Unmanned intelligent monitoring rescue system based on bidirectional information flow management

Abstract

The invention provides an unmanned intelligent monitoring rescue system based on bidirectional information flow management, and relates to the technical field of emergency rescue and intelligent monitoring, and the system comprises the following modules: an unmanned aerial vehicle monitoring module which is used for carrying out investigation and data collection in the air; the information acquisition module is used for collecting information of a preset area from different angles and aspects; the data transmission module and the central server are used for receiving various data in the data transmission module and preprocessing the various data; performing deep analysis on the preprocessed data through a preset intelligent game algorithm; the rescue module is used for executing the instruction of the central server; the two-way information exchange module is used for a user to coordinate the work among the modules; according to the invention, a more scientific and reasonable rescue strategy can be formulated through an intelligent game algorithm in the central server, and the rescue module is used for replacing rescue workers for rescue, so that the success rate of rescue and the intelligent level of rescue are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency rescue and intelligent monitoring, and in particular to an unmanned intelligent monitoring and rescue system based on bidirectional information flow management. Background Art

[0002] In the hot summer, swimming has become a popular activity for people to seek coolness and fun. However, drowning during swimming is a common and dangerous accident. According to national statistics in 2023, drowning accidents occurred frequently, with a total of 17,504 cases, resulting in the loss of 30,456 precious lives and 54,132 injuries. Among them, children and adolescents are at extremely high risk. It can be seen that in drowning rescue work, it is crucial to detect the drowning person in time, accurately judge his condition and quickly take effective rescue measures.

[0003] Traditional drowning rescue methods mostly rely on manual rescue. However, drowning incidents are often difficult to detect in a timely manner, resulting in serious delays in the start of rescue. In addition, rescuers have difficulty obtaining information after arriving at the scene. Rescuers have limited knowledge of key information such as the drowning person's location, underwater conditions, and drowning time, and rely heavily on experience and intuition for rescue, which makes the rescue time longer and the success rate lower. Moreover, during the rescue process, the information communication between on-site rescue personnel and the command center is often not timely and accurate enough. The command center finds it difficult to grasp the dynamic changes at the rescue site in real time and cannot make timely decision adjustments and resource allocation based on actual conditions. Rescuers often suffer casualties due to physical exhaustion, affecting the efficiency and effectiveness of the rescue. Summary of the Invention

[0004] The purpose of the present invention is to provide an unmanned intelligent monitoring and rescue system based on two-way information flow management to solve the technical problems existing in the prior art.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] An unmanned intelligent monitoring and rescue system based on two-way information flow management includes the following modules: a drone monitoring module equipped with multiple sets of high-definition camera equipment for aerial reconnaissance and data collection, and capturing images and environmental parameters in real time through comprehensive scanning of a preset area; an information acquisition module, which uses multiple sets of the high-definition camera equipment to collect information about the preset area from different angles and aspects, and can obtain dynamic changes in the preset area in real time; a data transmission module, which transmits the data collected in the information acquisition module to a central server through advanced communication technology, and has a data encryption terminal; a central server, which receives various data in the data transmission module and processes the various data The system pre-processes the data; performs in-depth analysis on the pre-processed data through a preset intelligent game algorithm; based on the results of the in-depth analysis, the intelligent game algorithm generates multiple sets of rescue strategies and notifies the local rescue department; generates corresponding instructions for the multiple sets of rescue strategies, and transmits the instructions to the rescue module to instruct it to perform corresponding rescue tasks; the rescue module is set on the drone and includes several rescue equipment, which is used to execute the instructions of the central server; the two-way information exchange module allows users to coordinate the work between the above modules, reasonably allocate the order of information transmission according to the priority and urgency of the information, and ensure the coordination of the rescue modules and the signal interaction with the affected people.

[0007] Furthermore, the high-definition camera equipment in the drone monitoring module includes:

[0008] A high-definition camera is used to obtain high-quality images of a preset area to help the central server perform visual analysis; a thermal imager is used to penetrate obstacles and dark environments to promptly detect hot spots and signs of life in a preset area; a gas detector is used to measure environmental parameters; and a wind speed sensor is used to measure and predict wind speed.

[0009] Furthermore, the data encryption terminal in the data transmission module includes the following units: a key generation unit, which is responsible for generating keys for encrypting and decrypting data, and is located on the sender; an encryption algorithm unit, which contains multiple sets of encryption algorithms, and is used to encrypt the data in the data transmission module; a decryption algorithm unit, which corresponds to the encryption algorithm and is used to decrypt the encrypted data and restore the original data, and is located on the receiver; an identity authentication unit, which is used to verify the identities of the sender and the receiver, so that data can be transmitted and processed between the two legal parties.

[0010] Furthermore, the intelligent game algorithm in the central server includes: cooperative game intelligence, all modules share a global reward function, and generate different rescue strategies for different situations, and integrate different rescue strategies to work together to speed up rescue operations; from the perspective of learning paradigm, it can be divided into: independent learning, joint learning and collaborative learning; adversarial game intelligence, used to combat adverse factors in the rescue process, timely adjust the confrontation strategy according to the dynamic changes of the preset area, and continuously optimize the rescue strategy; hybrid game intelligence, organically combine situations that require cooperation or confrontation in the rescue process, form a comprehensive response plan, and flexibly adjust according to the complex situations that arise in the actual rescue process.

[0011] Furthermore, the preprocessing in the central server includes: a data cleaning unit, which is used to remove duplicate data and redundant information, supplement missing data using a preset method for missing values, and compress them; a noise removal unit, which identifies the compressed data, removes noise and outliers, and classifies and labels the data according to its type and different sources; a deep processing unit, which sorts and synchronizes the classified and labeled data through a preset deep learning algorithm.

[0012] Furthermore, the rescue equipment in the rescue module includes: lifebuoy, life rope, rescue robot, unmanned vehicle, fire-fighting robot, AED and underwater small excavation equipment.

[0013] Compared with the prior art, the present invention has the following beneficial effects:

[0014] The present invention can quickly and accurately collect and process information on the area requiring rescue through the setting of the drone monitoring module, the information collection module and the data transmission module, and transmit the collected and processed data to the central server. The intelligent game algorithm in the central server can formulate a more scientific and reasonable rescue strategy, and the rescue module can replace the rescue personnel for rescue, thereby improving the success rate of rescue and the intelligence level of rescue; by replacing manual rescue with machines, the risk of casualties is reduced and the rescue time is shortened. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the present invention; DETAILED DESCRIPTION

[0016] To make the contents of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Identical parts are denoted by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, and the terms "inward" and "outward" refer to directions toward or away from the geometric center of a particular component, respectively.

[0017] like Figure 1 As shown, this embodiment provides an unmanned intelligent monitoring and rescue system based on two-way information flow management, including the following modules:

[0018] The drone monitoring module is equipped with multiple sets of high-definition camera equipment, which include: a high-resolution high-definition camera for obtaining high-quality graphics of the preset area to help the central server perform visual analysis; a thermal imager for penetrating obstacles and dark environments to promptly detect hot spots and signs of life in the preset area. The thermal imager is suitable for nighttime or large water areas; a gas detector for measuring environmental parameters; a wind speed sensor for measuring and predicting wind speed; the gas detector can detect whether there are harmful gases in the water, such as hydrogen sulfide and carbon monoxide, to help rescue equipment take protective measures in advance to avoid damage to equipment parts caused by gas pollution; and detect polluted gases in the water to determine the cause of drowning and rescue strategies; on the one hand, the wind speed sensor can help drones The flight posture and speed are adjusted according to the wind speed to improve the stability and safety of the UAV, thereby ensuring the smooth progress of monitoring and rescue missions. On the other hand, the wind speed sensor can infer the waves and water flow conditions on the water surface. Through the data of the wind speed sensor, the drift direction and speed of the drowning person in the water can be inferred, providing important reference parameters for subsequent rescue, thereby narrowing the search range and improving rescue efficiency; this module is used for aerial reconnaissance and data collection, and through a comprehensive scan of the preset area, it captures images and environmental parameters in real time; it provides continuous images and data to the central server, which can quickly rescue drowning people and shorten the time from discovering the drowning incident to starting the rescue operation. In addition, the UAV monitoring module can obtain a wide field of view from the air, without being hindered by complex terrain, increasing the possibility of successful rescue.

[0019] The information acquisition module collects information about a preset area from different angles and aspects through multiple sets of high-definition camera equipment. The information includes environmental information and the status of the drowning person. The information acquisition module can obtain dynamic changes in the preset area in real time, which helps to quickly determine the key areas for rescue and improve the success rate of rescue.

[0020] The data transmission module transmits the data collected in the information collection module to the central server through advanced communication technology, and the communication technology includes: 5G communication, Beidou positioning system and Wi-Fi; the data transmission module has a data encryption terminal; the data encryption terminal includes the following units: a key generation unit, responsible for generating keys for encrypting and decrypting data, the quality and security of the keys are particularly important for the encryption effect, and it is located on the sender; an encryption algorithm unit, including multiple groups of encryption algorithms, which are used to encrypt the data in the data transmission module, and the encryption algorithms are advanced encryption standards, asymmetric encryption algorithms, etc.; a decryption algorithm unit, which corresponds to the encryption algorithm, is used to decrypt the encrypted data and restore the original data, and it is located on the receiver; an identity authentication unit, which is used to verify the identities of the sender and the receiver, so that the data can be transmitted and processed between the two legal parties; the encryption terminal is set to protect the integrity of the data and prevent the data from being attacked; thereby avoiding damage and interference to the rescue process caused by hackers or other malicious attackers, thereby ensuring the smooth progress of the entire process of the present invention.

[0021] The central server receives various data in the data transmission module and pre-processes the various data; the pre-processing includes: a data cleaning unit for removing duplicate data and redundant information, and using a preset method to supplement missing data for missing values, the preset method is to fill the mean value, median and infer based on other relevant data; and compress the data, thereby reducing storage space and improving processing efficiency; a noise removal unit for identifying the compressed data, removing noise and outliers, and classifying and labeling the data according to the type and different sources of the data, and providing a basis for supervised learning for the machine learning or deep learning algorithm of the deep processing unit through the labeled data; the deep processing unit for the data after the classification through the preset deep learning algorithm The system sorts and synchronizes the classified and annotated data to ensure the consistency and accuracy of the data in time; conducts in-depth analysis of the pre-processed data through a preset intelligent game algorithm, which models the interaction relationship between each intelligent agent in a complex dynamic multi-agent system, and effectively solves the optimal goals or strategies of different game participants; based on the results of the in-depth analysis, the intelligent game algorithm generates multiple sets of rescue strategies and notifies the local rescue department, which includes: fire brigade, rescue medical team, additional volunteers, etc.; generates corresponding instructions for the multiple sets of rescue strategies, and transmits the instructions to the rescue module to command it to perform the corresponding rescue mission; the central server can control the flight path and status of the drone.

[0022] The rescue module is installed on the drone and includes several rescue equipment, which is used to execute the instructions of the central server; the rescue equipment includes: lifebuoy, life rope, rescue robot, unmanned vehicle, fire-fighting robot, AED and underwater small excavation equipment; the underwater small excavation equipment can clear underwater obstacles near the drowning person, so that the rescue equipment can provide sufficient rescue space in the narrow underwater space, ensuring the safety of the rescue process.

[0023] A two-way information exchange module includes: a data acquisition unit, a data conversion unit, a data analysis unit, a data storage unit, a data reporting unit, a data visualization unit, a data security unit, a data update unit, a data sharing unit and a data sharing unit; the two-way information exchange module user coordinates the work between the above modules, reasonably allocates the order of information transmission according to the priority and urgency of the information, and ensures the coordination and cooperation of the rescue modules and the signal interaction with the affected people.

[0024] Specifically, the intelligent game algorithm in the central server includes:

[0025] Cooperative game intelligence, all modules share a global reward function, that is, r = r1 = r2 = ... = r n From the perspective of game solving, since all agents share the reward function, this requires all agents to cooperate with each other and learn an optimal joint strategy π(a|s)=π( <a1,a2…,a n >|s) to maximize the common cumulative reward, which is essentially to find a social optimal solution; and generate different rescue strategies for different situations, and integrate different rescue strategies to work together to speed up the rescue operation; from the perspective of learning paradigm, it can be divided into: independent learning, joint learning and collaborative learning; in the independent learning, taking Q learning as an example, each agent i only needs to learn its own Q i Value function;

[0026] And adopt a greedy strategy To select the game action to execute; the essence of the independent learning method is to make an independent assumption about the two box strategies, that is, π(a|s)=π( <a1,a2…,a n >|s)=Π i π i (a i |s), and the cooperation between agents depends only on the shared reward value r=r1=r2=…=r n To achieve;

[0027] In the joint learning, taking the Q-learning method as an example, the super agent needs to learn a joint Q-value function: And adopt the greedy strategy π(a|s)=argmax a∈A Q(s,a) is used to extract the optimal joint action a= <a1,a2,…,a n >, each action component a i Send it to the corresponding agent i for execution;

[0028] The collaborative learning described above combines the advantages of independent learning and joint learning. In a general collaborative graph, one agent can be connected to more than two agents. However, to simplify the interaction between agents, we can further assume that each reward function involves at most two agents. In this case, the joint Q-value function can be further simplified to the following form:

[0029] Where i∈E represents agent i, and the edge (i,j)∈E of the graph indicates that there is an interactive dependency relationship between agents i and j. When each agent i makes an action selection, it only needs to coordinate with the agent j∈Γ(i) that has a dependency relationship with it, without considering unrelated agents. This limits the learning process to a local joint strategy space.

[0030] Adversarial game intelligence is used to combat adverse factors during the rescue process, timely adjust the confrontation strategy based on the dynamic changes in the preset area, and continuously optimize the rescue strategy. In this adversarial game intelligence, the two parties are in an adversarial and competitive relationship. The adversarial game aims at a more complex implicit equilibrium, which poses challenges to the efficiency, diversity, and robustness of the strategy solution.

[0031] Hybrid game intelligence organically combines situations that require cooperation or confrontation during the rescue process to form a comprehensive response plan and flexibly adjusts according to the complex situations that arise during the actual rescue process. In hybrid game intelligence, all intelligent agents have their own optimization goals, and organic collaboration is required between the participants to ensure both the optimization of individual goals and the maximization of the overall benefits of the system.

[0032] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. An unmanned intelligent monitoring and rescue system based on two-way information flow management, characterized by: Includes the following modules: The drone monitoring module is equipped with multiple sets of high-definition cameras for aerial reconnaissance and data collection. It captures images and environmental parameters in real time by comprehensively scanning the preset area. An information acquisition module collects information of a preset area from different angles and aspects through multiple sets of high-definition camera devices, and can obtain dynamic changes of the preset area in real time; A data transmission module, which transmits the data collected by the information collection module to a central server through advanced communication technology and has a data encryption terminal; The central server receives various data from the data transmission module and pre-processes the data; performs in-depth analysis on the pre-processed data using a preset intelligent game algorithm; based on the results of the in-depth analysis, the intelligent game algorithm generates multiple sets of rescue strategies and notifies the local rescue department; generates corresponding instructions for the multiple sets of rescue strategies, and transmits the instructions to the rescue module to instruct it to execute the corresponding rescue mission; A rescue module, provided on the drone, includes several rescue devices and is used to execute instructions from the central server; In the two-way information exchange module, users coordinate the work between the above modules, reasonably allocate the order of information transmission according to the priority and urgency of the information, and ensure the coordination of the rescue modules and the signal interaction with the affected people.

2. The unmanned intelligent monitoring and rescue system based on two-way information flow management according to claim 1 is characterized by: The high-definition camera equipment in the drone monitoring module includes: A high-definition camera for obtaining high-quality images of a preset area to assist the central server in performing visual analysis; Thermal imagers, used to penetrate obstacles and dark environments to promptly detect hot spots and signs of life within a preset area; Gas detectors, used to measure environmental parameters; Wind speed sensor, used to measure and predict wind speed.

3. The unmanned intelligent monitoring and rescue system based on two-way information flow management according to claim 1 is characterized by: The data encryption terminal in the data transmission module includes the following units: The key generation unit is responsible for generating keys for encrypting and decrypting data and is located at the sender; An encryption algorithm unit, comprising multiple encryption algorithms, for encrypting data in the data transmission module; A decryption algorithm unit, which corresponds to the encryption algorithm and is used to decrypt the encrypted data and restore the original data, and is located at the receiving end; The identity authentication unit is used to verify the identities of the sender and receiver, so that data can be transmitted and processed between the two legitimate parties.

4. The unmanned intelligent monitoring and rescue system based on two-way information flow management according to claim 1 is characterized by: The intelligent game algorithm in the central server includes: Cooperative game intelligence: All modules share a global reward function and generate different rescue strategies for different situations. They integrate different rescue strategies and make them work together to accelerate rescue operations. From the perspective of learning paradigm, it can be divided into: independent learning, joint learning, and collaborative learning. Countermeasure game intelligence is used to combat adverse factors during the rescue process, adjust the countermeasure strategy in a timely manner according to the dynamic changes of the preset area, and continuously optimize the rescue strategy; Hybrid game intelligence organically combines situations that require cooperation or confrontation during the rescue process to form a comprehensive response plan, and makes flexible adjustments based on the complex situations that arise during the actual rescue process.

5. The unmanned intelligent monitoring and rescue system based on two-way information flow management according to claim 1 is characterized in that: The pre-processing in the central server includes: Data cleaning unit, used to remove duplicate data and redundant information, supplement missing data using preset methods, and compress them; a noise removal unit, which identifies the compressed data, removes noise and outliers, and classifies and labels the data according to its type and origin; A deep processing unit sorts and synchronizes the classified and labeled data through a preset deep learning algorithm.

6. The unmanned intelligent monitoring and rescue system based on two-way information flow management according to claim 1 is characterized by: The rescue equipment in the rescue module includes: lifebuoy, life rope, rescue robot, unmanned vehicle, fire-fighting robot, AED and underwater small excavation equipment.

Images