Secret-related place supervision robot system and working method thereof

Abstract

The invention discloses a confidential place supervision robot system and a working method thereof. The system comprises a supervision and management robot local platform and an intelligent control center. The supervision robot local platform comprises a first switch, and an autonomous walking platform, a special tool library, a first network bridge, a visual perception subsystem, a navigation positioning subsystem, a special storage cabinet system, a behavior identification subsystem, a driving subsystem and an alarm subsystem which are electrically connected with the first switch and perform information interaction with the first switch; the intelligent control center comprises a second switch, and a main control terminal, a second network bridge, a man-machine interaction subsystem and an expert knowledge system which are electrically connected with the second switch and perform information interaction with the second switch; the first network bridge and the second network bridge perform information interaction to realize communication transmission between the intelligent control center and the supervision robot local platform; a key administrator and a behavior supervisor are replaced by the intelligent robot, intelligent management of keys or access control cards in secret-related places and personnel behavior supervision are completed, the management efficiency and standardization are improved, and manpower is liberated.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent mobile robots, and in particular relates to a robot system for monitoring confidential places and a working method thereof. Background Art

[0002] At present, in order to ensure the security of confidential information, some confidential places require that keys must not be carried around and kept. When entering, more than two people must enter and exit together. During this period, the role of one person is mainly limited to supervising the behavior of the entrants to avoid violations. This not only takes up a lot of work time for one staff member, but also makes it difficult to enter the place to work during special periods such as holidays. At present, there are relatively many personnel behavior recognition and monitoring systems in confidential places in China, which mainly achieve supervision purposes by installing a large number of cameras, but there are no intelligent robots that combine key control, behavior tracking and supervision, and real-time alarm for violations. Especially when the target object is in the monitoring blind spot, it is easy to lose control and lose control. The efficiency of equipment use and supervision and management needs to be improved. Summary of the invention

[0003] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a robot system for supervising confidential places and a working method thereof, so as to realize multi-angle intelligent management and control, high supervision efficiency, fast emergency response speed, and save manpower.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A robot system for monitoring confidential places, including a local platform for monitoring and managing robots and an intelligent control center;

[0006] The local platform of the secret-related place supervision robot includes a first switch and an autonomous walking platform electrically connected to it and interacting with it, a special tool library, a first bridge, a visual perception subsystem, a navigation and positioning subsystem, a special storage cabinet system, a behavior recognition subsystem, a drive subsystem and an alarm subsystem;

[0007] The intelligent control center includes a second switch and a main control terminal electrically connected thereto and performing information interaction, a second network bridge, a human-computer interaction subsystem and an expert knowledge system;

[0008] The first bridge and the second bridge interact with each other to realize the communication transmission between the intelligent control center and the local platform of the supervisory robot;

[0009] The visual perception subsystem is used to detect and photograph the position, environment, biological characteristics, behavioral characteristics, and movement trajectory of the supervision and management robot and the target object;

[0010] The navigation and positioning subsystem is used to obtain its own position and posture information, the target object position information, and the movement trajectory information, and perform path planning to give motion control instructions for the autonomous walking platform;

[0011] The dedicated storage cabinet subsystem is used to store access keys and access cards for confidential places, and to track and locate keys and access cards through RFID tags;

[0012] The behavior recognition subsystem confirms the target object's human skeleton key point information through the rules pre-set by the expert knowledge base and the artificial intelligence visual algorithm based on the neural network, and analyzes and processes the target object's behavior and movement trajectory obtained by the visual perception subsystem;

[0013] The alarm subsystem is used to issue an alarm;

[0014] The driving subsystem is used to drive the autonomous walking platform;

[0015] The human-computer interaction subsystem is used to display the operating status of each sensor in real time, visualize the data processing information sent back by the second switch, and send the command and control instructions directly input by people to the main control terminal through the second switch;

[0016] The expert knowledge system autonomously screens and determines whether the target object's characteristics, behaviors, and movement trajectories comply with instructions or permissions based on pre-set rules and neural network artificial intelligence visual algorithms.

[0017] The present invention also has the following technical features:

[0018] Preferably, the visual perception subsystem includes a binocular high-definition camera module and a multi-degree-of-freedom liftable visual bearing platform. The binocular high-definition camera module is mounted on the multi-degree-of-freedom liftable visual bearing platform to reconstruct three-dimensional scenes, accurately locate target positions, and extract target object features, trajectories, and behavioral actions.

[0019] Preferably, the navigation and positioning subsystem includes an inertial sensor, a Beidou satellite navigation system positioning module and a laser radar, which can realize the position and posture measurement of the monitoring robot and target objects in confidential places, as well as the detection of obstacle information.

[0020] Preferably, the dedicated storage cabinet subsystem includes a key storage cabinet, an identity recognition device, an RFID smart reading tool, and a display screen to realize the storage, retrieval, and tracking of keys or access cards.

[0021] Preferably, the alarm subsystem includes an alarm controller, an audible and visual alarm, and a voice announcer, which analyze and process the threat level determination information to realize real-time alarm and supervision reminder of abnormal behavior of the target object.

[0022] Preferably, the expert knowledge system can also be used to establish an archiving, organization, analysis, and self-learning system for equipment control and human facial features and motion postures. During the treatment process, the system automatically identifies facial features and motion actions, thereby recommending or selecting the optimal treatment plan.

[0023] Furthermore, the expert knowledge system establishes an expert database system for real-time tracking of target objects, behavior detection, and disposal and response based on autonomous learning, human-machine assisted decision-making, and rapid reasoning mechanisms.

[0024] The present invention also protects a working method of the above-mentioned confidential place monitoring robot system, which specifically includes the following steps:

[0025] 1) The administrator or supervisor authorizes the target object to enter the confidential place to collect the key or access card through the main control terminal;

[0026] 2) The visual perception subsystem extracts and identifies the identity features of the target object through a binocular high-definition camera module;

[0027] 3) The dedicated storage cabinet subsystem reviews the target object’s permission information, issues keys or access cards to users who meet the permission requirements, terminates tasks for users who do not meet the permission requirements, and issues an alarm message;

[0028] 4) The visual perception subsystem analyzes the target object's location information and the environment information, and reconstructs the environment in three dimensions. According to the target object's location and environment, the camera is adjusted to a suitable angle and height, and the target object's body movements and movement trajectories are recorded and photographed;

[0029] 5) The navigation and positioning subsystem performs path planning, guides the robot's local platform to automatically track the target object and supervise its operation;

[0030] 6) The behavior recognition subsystem confirms the key point information of the target object's human skeleton through a neural network artificial intelligence visual algorithm, outputs the behavior action and movement trajectory data to match the standard data, and determines whether the target object's behavior action is abnormal;

[0031] 7) The alarm subsystem analyzes and processes the determined threat level through the alarm controller. For low-threat-level abnormal behaviors, the alarm subsystem uses voice reminder mode to urge the target object to correct the abnormal behavior. For high-threat-level or recurring low-threat-level abnormal behaviors, the alarm subsystem reminds the target object to correct its behavior. At the same time, it sends an alarm to the on-duty personnel and supervisors, pushes the disposal plan, and prompts them to assign a special person to deal with it on the spot;

[0032] 8) The manual backstage confirms the disposal plan and dispatches personnel to carry out on-site disposal;

[0033] 9) The intelligent control center plans the action path and displays the disposal screen in real time;

[0034] 10) The manual background confirms that the threat has been eliminated, passes the data processing information to the main control terminal, and cancels the alarm information;

[0035] 11) After the target object completes the task and leaves the confidential place, the dedicated storage cabinet subsystem will identify the returned key or access card and confirm its return, and the task is completed.

[0036] Compared with the prior art, the present invention has the following technical effects:

[0037] The confidential place supervision robot system of the present invention realizes intelligent management of access keys or access cards of confidential places by a special storage cabinet system, and the intelligent vision module provides facial feature acquisition, precise position positioning, three-dimensional reconstruction of environmental information and capture of human body movements and movement trajectories of the target object, guiding the robot to conduct full-process and all-round supervision and management of the target object's behavior after entering the confidential place, effectively solving the problems of current access key control mode for confidential places, backward behavior supervision and management means, slow perception of abnormal behavior, and delayed emergency response.

[0038] The robot system for supervising confidential places of the present invention has a dedicated storage cabinet system based on biometrics and RFID recognition technology, which can perform key location positioning, trajectory restoration, authority management, remote approval, emergency management and log management, thereby realizing intelligent control of access keys or access cards in confidential places.

[0039] The robot system for monitoring confidential places of the present invention, its visual perception subsystem, based on the target intelligent detection and positioning technology of three-dimensional vision, can reconstruct three-dimensional scenes, accurately locate the position of target objects, and capture the facial features, human body movements and behavioral trajectories of target objects, and can accurately obtain the three-dimensional information, position information, environmental information, and behavioral action and trajectory information of the target object.

[0040] The confidential place monitoring robot system of the present invention has a behavior recognition subsystem, which confirms the key point information of the human skeleton of the target object based on a neural network artificial intelligence visual algorithm, outputs motion trajectory data to match standard motion trajectory data, and can perform behavior analysis and recognition.

[0041] The confidential place supervision robot system of the present invention is aimed at the security control of access control keys and personnel behavior control in confidential places. Through multiple key technologies such as facial feature and RFID recognition, three-dimensional scene intelligent analysis and construction, intelligent image and video detection, artificial intelligence visual algorithm, autonomous path planning, multi-sensor information fusion, etc., it can realize unmanned and intelligent management of the whole process of "authorized entry-key access control supervision, behavior supervision-early warning disposal-leaving" in confidential places. In the entry stage, it is determined whether the target object meets the authority requirements; in the supervision stage, it is determined whether the behavior and trajectory of the target object after entering the confidential place meet the requirements of the rules; in the early warning stage, it realizes the identification and judgment of abnormal behavior, real-time alarm and emergency disposal; in the departure stage, it realizes the tracking and management of access control keys, which can meet the requirements of the whole process of security control of entering confidential places in different scenarios and under different conditions, and the requirements of real-time and standardized behavior supervision and response disposal;

[0042] The confidential place supervision system is based on a neural network visual intelligent analysis algorithm. It uses intelligent robots to replace key managers and behavior supervisors to complete the intelligent management of keys or access cards in confidential places and the behavior supervision of people entering, completely freeing people from inefficient work. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a structural block diagram of the present invention;

[0044] Figure 2 Workflow diagram of the present invention. DETAILED DESCRIPTION

[0045] The specific contents of the present invention are further explained in detail below in conjunction with embodiments.

[0046] Reference Figure 1 ,This embodiment provides a confidential place supervision robot system, including a supervision and management robot local platform and an intelligent control center;

[0047] The local platform of the secret-related place supervision robot includes a first switch and an autonomous walking platform electrically connected to and interacting with the switch, a special tool library, a first network bridge, a visual perception subsystem, a navigation and positioning subsystem, a special storage cabinet system, a behavior recognition subsystem, a drive subsystem, and an alarm subsystem;

[0048] The intelligent control center includes a second switch and a main control terminal electrically connected thereto and performing information interaction, a second network bridge, a human-computer interaction subsystem and an expert knowledge system;

[0049] The first bridge and the second bridge exchange information to realize the communication transmission between the intelligent control center and the local platform of the supervisory robot;

[0050] The visual perception subsystem is used to detect and photograph the location, environment, biological characteristics, behavioral characteristics, and movement trajectory of the supervisory management robot and target object;

[0051] The navigation and positioning subsystem is used to obtain its own position and posture information, the target object's position information, and the movement trajectory information, and to perform path planning and give motion control instructions for the autonomous walking platform;

[0052] The dedicated storage cabinet system is used to store access keys and access cards in confidential places, and to track and locate keys and access cards through RFID tags;

[0053] The behavior recognition subsystem confirms the target object's human skeleton key point information through the rules pre-set by the expert knowledge base and the artificial intelligence visual algorithm based on the neural network, and analyzes and processes the target object's behavior and movement trajectory obtained by the visual perception subsystem;

[0054] The alarm subsystem is used to issue an alarm;

[0055] The driving subsystem is used to drive the autonomous walking platform;

[0056] The human-computer interaction subsystem is used to display the operating status of each sensor in real time, visualize the data processing information sent back by the second switch, and send the command and control instructions directly input by people to the main control terminal through the second switch;

[0057] The expert knowledge system autonomously screens and determines whether the target object's characteristics, behaviors, and movement trajectories comply with instructions or permissions based on pre-set rules and neural network artificial intelligence visual algorithms.

[0058] The visual perception subsystem includes a binocular high-definition camera module and a multi-degree-of-freedom liftable visual carrier platform. The binocular high-definition camera module is mounted on the multi-degree-of-freedom liftable visual carrier platform and is used to reconstruct three-dimensional scenes, accurately locate target positions, and extract target object features, trajectories, and behavioral actions.

[0059] The navigation and positioning subsystem includes inertial sensors, Beidou satellite navigation system positioning module and lidar, which can realize the position and posture measurement of the monitoring robot and target objects in confidential places, as well as obstacle information detection.

[0060] The dedicated storage cabinet subsystem includes a key storage cabinet, identity recognition equipment, RFID smart reading tools, and a display screen to enable the storage, retrieval, and tracking of keys or access cards.

[0061] The alarm subsystem includes an alarm controller, an audible and visual alarm, and a voice announcer, which analyzes and processes the threat level determination information to achieve real-time alarm and supervision reminders for abnormal behavior of the target object.

[0062] The expert knowledge system can also be used to establish equipment control and the archiving, organization, analysis, and self-learning systems of human facial features and motion postures. During the treatment process, the system automatically identifies facial features and motion actions to recommend or select the optimal treatment plan.

[0063] The expert knowledge system is based on autonomous learning, human-machine assisted decision-making and rapid reasoning mechanisms to establish an expert database system for real-time tracking of target objects, behavior detection, and disposal and response.

[0064] Reference Figure 2 The working method of the confidential place monitoring robot system of the present invention specifically includes the following steps:

[0065] 1) The administrator or supervisor authorizes the target object to enter the confidential place to collect the key or access card through the main control terminal;

[0066] 2) The visual perception subsystem extracts and identifies the identity features of the target object through a binocular high-definition camera module;

[0067] 3) The dedicated storage cabinet subsystem reviews the target object’s permission information, issues keys or access cards to users who meet the permission requirements, terminates tasks for users who do not meet the permission requirements, and issues an alarm message;

[0068] 4) The visual perception subsystem analyzes the target object's location information and the environment information, and reconstructs the environment in three dimensions. According to the target object's location and environment, the camera is adjusted to a suitable angle and height, and the target object's body movements and movement trajectories are recorded and photographed;

[0069] 5) The navigation and positioning subsystem performs path planning, guides the robot's local platform to automatically track the target object and supervise its operation;

[0070] 6) The behavior recognition subsystem confirms the key point information of the target object's human skeleton through a neural network artificial intelligence visual algorithm, outputs the behavior action and movement trajectory data to match the standard data, and determines whether the target object's behavior action is abnormal;

[0071] 7) The alarm subsystem analyzes and processes the determined threat level through the alarm controller. For low-threat-level abnormal behaviors, the alarm subsystem uses voice reminder mode to urge the target object to correct the abnormal behavior. For high-threat-level or recurring low-threat-level abnormal behaviors, the alarm subsystem reminds the target object to correct its behavior. At the same time, it sends an alarm to the on-duty personnel and supervisors, pushes the disposal plan, and prompts them to assign a special person to deal with it on the spot;

[0072] 8) The manual backstage confirms the disposal plan and dispatches personnel to carry out on-site disposal;

[0073] 9) The intelligent control center plans the action path and displays the disposal screen in real time;

[0074] 10) The manual background confirms that the threat has been eliminated, passes the data processing information to the main control terminal, and cancels the alarm information;

[0075] 11) After the target object completes the task and leaves the confidential place, the dedicated storage cabinet subsystem will identify the returned key or access card and confirm its return, and the task is completed.

[0076] The present invention has been described in detail with reference to the above embodiments. Those skilled in the art should understand that the specific embodiments of the present invention may still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be included in the scope of the present invention.

Claims

1. A robot system for monitoring confidential places, characterized in that: Including supervision and management of robot local platform and intelligent control center; The local platform of the secret-related place supervision robot includes a first switch and an autonomous walking platform electrically connected to it and interacting with it, a special tool library, a first bridge, a visual perception subsystem, a navigation and positioning subsystem, a special storage cabinet system, a behavior recognition subsystem, a drive subsystem and an alarm subsystem; The intelligent control center includes a second switch and a main control terminal electrically connected thereto and performing information interaction, a second network bridge, a human-computer interaction subsystem and an expert knowledge system; The first bridge and the second bridge interact with each other to realize the communication transmission between the intelligent control center and the local platform of the supervisory robot; The visual perception subsystem is used to detect and photograph the position, environment, biological characteristics, behavioral characteristics, and movement trajectory of the supervision and management robot and the target object; The navigation and positioning subsystem is used to obtain its own position and posture information, the target object position information, and the movement trajectory information, and perform path planning to give motion control instructions for the autonomous walking platform; The dedicated storage cabinet subsystem is used to store access keys and access cards for confidential places, and to track and locate keys and access cards through RFID tags; The behavior recognition subsystem confirms the target object's human skeleton key point information through the rules pre-set by the expert knowledge base and the artificial intelligence visual algorithm based on the neural network, and analyzes and processes the target object's behavior and movement trajectory obtained by the visual perception subsystem; The alarm subsystem is used to issue an alarm; The driving subsystem is used to drive the autonomous walking platform; The human-computer interaction subsystem is used to display the operating status of each sensor in real time, visualize the data processing information sent back by the second switch, and send the command and control instructions directly input by people to the main control terminal through the second switch; The expert knowledge system autonomously screens and determines whether the target object's characteristics, behaviors, and movement trajectories comply with instructions or permissions based on pre-set rules and neural network artificial intelligence visual algorithms.

2. The robot system for monitoring confidential places according to claim 1, characterized in that: The visual perception subsystem includes a binocular high-definition camera module and a multi-degree-of-freedom liftable visual bearing platform. The binocular high-definition camera module is mounted on the multi-degree-of-freedom liftable visual bearing platform to reconstruct three-dimensional scenes, accurately locate target positions, and extract target object features, trajectories, and behavioral actions.

3. The robot system for monitoring confidential places as claimed in claim 1, characterized in that: The navigation and positioning subsystem includes an inertial sensor, a Beidou satellite navigation system positioning module and a laser radar, which can realize the position and posture measurement of the monitoring robot and target objects in confidential places, as well as the obstacle information detection.

4. The robot system for monitoring confidential places according to claim 1, characterized in that: The dedicated storage cabinet subsystem includes a key storage cabinet, an identity recognition device, an RFID smart reading tool, and a display screen to realize the storage, access, and tracking of keys or access cards.

5. The robot system for monitoring confidential places as claimed in claim 1, characterized in that: The alarm subsystem includes an alarm controller, an audible and visual alarm, and a voice announcer, which analyze and process threat level determination information to achieve real-time alarm and supervision reminders for abnormal behaviors of target objects.

6. The robot system for monitoring confidential places as claimed in claim 1, characterized in that: The expert knowledge system can also be used to establish an archiving, organization, analysis, and self-learning system for equipment control and human facial features and motion postures. During the treatment process, the system automatically identifies facial features and motion actions, and recommends or selects the optimal treatment plan.

7. The robot system for monitoring confidential places as claimed in claim 6, characterized in that: The expert knowledge system is based on autonomous learning, human-machine assisted decision-making and rapid reasoning mechanisms to establish an expert database system for real-time tracking of target objects, behavior detection, and disposal and response.

8. A working method of the robot system for monitoring confidential places as claimed in any one of claims 1 to 7, characterized in that: The following steps are involved: 1) The administrator or supervisor authorizes the target object to enter the confidential place to collect the key or access card through the main control terminal; 2) The visual perception subsystem extracts and identifies the identity features of the target object through a binocular high-definition camera module; 3) The dedicated storage cabinet subsystem reviews the target object’s permission information, issues keys or access cards to users who meet the permission requirements, terminates tasks for users who do not meet the permission requirements, and issues an alarm message; 4) The visual perception subsystem analyzes the target object's location information and the environment information, and reconstructs the environment in three dimensions. According to the target object's location and environment, the camera is adjusted to a suitable angle and height, and the target object's body movements and movement trajectories are recorded and photographed; 5) The navigation and positioning subsystem performs path planning, guides the robot's local platform to automatically track the target object and supervise its operation; 6) The behavior recognition subsystem confirms the key point information of the target object's human skeleton through a neural network artificial intelligence visual algorithm, outputs the behavior action and movement trajectory data to match the standard data, and determines whether the target object's behavior action is abnormal; 7) The alarm subsystem analyzes and processes the determined threat level through the alarm controller. For low-threat-level abnormal behaviors, the alarm subsystem uses voice reminder mode to urge the target object to correct the abnormal behavior. For high-threat-level or recurring low-threat-level abnormal behaviors, the alarm subsystem reminds the target object to correct its behavior. At the same time, it sends an alarm to the on-duty personnel and supervisors, pushes the disposal plan, and prompts them to assign a special person to deal with it on the spot; 8) The manual backstage confirms the disposal plan and dispatches personnel to carry out on-site disposal; 9) The intelligent control center plans the action path and displays the disposal screen in real time; 10) The manual background confirms that the threat has been eliminated, passes the data processing information to the main control terminal, and cancels the alarm information; 11) After the target object completes the task and leaves the confidential place, the dedicated storage cabinet subsystem will identify the returned key or access card and confirm its return, and the task is completed.

Images