JP2020042600A - Robot group control system - Google Patents

Abstract

To provide a robot group control system that can appropriately perform subsequent guard work in accordance with the state of a work area acquired in guard work when performing a plurality of works including guard work.SOLUTION: A robot group control system 1 comprises: a plurality of self-propelled robots 2A to 2D that perform a first work related to guarding and a second work different from the first work in parallel or switchably in a work area 10; and a control device 3 that controls the plurality of self-propelled robots 2A to 2D. Each of the plurality of self-propelled robots 2A to 2D include a guard unit 23 that acquires state information of the work area 10 by the first work. The control device 3 includes a work mode determination unit 325 that determines a work to be performed by each of the plurality of self-propelled robots 2A to 2D among the first work and the second work based on the status information obtained by the guard unit 23.SELECTED DRAWING: Figure 1

Description

  The present invention provides, in a work area, a plurality of self-propelled robots that perform a first work related to security and a second work different from security in a parallel or switchable manner, a management device that manages a plurality of self-propelled robots, The present invention relates to a robot group management system including a robot group.

  2. Description of the Related Art Conventionally, a self-propelled robot that performs various operations such as security, cleaning, and transport while moving these various operations has been used in order to save labor and increase efficiency.

  For example, in Patent Document 1, in an automatic cleaning system including a plurality of self-propelled robots, each self-propelled robot transmits image information captured by a camera to a management room as security information while performing cleaning work. It is disclosed that a security operation is performed by transmitting a mobile object detection signal to a management room when a mobile object is detected based on image information.

JP-A-2000-342498

  However, a plurality of self-propelled robots disclosed in Patent Literature 1 perform security work while performing cleaning work in a cleaning area where each self-propelled robot is arranged. The robot does not cooperate in cleaning and security work.

  Therefore, even if an abnormal state occurs in which a mobile object is detected by one self-propelled robot during security work, the other self-propelled robots continue cleaning work as it is. For example, even if the security work around the point where the abnormal state is detected is focused on or the security work in the entire work area is strengthened, the security work is appropriately performed by only one self-propelled robot. There was a problem that it was not possible.

  The present invention has been made in view of such circumstances, and when performing a plurality of operations including security work, appropriate subsequent security work according to the state of the work area acquired in the security work. It is an object of the present invention to provide a robot group management system that can be performed at a time.

The present invention is to solve the above problems, and a robot group management system according to an embodiment of the present invention,
In the work area, a plurality of self-propelled robots that perform first work related to security and a second work different from the first work in a parallel or switchable manner, and a management device that manages the plurality of self-propelled robots, A robot group management system including
Each of the plurality of self-propelled robots includes:
An acquisition unit that acquires state information indicating a state of the work area by the first work,
The management device,
A determining unit that determines a task to be performed by each of the plurality of self-propelled robots among the first task and the second task based on the state information acquired by the acquiring unit. .

Further, the robot group management system according to one embodiment of the present invention,
Each of the plurality of self-propelled robots performs the first work and the second work in a parallel or switchable manner according to a plurality of work modes,
The determination unit determines, as the work, a work mode for each of the plurality of self-propelled robots among the plurality of work modes,
The working mode is
A parallel mode for performing the first work and the second work in parallel;
A first work alone mode for performing the first work alone;
A first work cooperation mode in which the first work is performed in cooperation with the other self-propelled robot;
At least.

Further, the robot group management system according to one embodiment of the present invention,
The management device,
Based on the state information acquired by the acquisition unit, further comprising a security level determination unit that determines a security level indicating the degree of security required,
The determination unit determines the work based on the security level determined by the security level determination unit.

Further, the robot group management system according to one embodiment of the present invention,
The security level determination unit, based on the learning model created based on the past state information acquired by the acquisition unit and the current state information acquired by the acquisition unit, the security level, Determination.

Further, the robot group management system according to one embodiment of the present invention,
The self-propelled robot performs the first work among the plurality of self-propelled robots in cooperation with another self-propelled robot based on the security level determined by the security level determination unit. The number or the ratio of the expression robots is determined.

Further, the robot group management system according to one embodiment of the present invention,
The determining unit is based on the security level determined by the security level determination unit, the higher the degree that the security is required, the higher the number or the higher the ratio, the number or The ratio is determined.

  According to the robot group management system according to the embodiment of the present invention, the determination unit indicates the state of the work area acquired by any of the plurality of self-propelled robots performing the first work related to security. Since the work performed by each of the plurality of self-propelled robots among the first work and the second work is determined based on the information, for example, an abnormal state may occur in any of the plurality of self-propelled robots performing the first work. When detected, the work performed by each of the plurality of self-propelled robots is determined according to the abnormal state, so that subsequent security work is appropriately performed according to the state of the work area acquired in the security work. Can be done.

1 is an overall configuration diagram illustrating a robot group management system 1 according to an embodiment of the present invention. It is a block diagram showing self-propelled robots 2A-2D concerning an embodiment of the present invention. FIG. 4 is an explanatory diagram showing a relationship among state information, abnormal state (abnormality detection factor), security level, and work mode in the robot group management system 1 according to the embodiment of the present invention. It is a block diagram showing management device 3 concerning an embodiment of the present invention. It is a flowchart which shows operation | movement of security work and cleaning work in the robot group management system 1 which concerns on embodiment of this invention. 6 is a flowchart illustrating an operation of a learning step of a learning model 314 in the robot group management system 1 according to the embodiment of the present invention. 9 is a flowchart illustrating an operation of a use stage of a learning model 314 in the robot group management system 1 according to the embodiment of the present invention.

  Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is an overall configuration diagram showing a robot group management system 1 according to an embodiment of the present invention. The robot group management system 1 includes, in a work area 10, a robot group 2 including a plurality of self-propelled robots 2A to 2D that perform a first work related to security and a second work different from the first work in a parallel or switchable manner. , A management device 3 that manages the robot group 2, a manager terminal 4 used by a manager M of the work area 10, and a network 5 that connects the devices so that they can communicate with each other.

  The “work area 10” is an area in which a plurality of self-propelled robots 2A to 2D work, and is, for example, an office building, a commercial facility, an apartment house, a public facility, a factory, a station, an airport, a port, an event. Venues, stadiums, exhibition halls, halls, theme marks, landmarks, etc. The work area 10 may be any of an indoor area and an outdoor area, and may be an area including both an indoor area and an outdoor area. When the work area 10 is indoors, the work area 10 may include, for example, a plurality of floors of a building, a basement of a building, and an indoor parking lot. When the work area 10 is outdoor, for example, a rooftop floor of a building, an outdoor parking lot May be included. In the present embodiment, it is assumed that the work area 10 is an indoor area.

  The “first operation related to security” includes an abnormal state in the work area 10, for example, a suspicious person (fighting, abusive language, burglary, robbery, etc.) to prepare for an event such as an accident, destruction, or theft. People, such as those who are sitting on the spot like tramps, suspicious objects (things that did not exist before, including moving objects), damage to equipment such as windows and doors, fires , Detecting a gas leak or the like, and coping with the abnormal state, and includes, for example, operations such as monitoring, patrol, tracking, reporting, warning, collecting information, and recording and transmitting the collected information. In the present embodiment, the first work is referred to as “security work”.

  The “second operation different from the first operation” is an operation performed in parallel or switchably with the security operation, and includes, for example, operations such as cleaning, transport, and watering. In the present embodiment, the “second operation” is referred to as “cleaning operation” that is an operation related to cleaning.

  Each of the plurality of self-propelled robots 2 </ b> A to 2 </ b> D is a self-propelled robot, and performs a guard operation and a cleaning operation in a parallel or switchable manner according to a plurality of operation modes. In the present embodiment, the number of the self-propelled robots 2A to 2D is described as four, but the number of the self-propelled robots may be appropriately changed.

  The “plurality of operation modes” include, for example, a “parallel mode” in which the security operation and the cleaning operation are performed in parallel, a “security operation alone mode” in which the security operation is performed independently (first operation only mode), and a security operation in other modes. At least a “security work cooperation mode” (first work cooperation mode) performed in cooperation with the self-propelled robot. In the “parallel mode”, the cleaning operation may be performed as a sub-operation while the security operation is performed as a main operation, or the security operation may be performed as a sub-operation while performing the cleaning operation as a main operation. The plurality of operation modes include a “cleaning operation cooperation mode” (second operation cooperation mode) in which the cleaning operation is performed in cooperation with another self-propelled robot, and a “cleaning operation independent mode” in which the cleaning operation is performed independently ( The second work alone mode) and a “manual mode” in which a manual operation is performed by the administrator terminal 4 may be further included.

  The management device 3 determines a work mode for each of the four self-propelled robots 2A to 2D among the plurality of work modes on the basis of the security level, so that four self-propelled robots out of the security work and the cleaning work are determined. The work performed by each of the traveling robots 2A to 2D is determined.

  The “security level” is an index indicating the degree of need for security, and is divided into, for example, a plurality of stages. In the present embodiment, the security level is classified into five levels of “critical level”, “high level”, “medium level”, “low level”, and “continuation level” in descending order of the level of need for security. Shall be.

  The administrator terminal 4 includes, for example, an information processing device such as a general-purpose computer installed in a management room of the work area 10, and a portable terminal device such as a tablet and a smartphone that the administrator M can carry. The administrator terminal 4 receives various inputs from the administrator M via the input screen, and presents various information to the administrator M via the display screen and voice. In the present embodiment, the number of the administrator terminals 4 is one, but the number of the administrator terminals 4 may be plural.

(Specific configuration of self-propelled robots 2A to 2D)
FIG. 2 is a block diagram showing the self-propelled robots 2A to 2D according to the embodiment of the present invention. Each of the four self-propelled robots 2A to 2D includes a position detection unit 20 that detects the current position of the self-propelled robots 2A to 2D as position information, an HDD, a memory, and the like, and stores various types of information. A storage unit 21, a cleaning unit 22 for cleaning the floor and wall surfaces of the work area 10 by a cleaning operation, a security unit 23 (acquisition unit) for acquiring state information indicating the state of the work area 10 by a security operation, A traveling unit 24 for moving the main body of each of the robots 2A to 2D back and forth, left and right, a control unit 25 including a processor such as a CPU, a communication unit 26 that is a communication interface with the network 5, and a self-propelled robot 2A to 2D. A battery 27 that supplies power to each unit of the 2D.

  The position detection unit 20 detects the current positions of the self-propelled robots 2A to 2D by receiving position information from position transmitters (not shown) installed at respective positions in the work area 10. The position detection unit 20 may be a GPS sensor, a position estimation sensor based on dead reckoning using speed, angular velocity, or the like, or a combination thereof.

  The storage unit 21 includes a map including information on the positions and sizes of passages, rooms, walls, doors, doorways, windows, elevators, stairs, off-limits areas, charging stations, waste disposal stations, and the like provided in the work area 10. Information 210, user information 211 including information on the characteristics of the user and the manager M, such as the face and height of the work area 10, and the self-propelled robots 2A to 2D when performing the security work and the cleaning work. Work plan information 212 including information on a work mode and a travel route, and a robot control program 213 for controlling operations of the self-propelled robots 2A to 2D are stored.

  The map information 210, the user information 211, and the work plan information 212 are appropriately updated by receiving the information transmitted by the management device 3, respectively.

  The travel route included in the work plan information 212 includes, for example, a plurality of position coordinates and a passing time at each position coordinate, and at a monitoring point where a specific place is continuously monitored, a predetermined position coordinate is used. Is instructed to stop for a predetermined time. The travel route may be calculated by any of the self-propelled robots 2A to 2D and the management device 3. The traveling route may be a traveling route calculated in advance so as to reach a destination from a departure place via a plurality of transit points, or, for example, track a monitoring target (eg, a user or a suspicious person). When performing such security work, it may be instructed to calculate the traveling route in real time according to the position or movement of the monitored object.

  The security unit 23 includes an omnidirectional camera 230 that obtains 360-degree wide-angle image information in all directions, a high-resolution camera 231 that obtains narrow-angle image information having a predetermined angle of view, and a self-propelled type using, for example, infrared rays. A human sensor 232 that acquires the position and direction of a person existing around the robots 2A to 2D as human information, a temperature / humidity sensor 233 that acquires the temperature and humidity of the work area 10 as temperature / humidity information, A gas leak sensor 234 for acquiring the presence or absence as gas leak information, and a microphone 235 for acquiring environmental sounds around the self-propelled robots 2A to 2D as environmental sound information are provided. The security unit 23 may include, for example, a flame sensor that detects fire, smoke, and the like, instead of the temperature and humidity sensor 233. In addition, the security unit 23 includes a warning unit that issues a warning to the suspicious person by, for example, sound or light, and a notification unit that notifies the user of, for example, the presence of a suspicious object by sound or light. You may.

  Then, the security unit 23 includes, as state information of the work area 10, wide-angle image information from the omnidirectional camera 230, narrow-angle image information from the high-resolution camera 231, human feeling information from the human sensor 232, and temperature information from the temperature and humidity sensor 233. The humidity information, the gas leak information from the gas leak sensor 234, and the environmental sound information from the microphone 235 are acquired.

  The security unit 23 includes an abnormal state detection unit 236 that detects an abnormal state in the work area 10 based on the state information of the work area 10.

  FIG. 3 is an explanatory diagram showing a relationship among state information, an abnormal state (an abnormality detection factor), a security level, and a work mode in the robot group management system 1 according to the embodiment of the present invention. The abnormal state detection unit 236 detects an abnormal state in the work area 10 based on the state information of the work area 10 as in the relation between the state information and the abnormal state (abnormality detection factor) shown in FIG. An abnormality detection factor (in the present embodiment, any one of a suspicious person, a suspicious object, entry, a fire, a gas leak, a destruction sound, and a scream) that is a cause of detecting an abnormal state is detected. Note that the abnormal state detection unit 236 may simultaneously detect a plurality of abnormality detection causes.

  The abnormal state detection unit 236 includes, for example, the user information 211 by using the wide-angle image information from the omnidirectional camera 230 and the narrow-angle image information from the high-resolution camera 231 as the state information of the work area 10. An abnormal state in the work area 10 is detected by detecting a "suspicious person" who is not a person and an "suspicious object" which is an object not included in the map information 210.

  The abnormal state detection unit 236 uses the human presence information from the human presence sensor 232 as the state information of the work area 10, and detects, for example, “entrance” to a restricted area included in the map information 210 to perform the work. An abnormal state in the area 10 is detected. In the present embodiment, it is assumed that “entrance” to a restricted area detects “entrance” by a user, and “entrance” by a suspicious person detects “suspicious person”.

  The abnormal state detection unit 236 detects, for example, “fire” in which the temperature of the work area 10 is higher than a predetermined temperature by using the temperature and humidity information from the temperature and humidity sensor 233 as the state information of the work area 10. By doing so, an abnormal state in the work area 10 is detected.

  The abnormal state detection unit 236 uses the gas leak information from the gas leak sensor 234 as the state information of the work area 10, and detects, for example, “gas leak” in which the gas leak has occurred, thereby detecting the work area 10. Detect an abnormal state.

  The abnormal state detection unit 236 uses the environmental sound information from the microphone 235 as the state information of the work area 10, for example, a “destruction sound” when a window or a door is destroyed, or a “scream” generated by a user. Is detected, an abnormal state in the work area 10 is detected.

  The traveling unit 24 travels by rotating an electric motor (not shown) with electric power supplied from the battery 27. At that time, the traveling unit 24 travels along the traveling route by referring to the traveling route included in the map information 210 and the work plan information 212 based on the position information detected by the position detecting unit 20. The vehicle travels while avoiding a collision with an obstacle by recognizing a nearby obstacle based on state information acquired by the security unit 23, for example, wide-angle image information or narrow-angle image information.

  The control unit 25 functions as the work mode switching unit 250, the parallel mode execution unit 251, the security work independent mode execution unit 252, and the security work cooperation mode execution unit 253 by executing the robot control program 213.

  Then, the control unit 25 controls the position information detected by the position detection unit 20 and the map information 210 stored in the storage unit 21 so that the security operation and the cleaning operation can be performed in parallel or switchably in accordance with the plurality of operation modes. And controlling the cleaning unit 22, the security unit 23, and the traveling unit 24 based on the work plan information 212, the state information of the work area 10 acquired by the security unit 23, and the information received from the management device 3 via the communication unit 26. I do.

  The work mode switching unit 250 selects one of the parallel mode execution unit 251, the security work single mode execution unit 252, and the security work cooperation mode execution unit 253 according to the work mode included in the work plan information 212 among the plurality of work modes. Switch to the running state.

  When switched to the execution state by the work mode switching unit 250, the parallel mode execution unit 251 operates the cleaning unit 22 by operating the traveling unit 24 and traveling along the traveling route included in the work plan information 212. In addition to performing the cleaning operation, the security unit 23 is operated to perform the security operation. Then, when the abnormal state is detected by the abnormal state detection unit 236, the parallel mode execution unit 251 sends the abnormal state detection notification including the time, the position information, the state information, and the abnormality detection factor when the abnormal state is detected. Is transmitted to the management device 3.

  When switched to the execution state by the work mode switching unit 250, the security work independent mode execution unit 252 operates the traveling unit 24 to operate the traveling route included in the work plan information 212 or the monitoring target (for example, a user, The security unit 23 is operated while traveling along a traveling route calculated in real time to track the monitoring target according to the position and movement of the (suspicious person) or to move ahead in the traveling direction of the monitoring target. To perform independent security work. Then, the security work independent mode execution unit 252 transmits the status information acquired by the security unit 23 to the management device 3 in real time, and when the abnormal status detection unit 236 detects an abnormal status, notifies the abnormal status detection notification. Is transmitted to the management device 3.

  When switched to the execution state by the work mode switching unit 250, the security work cooperation mode execution unit 253 operates the traveling unit 24 to operate the traveling route included in the work plan information 212 or the monitoring target (for example, a user, The security unit 23 is operated while traveling along a traveling route calculated in real time to track the monitoring target according to the position and movement of the (suspicious person) or to move ahead in the traveling direction of the monitoring target. This allows security work in cooperation with other self-propelled robots. Then, the security work cooperation mode execution unit 253 transmits the status information acquired by the security unit 23 to the management device 3 in real time, and when the abnormal status detection unit 236 detects an abnormal status, notifies the abnormal status detection notification. Is transmitted to the management device 3.

  Note that the security work cooperation mode execution unit 253 communicates with another self-propelled robot and exchanges position information and work plan information 212 with each other when performing security work in cooperation with another self-propelled robot. Thus, the monitoring target may be tracked in a relay format.

  The work plan update unit 326 determines that the work mode for all four self-propelled robots 2A to 2D is “security work” because the abnormality detection factor is “suspicious person” and the security level is “severe level”. If the “cooperative mode” is determined, the self-propelled robot 2A that has detected the abnormal state is instructed to calculate the travel route in real time according to the position and movement of the suspicious person, and the other three In contrast to the self-propelled robots 2B to 2D, the self-propelled robot 2A that sets a passage or door ahead of the suspicious person's traveling direction as a destination or tracks the suspicious person, The position information is exchanged with each other, and an instruction is issued to calculate a traveling route based on the position information of the self-propelled robot 2A or an instruction from the self-propelled robot 2A.

(Specific configuration of management device 3)
FIG. 4 is a block diagram illustrating the management device 3 according to the embodiment of the present invention. The management device 3 is a device that functions as a server or a cloud, and includes an input unit 30 configured by a keyboard, a touch panel, and the like, a storage unit 31 configured by an HDD, a memory, and the like, and a control configured by a processor such as a CPU. A communication unit 33 serving as a communication interface with the network 5; and an output unit 34 including a display, a speaker, and the like. Note that the input unit 30 and the output unit 34 may be omitted. In that case, the administrator terminal 4 may function as the input unit 30 and the output unit 34.

  The storage unit 31 has master map information 310 to be transmitted to the self-propelled robots 2A to 2D as map information 210, and master user information 311 to be transmitted to the self-propelled robots 2A to 2D as user information 211. Master work plan information 312 to be transmitted to the self-propelled robots 2A to 2D as work plan information 212, time, position information, state information, abnormality detection factor, and time when an abnormal state in the work area 10 is detected. The abnormal state history information 313 which manages the security level judgment result information indicating the judgment result of the security level judged by the manager M and the abnormal state history information 313 which is past state information are used as training data, and the current state information is used. The robot group 2 is controlled by controlling the operation of the management device 3 and the learning model 314 created for determining the security level. A robot group control program 315 to sense, are stored.

  The master map information 310, the master user information 311 and the master work plan information 312 are transmitted from the management device 3 to the four self-propelled robots 2A to 2D when the information is created or updated by the management device 3. You. At that time, the management device 3 may transmit only information used by each of the self-propelled robots 2A to 2D, instead of transmitting all information. For example, the management device 3 may transmit the master work plan information 312 to the self-propelled robot 2A. , The information corresponding to the work plan information 212 of the self-propelled robot 2A among the master work plan information 312 may be transmitted.

  The control unit 32 executes the robot group management program 315 to execute a work plan creation unit 320, an abnormal state history management unit 321, a security level reception unit 322, a learning unit 323, a security level determination unit 324, and a work mode determination unit 325. , Function as a work plan updating unit 326 and a notification processing unit 327.

  Based on the master map information 310, the work plan creator 320 sets each of the four self-propelled robots 2A to 2D to perform the security work and the cleaning work by sharing each area with the four self-propelled robots 2A to 2D. Each of the four self-propelled robots 2A to 2D is assigned to each of the mobile robots 2A to 2D, and a work mode in each sharing area and a travel route for executing the work mode are set. The master work plan information 312 including the plan information 212 is created. The work plan creation unit 320 newly creates master work plan information 312 based on information input by the manager M via the manager terminal 4 or a part of the created master work plan information 312. May be modified.

  The abnormal state history management unit 321 monitors the abnormal state in the work area 10 by determining whether an abnormal state detection notification has been received from any of the four self-propelled robots 2A to 2D. When determining that the abnormal state detection notification has been received, the abnormal state history management unit 321 determines whether or not the learning model 314 has been created. The detection notification is sent to the security level reception unit 322, and if it is determined that the information has been created, the abnormal state detection notification is sent to the security level determination unit 324.

  Further, the abnormal state history management unit 321 associates the time, the position information, the state information, and the abnormality detection factor included in the abnormal state detection notification with the security level determination result information received by the security level receiving unit 322. To add to the abnormal state history information 313.

  The security level receiving unit 322 determines whether an abnormal state has been detected in any of the four self-propelled robots 2A to 2D and has received an abnormal state detection notification, and has not created the learning model 314. Output information for outputting the position information and the state information included in the abnormal state detection notification to the administrator terminal 4 is transmitted to the administrator terminal 4.

  Then, the security level receiving unit 322 checks the position information and the state information output to the administrator terminal 4 by the manager M, determines the security level based on the position information and the state information, and determines the security level. When the determined result is input via the administrator terminal 4, the determined result is accepted as security level determination result information.

  The administrator M determines the security level according to a predetermined security level determination criterion by checking the position information and the state information when an abnormal state is detected in any of the four self-propelled robots 2A to 2D. I do.

  As a predetermined security level determination criterion, the administrator M may use, for example, wide-angle image information and a narrow angle when an abnormal state is detected, such as the relationship between the abnormal state (the abnormality detection factor) and the security level shown in FIG. If "suspicious person" is confirmed in the image information, "fire" is confirmed in the temperature and humidity information, or "gas leak" is confirmed in the gas leak information, it is judged as "serious level" and the environmental sound If "destruction sound" or "scream" is confirmed in the information, it is determined as "high level". If "suspicious object" is confirmed in the wide-angle image information and narrow-angle image information, it is determined as "medium level". On the other hand, if "entrance" is confirmed from the human feeling information, it is determined to be "low level", and if it is confirmed that detection as an abnormal state is erroneous detection, it is determined to be "continuation level".

  The learning unit 323 creates a learning model 314 based on the past state information acquired by the security unit 23. Specifically, the learning unit 323 determines whether the abnormal state history information 313 satisfies a predetermined learning condition (for example, whether the number of data stored in the abnormal state history information 313 exceeds a predetermined number). When it is determined that the predetermined learning condition is satisfied, the correlation between the state information when the abnormal state is detected and the security level determination result information is determined based on the abnormal state history information 313. And a learning model 314 is created. Accordingly, the learning model 314 has learned a predetermined security level determination criterion when the manager M determines the security level.

  The security level determination unit 324 determines a security level indicating the degree of need for security based on the status information acquired by the security unit 23. The security level determination unit 324 determines the security level based on the learning model 314 created by the learning unit 323 based on the past state information and the current state information acquired by the security unit 23.

  More specifically, the security level determination unit 324 generates a learning model 314 when an abnormal state is detected in any of the four self-propelled robots 2A to 2D and an abnormal state detection notification is received. If the abnormal state is detected, the state information included in the abnormal state detection notification is input to the learning model 314, and the past state information that is the same as or similar to the state information when the abnormal state is detected is determined by the administrator. The security level is determined so that M is the same as or similar to the security level determined in the past.

  The work mode determination unit 325 determines the work to be performed by each of the four self-propelled robots 2A to 2D among the security work and the cleaning work based on the state information acquired by the security unit 23. When an abnormal state is detected in any one of the four self-propelled robots 2A to 2D and the security level is determined by the security level determination unit 324, the work mode determination unit 325 determines based on the security level. As work performed by each of the four self-propelled robots 2A to 2D, a work mode for each of the four self-propelled robots 2A to 2D among a plurality of work modes is determined.

  In addition, when an abnormal state is detected in any of the four self-propelled robots 2A to 2D and the security level receiving unit 322 receives the security level determination result information, The work performed by each of the four self-propelled robots 2 </ b> A to 2 </ b> D based on the security level indicated by the security level determination result information includes the work performed on each of the four self-propelled robots 2 </ b> A to 2 </ b> D among a plurality of work modes. Determine the mode.

  When determining the work mode for each of the four self-propelled robots 2A to 2D based on the security level, the work mode determination unit 325 uses the security level and the work mode shown in FIG. Based on the security level, the number or the ratio of the self-propelled robots to be set in the “security work cooperation mode” among the four self-propelled robots 2A to 2D is determined.

  Specifically, the work mode determination unit 325 determines that the higher the security level, the larger the number of self-propelled robots in the “security work cooperation mode” or the higher the ratio of the four self-propelled robots. A work mode for each of the expression robots 2A to 2D is determined. In the present embodiment, the work mode determination unit 325 determines the number of self-propelled robots to be set in the “security work cooperation mode”.

  Therefore, when the security level is the “critical level”, the work mode determination unit 325 determines the work mode for all four self-propelled robots 2A to 2D to be the “security work cooperation mode”.

  When the security level is “high level”, the work mode determination unit 325 determines the work mode for some (three in this embodiment) self-propelled robots 2A to 2C to be “security work cooperation mode”. Then, the work mode for the other self-propelled robot 2D is determined to be the “parallel mode”. The work mode determining unit 325 determines, as a part of the self-propelled robots 2A to 2C, a predetermined number (for example, in the present embodiment) located within a predetermined range around the position of the self-propelled robot 2A that has detected an abnormal state. In this case, a predetermined number (two in the present embodiment) of the self-propelled robots 2B are selected in the order of proximity to the position of the self-propelled robot 2A that has detected the abnormal state. Or 2C.

  When the security level is “medium level”, the work mode determination unit 325 determines the work mode for some (two in this embodiment) self-propelled robots 2A and 2B to be “security work cooperation mode”. Then, the work mode for the other self-propelled robots 2C and 2D is determined to be the “parallel mode”. The work mode determination unit 325 determines, as a part of the self-propelled robots 2A and 2B, for example, a predetermined number (in the present embodiment) located within a predetermined range around the position of the self-propelled robot 2A that has detected an abnormal state. In this case, one (1) self-propelled robot 2B is selected, or a predetermined number (one in this embodiment) of self-propelled robots 2B is selected in order of proximity to the position of the self-propelled robot 2A that has detected an abnormal state. Or

  When the security level is “low level”, the work mode determination unit 325 determines the work mode for the self-propelled robot 2A that has detected the abnormal state to be “security work alone mode”, and the other self-propelled robots The work mode for 2B to 2D is determined to be “parallel mode”.

  When the security level is the “continuation level”, the work mode determination unit 325 determines the work mode for the four self-propelled robots 2A to 2D to be the “parallel mode”.

  When the work mode is determined by the work mode determination unit 325, the work plan update unit 326 resets the determined work mode and the travel route for executing the determined work mode. Thus, the master work plan information 312 including the work plan information 212 for each of the four self-propelled robots 2A to 2D is updated.

  The work plan update unit 326 determines that the work mode for all four self-propelled robots 2A to 2D is “security work” because the abnormality detection factor is “suspicious person” and the security level is “severe level”. If the “cooperative mode” is determined, the self-propelled robot 2A that has detected the abnormal state is instructed to calculate the travel route in real time according to the position and movement of the suspicious person, and the other three In contrast to the self-propelled robots 2B to 2D, the self-propelled robot 2A that sets a passage or door ahead of the suspicious person's traveling direction as a destination or tracks the suspicious person, The position information is exchanged with each other, and an instruction is issued to calculate a traveling route based on the position information of the self-propelled robot 2A or an instruction from the self-propelled robot 2A.

  Further, the work plan updating unit 326 determines that the work mode for the three self-propelled robots 2A to 2C is “security work” because the abnormality detection factor is “scream” and the security level is “high”. If it is determined to be the “cooperation mode”, the self-propelled robot 2A that has detected the abnormal state is instructed to calculate in real time a travel route for searching for a screaming user. For the two self-propelled robots 2B and 2C, a traveling route for searching for a suspicious person or a suspicious object is set around the position where the scream is detected.

  Further, the work plan updating unit 326 sets the work mode for the two self-propelled robots 2A and 2B to “security”, for example, when the abnormality detection factor is “suspicious object” and the security level is “medium level”. When the operation mode is determined, the self-propelled robot 2A that has detected the abnormal state sets a traveling path for approaching the suspicious object and monitoring the suspicious object. In 2B, a traveling route for searching for a suspicious person or another suspicious object is set around the position where the suspicious object is detected.

  Further, the work plan update unit 326 sets the work mode for the self-propelled robot 2A that has detected the abnormal state to “security work”, for example, because the abnormality detection factor is “entry” and the security level is “low”. When the mode is determined to be the “single mode”, the self-propelled robot 2A that has detected the abnormal state is instructed to calculate the traveling route in real time according to the position and movement of the user who “enters”. Or set a traveling route to be monitored around the off-limits area where “entering” was performed.

  When an abnormal state is detected in any of the four self-propelled robots 2A to 2D and an abnormal state detection notification is received, the notification processing unit 327 performs a process based on the abnormality detection factor included in the abnormal state detection notification. Then, it is determined whether or not it is necessary to notify the detection of an abnormal state in the work area 10 to an external organization (for example, police, fire department, security company, etc.). Performs a report process of transmitting report information to an external organization. For example, when the abnormality detection factor included in the abnormal state detection notification is “fire”, the notification processing unit 327 determines that notification is necessary, and transmits notification information indicating that a fire has been detected to an external organization. I do.

(Operation of robot group management system 1)
Next, an operation of the robot group management system 1 according to the embodiment of the present invention will be described. FIG. 5 is a flowchart showing operations of a security operation and a cleaning operation in the robot group management system 1 according to the embodiment of the present invention. FIG. 6 is a flowchart illustrating the operation of the learning step of the learning model 314 in the robot group management system 1 according to the embodiment of the present invention. FIG. 7 is a flowchart showing the operation of the use stage of the learning model 314 in the robot group management system 1 according to the embodiment of the present invention.

  First, as shown in FIG. 5, the work plan creation unit 320 of the management device 3 creates master work plan information 312 and, for the four self-propelled robots 2A to 2D, each of the self-propelled robots 2A to 2D. Is transmitted (step S1). Here, the description will be given assuming that the work plan creation unit 320 has set “parallel mode” as the work mode for the four self-propelled robots 2A to 2D.

  The abnormal state history management unit 321 of the management device 3 determines whether an abnormal state detection notification has been received from any of the four self-propelled robots 2A to 2D, and thereby determines whether the abnormal state in the work area 10 has been detected. Is monitored (step S2).

  On the other hand, when each of the self-propelled robots 2 </ b> A to 2 </ b> D receives the work plan information 212 from the management device 3, the work plan information 212 is stored in the storage unit 21. Then, the work mode switching unit 250 of each of the self-propelled robots 2A to 2D switches the parallel mode execution unit 251 to the execution state according to the work mode (“parallel mode”) included in the work plan information 212 (steps S101, S201, S301, S401).

  Next, when the parallel mode execution unit 251 of each of the self-propelled robots 2A to 2D is switched to the execution state, the cleaning operation by the cleaning unit 22 and the security while traveling along the traveling route included in the work plan information 212 are performed. The security work by the unit 23 is performed in parallel.

  When the abnormal state detecting unit 236 detects an abnormal state ("Yes" in step S102, "Yes" in S202, "Yes" in S302, and "Yes" in S402), the parallel mode execution unit 251 Then, an abnormal state detection notification including the time when the abnormal state is detected, position information, state information, and an abnormality detection factor is transmitted to the management device 3 (steps S103, S203, S303, and S403).

  Next, when the abnormal state history management unit 321 of the management device 3 determines that an abnormal state detection notification has been received from any of the four self-propelled robots 2A to 2D, it is determined whether the learning model 314 has been created. It is determined whether or not it is (step S3). When the abnormal state history management unit 321 determines that the learning model 314 has not been created (“No” in step S3), the abnormal state detection notification is sent to the security level receiving unit 322, and the process illustrated in FIG. Move to S11. On the other hand, when the abnormal state history management unit 321 determines that the learning model 314 has been created (“Yes” in step S3), the abnormal state detection notification is sent to the security level determination unit 324, and FIG. Move to step S13.

  As described above, when it is determined that the learning model 314 has not been created in step S3 in FIG. 5 (“No” in step S3), as shown in FIG. An abnormal state detection notification is received from the state history management unit 321, and a determination result of the security level determined by the administrator M with respect to the abnormal state detection notification is received as security level determination result information (step S <b> 11).

  Next, the work mode determination unit 325 determines a work mode for each of the four self-propelled robots 2A to 2D based on the security level indicated by the security level determination result information received by the security level reception unit 322. (Step S12).

  Here, the management device 3 receives the abnormal state detection notification from the self-propelled robot 2A, and the security level indicated by the security level determination result information received by the security level receiving unit 322 is “medium level”. The mode determination unit 325 determines the work mode for the two self-propelled robots 2A and 2B to be “security work cooperation mode”, and sets the work mode for the other two self-propelled robots 2C and 2D to “parallel mode”. The description will be made assuming that it has been determined.

  Then, when the work mode is determined by the work mode determination unit 325, the work plan update unit 326 resets the determined work mode and the travel route for executing the determined work mode. As a result, the master work plan information 312 is updated, and the work plan information 212 corresponding to each of the self-propelled robots 2A to 2D is transmitted to the four self-propelled robots 2A to 2D (step S21).

  Next, when each of the self-propelled robots 2 </ b> A to 2 </ b> D receives the updated work plan information 212 from the management device 3, it stores the updated work plan information 212 in the storage unit 21. Then, the work mode switching unit 250 of the two self-propelled robots 2A and 2B sets the security work cooperation mode execution unit 253 to the execution state according to the work mode (“security work cooperation mode”) included in the work plan information 212. Switching (steps S111 and S211), the work mode switching units 250 of the other two self-propelled robots 2C and 2D perform the parallel mode execution unit 251 according to the work mode (“parallel mode”) included in the work plan information 212. Is continuously executed (steps SS311 and 411).

  Therefore, the two self-propelled robots 2A and 2B cooperate in the security work according to the security work coordination mode (steps S112 and S212), and the other two self-propelled robots 2C and 2D perform the parallel mode. , The preparatory work and the cleaning work are performed in parallel (steps S312 and S412).

  In addition, the notification processing unit 327 externally notifies that the abnormal state in the work area 10 has been detected based on the abnormality detection factor included in the abnormal state detection notification received from any of the four self-propelled robots 2A to 2D. It is determined whether or not it is necessary to notify the institution (step S31). If it is determined that the notification is necessary ("Yes" in step S31), a notification process is performed (step S32).

  Next, the abnormal state history management unit 321 determines the time, the position information, the state information, and the abnormality detection factor included in the abnormal state detection notification received from any of the four self-propelled robots 2A to 2D, and proceeds to step S11. The security level determination result information received by the security level reception unit 322 is added to the abnormal state history information 313 in a linked state (step S33).

  Then, the learning unit 323 determines whether or not the abnormal state history information 313 satisfies a predetermined learning condition (Step S41), and when it determines that the predetermined learning condition is satisfied (“Yes” in Step S41). Based on the abnormal state history information 313, the correlation between the state information when the abnormal state is detected and the security level determination result information is learned, and a learning model 314 is created (step S42). On the other hand, when the learning unit 323 determines that the predetermined learning condition is not satisfied (“No” in step S41), the process returns to step S2 in FIG. 5 without creating the learning model 314.

  On the other hand, when it is determined in step S3 of FIG. 5 that the learning model 314 has been created (“Yes” in step S3), as shown in FIG. The security level is determined by receiving the abnormal state detection notification from the management unit 321 and inputting the state information included in the abnormal state detection notification to the learning model 314 (step S13).

  Next, the work mode determination unit 325 determines a work mode for each of the four self-propelled robots 2A to 2D based on the security level determined by the security level determination unit 324 (Step S14).

  Here, since the management device 3 receives the abnormal state detection notification from the self-propelled robot 2A and the security level determined by the security level determination unit 324 is “critical level”, the work mode determination unit 325 Description will be made assuming that the work mode for all the self-propelled robots 2A to 2D has been determined to be the "security work cooperation mode".

  Then, when the work mode is determined by the work mode determination unit 325, the work plan update unit 326 resets the determined work mode and the travel route for executing the determined work mode. As a result, the master work plan information 312 is updated, and the work plan information 212 corresponding to each of the self-propelled robots 2A to 2D is transmitted to the four self-propelled robots 2A to 2D (step S21).

  Next, when each of the self-propelled robots 2 </ b> A to 2 </ b> D receives the updated work plan information 212 from the management device 3, it stores the updated work plan information 212 in the storage unit 21. Then, the work mode switching unit 250 of the four self-propelled robots 2A to 2D sets the security work cooperation mode execution unit 253 to the execution state according to the work mode (“security work cooperation mode”) included in the work plan information 212. Switch (steps S121, S221, S321, S421).

  Therefore, the four self-propelled robots 2A to 2D cooperate in security work according to the security work cooperation mode (steps S122, S222, steps S322, S422).

  In addition, the notification processing unit 327 externally notifies that the abnormal state in the work area 10 has been detected based on the abnormality detection factor included in the abnormal state detection notification received from any of the four self-propelled robots 2A to 2D. It is determined whether or not it is necessary to notify the institution (step S31). If it is determined that the notification is necessary ("Yes" in step S31), a notification process is performed (step S32).

  As described above, the management device 3 determines the security level according to the abnormal state in the work area 10, and determines the work mode for the four self-propelled robots 2A to 2D according to the security level. Thereby, each self-propelled robot 2A performs a security operation and a cleaning operation according to the operation mode.

  Therefore, according to the robot group management system 1 according to the present embodiment, the work mode determination unit 325 determines that the work area 10 acquired by the security unit 23 of any one of the plurality of self-propelled robots 2A to 2D performing the security work. Of the security work and the cleaning work, each of the plurality of self-propelled robots 2A to 2D is determined based on the state information indicating the state of the robot. For example, the plurality of self-propelled robots 2A to perform the security work are determined. When an abnormal state is detected in any of the 2Ds, the work performed by each of the plurality of self-propelled robots 2A to 2D is determined according to the abnormal state. The subsequent security work can be appropriately performed according to the state.

  Further, according to the robot group management system 1 according to the present embodiment, the work mode determination unit 325 performs the work for each of the plurality of self-propelled robots based on the state information of the work area 10 acquired by the security unit 23. Since the mode is determined, for example, if the work mode for a plurality of self-propelled robots is determined to be the security cooperative mode, one self-propelled It is possible to perform a wide range of security work more appropriately than a type robot performs a security work.

  Further, according to the robot group management system 1 according to the present embodiment, the security level determination unit 324 sets the security level indicating the degree of need for security based on the state information of the work area 10 acquired by the security unit 23. Then, the work mode determination unit 325 determines the work to be performed by each of the plurality of self-propelled robots 2A to 2D among the security work and the cleaning work based on the security level determined by the security level determination unit 324. Since the work performed by each of the plurality of self-propelled robots 2A to 2D is determined according to the degree to which security is required, necessary security work can be reliably performed according to the state of the work area.

  Further, according to the robot group management system 1 according to the present embodiment, the security level determination unit 324 acquires the learning model 314 created based on the past state information acquired by the security unit 23, and acquires the learning model 314 from the security unit 23. Since the security level is determined based on the current status information that has been obtained, the security level is determined based on the current status information in the same manner as when the administrator determines the security level based on the past status information. Therefore, it is possible to accurately determine the security level without previously setting a reference for determining the security level, and perform the same security work as that performed in the past according to the current state of the work area. be able to.

  Further, according to the robot group management system 1 according to the present embodiment, the work mode determination unit 325 links security work among a plurality of self-propelled robots based on the security level determined by the security level determination unit 324. The number or ratio of self-propelled robots to perform is determined, so the number or ratio of self-propelled robots that cooperate with security work is determined according to the degree of security required. Can be performed reliably.

  Further, according to the robot group management system 1 according to the present embodiment, the work mode determination unit 325 performs the security work based on the security level determined by the security level determination unit 324 as the degree to which security is required is higher. The number or ratio is determined so that the number of self-propelled robots to perform in coordination increases or the ratio increases, so that the higher the degree of need for security, the more self-propelled robots cooperate in security work Since the number of robots is large, or the proportion of self-propelled robots that perform security work in cooperation increases, necessary security work can be reliably performed.

(Other embodiments)
Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and can be appropriately changed without departing from the technical idea of the present invention.

  In the above embodiment, the self-propelled robots 2A to 4D have been described as including the abnormal state detecting unit 236, but the management device 3 may include the abnormal state detecting unit. In that case, the self-propelled robots 2 </ b> A to 4 </ b> D may transmit the status information to the management device 3 in real time, so that the abnormal status detection unit included in the management device 3 detects the abnormal status.

  In the above embodiment, the storage unit 31 of the management device 3 has been described as storing the abnormal state history information 313 for managing the state information and the like when the abnormal state in the work area 10 is detected. Alternatively, the storage unit 31 may store state information and the like when no abnormal state is detected as a history.

  Further, in the above embodiment, the learning unit 323 of the management device 3 determines that the abnormal state history information 313 satisfies a predetermined learning condition (for example, the number of data stored in the abnormal state history information 313 exceeds a predetermined number). ), It has been described that the learning model 314 is created based on the data accumulated in the abnormal state history information 313. However, even after the learning unit 323 satisfies a predetermined learning condition, the learning state The learning model 314 may be updated based on the data further stored in the 313.

  In the above embodiment, the security level determination unit 324 of the management device 3 determines the security level based on the learning model 314 that has learned a predetermined security level determination criterion when the manager M determines the security level. As described above, the security level determination unit 324 determines the security level based on the abnormal state detected by the abnormal state detection unit 236 of the security unit 23 and the abnormality detection factor without using the learning model 314. It may be.

  Further, in the above embodiment, the robot control program 213 is stored in the storage unit 21 and the robot group management program 315 is described as being stored in the storage unit 31. The file may be provided by being recorded on a computer-readable recording medium such as a CD-ROM and a DVD. Further, the robot control program 213 and the robot group management program 315 may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network.

DESCRIPTION OF SYMBOLS 1 ... Robot group management system, 2 ... Robot group 2A-2D ... Self-propelled robot, 3 ... Management device, 4 ... Administrator terminal,
5 ... network, 10 ... work area,
20 position detection unit, 21 storage unit, 22 cleaning unit,
23: security unit (acquisition unit), 24: traveling unit,
25: control unit, 26: communication unit, 27: battery,
Reference numeral 30: input unit, 31: storage unit, 32: control unit, 33: communication unit,
34 output unit, 210 map information, 211 user information,
212: work plan information, 213: robot control program,
230: spherical camera, 231, high-resolution camera, 232: human sensor,
233 temperature / humidity sensor 234 gas leak sensor 235 microphone
236: Abnormal state detection unit, 250: Work mode switching unit
251, a concurrent mode execution unit; 252, a security work alone mode execution unit;
253: Security work cooperation mode execution unit
310: master map information, 311: master user information,
312: Master work plan information, 313: Abnormal state history information,
314: learning model, 315: robot group management program,
320: work plan creation unit, 321: abnormal state history management unit
322: security level reception unit, 323: learning unit,
324: security level determination unit, 325: work mode determination unit (determination unit)
326: work plan update unit, 327: report processing unit, M: manager

Claims (6)

In the work area, a plurality of self-propelled robots that perform first work related to security and a second work different from the first work in a parallel or switchable manner, and a management device that manages the plurality of self-propelled robots, A robot group management system including
Each of the plurality of self-propelled robots includes:
An acquisition unit that acquires state information indicating a state of the work area by the first work,
The management device,
Based on the state information acquired by the acquisition unit, the determination unit that determines a work to be performed by each of the plurality of self-propelled robots among the first work and the second work,
A robot group management system, characterized in that: Each of the plurality of self-propelled robots performs the first work and the second work in a parallel or switchable manner according to a plurality of work modes,
The determination unit determines, as the work, a work mode for each of the plurality of self-propelled robots among the plurality of work modes,
The working mode is
A parallel mode for performing the first work and the second work in parallel;
A first work alone mode for performing the first work alone;
A first work cooperation mode in which the first work is performed in cooperation with the other self-propelled robot;
Including at least
The robot group management system according to claim 1, wherein: The management device,
Based on the state information acquired by the acquisition unit, further comprising a security level determination unit that determines a security level indicating the degree of security required,
The determination unit determines the work based on the security level determined by the security level determination unit,
The robot group management system according to claim 1 or 2, wherein: The security level determination unit, based on the learning model created based on the past state information acquired by the acquisition unit and the current state information acquired by the acquisition unit, the security level, judge,
The robot group management system according to claim 3, wherein: The self-propelled robot performs the first work among the plurality of self-propelled robots in cooperation with another self-propelled robot based on the security level determined by the security level determination unit. Determine the number or proportion of mobile robots,
The robot group management system according to claim 3 or 4, wherein The determining unit is based on the security level determined by the security level determination unit, the higher the degree that the security is required, the higher the number or the higher the ratio, the number or Determine the proportion,
The robot group management system according to claim 5, wherein:

Images