unit acquires, as the security state, either one of an alert mode in which an alarm is output when a person is present in the adjacent area and a release mode in which the alarm is not output; The patrol security robot according to claim 1 .

Images