Heterogeneous multi-agent rescue system for chemical enterprise fire

Abstract

The invention provides a heterogeneous multi-agent rescue system for chemical enterprise fire disasters, and belongs to the technical field of agent application. The system comprises an unmanned vehicle subsystem which is used for sensing first environment data in a chemical enterprise where a fire occurs in real time and serves as a base station; the unmanned aerial vehicle subsystem is used for performing coverage detection on a fire scene area where an obstacle exists or an unmanned vehicle cannot pass, so as to obtain second environment data and position information of each person in distress; the mobile terminal subsystem is used for acquiring the position information of each rescue worker in the chemical enterprise; the upper computer subsystem is used for fusing the first environment data and the second environment data to construct an internal map of the chemical enterprise, and performing path planning by combining the position information of each person in distress and the position information of each rescue worker; and optimal rescue path information corresponding to each rescue worker is determined and sent to the mobile terminal subsystem for navigation. According to the invention, the fire emergency rescue response speed can be improved.

Description

Technical Field

[0001] This application relates to the field of intelligent agent application technology, and in particular to a heterogeneous multi-agent rescue system for chemical plant fires. Background Technology

[0002] Chemical plants have densely packed production facilities and use highly flammable and explosive materials, making fires there prone to suddenness, rapid spread, and immense damage, easily causing significant casualties, property losses, and environmental pollution. Current fire rescue operations in chemical plants heavily rely on manual labor, lacking real-time awareness of the plant environment and the location of those in distress, resulting in low rescue efficiency and high risks. Summary of the Invention

[0003] The main purpose of this application is to propose a heterogeneous multi-agent rescue system for chemical plant fires, which aims to improve the response speed of emergency rescue in chemical plant fires and reduce rescue risks.

[0004] To achieve the above objectives, one aspect of this application proposes a heterogeneous multi-agent rescue system for chemical plant fires, comprising: The unmanned vehicle subsystem is used to perceive the first environmental data inside the chemical enterprise where a fire has occurred in real time and send it to the host computer subsystem. It also serves as a base station to provide positioning support for drones and mobile terminals in complex indoor environments. The unmanned aerial vehicle (UAV) subsystem is used to conduct coverage detection of fire areas within the chemical enterprise where there are obstacles or where unmanned vehicles cannot pass, in order to obtain detection data and send it to the host computer subsystem; wherein, the detection data includes second environmental data of the fire area and location information of each person in distress; The mobile terminal subsystem is used to acquire the location information of each rescuer inside the chemical enterprise and send it to the host computer subsystem. The host computer subsystem is used to fuse the first environmental data and the second environmental data to construct the latest internal map of the chemical enterprise, and then perform path planning based on the internal map of the chemical enterprise, the location information of each person in distress and the location information of each rescuer to determine the optimal rescue path information corresponding to each rescuer and send it to the mobile terminal subsystem. The mobile terminal subsystem is also used to provide real-time rescue route navigation to each rescuer based on the optimal rescue route information corresponding to each rescuer.

[0005] Furthermore, the unmanned vehicle subsystem is also used to adjust the unmanned vehicle movement path and mobile base station deployment when the environment inside the chemical enterprise changes.

[0006] Furthermore, the unmanned vehicle subsystem includes multiple unmanned vehicles; Each of the unmanned vehicles is equipped with a first UWB positioning module, a first environmental perception module, and a first communication module. The first UWB positioning module is used to provide a positioning reference for the unmanned vehicle subsystem and the mobile terminal subsystem. The first environmental perception module is used to monitor the local environment of the area where the unmanned vehicle is located in real time when the unmanned vehicle is inside the chemical enterprise. The first communication module is used to transmit data with the host computer subsystem. Each of the unmanned vehicles is equipped with a first control module, which is used to receive the optimal deployment path issued by the host computer subsystem in order to control the unmanned vehicle to avoid obstacles and move and park within the chemical enterprise.

[0007] Furthermore, the unmanned vehicle subsystem includes multiple unmanned vehicles, and there is a one-to-one correspondence between the multiple unmanned vehicles and the multiple unmanned vehicles; Each of the UAVs is equipped with a second UWB positioning module, a second environmental perception module, an image acquisition module, and a second communication module. The second UWB positioning module is used to determine the location information of the UAV. The second environmental perception module is used to monitor the local environment of the area where the UAV is located when the UAV is located in the fire area. The image acquisition module is used to acquire image data of the area where the UAV is located. The second communication module is used to transmit data with the host computer subsystem. Each of the aforementioned UAVs has a built-in onboard processing module. The onboard processing module is used to receive the search path issued by the host computer subsystem to control the UAV to conduct detection in the fire area. It is also used to identify the image data and, when it is identified that there are people in distress in the image data, to determine the location information of the people in distress based on the image data and the location information of the UAV.

[0008] Furthermore, the mobile terminal subsystem includes smart safety helmets configured for each rescuer; The smart safety helmet is equipped with a third UWB positioning module and a third communication module. The third UWB positioning module is used to determine the location information of the rescuer, and the third communication module is used to transmit data with the host computer subsystem. The smart safety helmet also has a display module integrated into the protective face shield, which is used to project and display the optimal rescue route information corresponding to the rescuer.

[0009] Furthermore, each of the unmanned vehicles is also equipped with a charging module, which is used to provide charging functionality for the unmanned aerial vehicle subsystem.

[0010] Furthermore, the first environmental perception module includes a first lidar device, a first gas sensor, a first temperature sensor, and a first smoke sensor. The first lidar device is used to acquire point cloud data of the local environment in the area where the unmanned vehicle is located. The first gas sensor is used to monitor the gas concentration in the area where the unmanned vehicle is located. The first temperature sensor is used to monitor the temperature in the area where the unmanned vehicle is located. The first smoke sensor is used to monitor the smoke concentration in the area where the unmanned vehicle is located.

[0011] Furthermore, the second environmental perception module includes a second lidar device, a second gas sensor, a second temperature sensor, and a second smoke sensor. The second lidar device is used to acquire point cloud data of the local environment in the area where the drone is located. The second gas sensor is used to monitor the gas concentration in the area where the drone is located. The second temperature sensor is used to monitor the temperature in the area where the drone is located. The second smoke sensor is used to monitor the smoke concentration in the area where the drone is located.

[0012] Furthermore, the mobile terminal subsystem is also used to acquire the physiological status information of each rescuer and send it to the host computer subsystem.

[0013] Furthermore, the host computer subsystem is also used to mark and display the locations of the distressed personnel and the rescue personnel on the internal map of the chemical enterprise.

[0014] This application includes at least the following beneficial effects: First, the unmanned vehicle subsystem perceives the first environmental data inside the chemical enterprise where the fire has occurred in real time. Then, the drone subsystem covers and detects the fire area inside the chemical enterprise where there are obstacles or where the unmanned vehicle cannot pass, in order to obtain the second environmental data of the fire area and the location information of each person in distress. The mobile terminal subsystem obtains the location information of each rescuer inside the chemical enterprise. Then, the host computer subsystem merges the first environmental data inside the chemical enterprise and the second environmental data of the fire area to construct the latest internal map of the chemical enterprise. Combined with the location information of each person in distress and the location information of each rescuer, path planning is performed to determine the optimal rescue path information for each rescuer. Subsequently, the mobile terminal subsystem provides real-time rescue path navigation to each rescuer. This can improve the response speed of emergency rescue in chemical enterprise fires and reduce rescue risks. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the composition of a heterogeneous multi-agent rescue system for chemical plant fires provided in an embodiment of this application; Figure 2This is a schematic diagram of the internal components of the unmanned vehicle provided in an embodiment of this application; Figure 3 This is a schematic diagram of the internal components of the drone provided in an embodiment of this application; Figure 4 This is a schematic diagram of the internal components of the smart safety helmet provided in the embodiments of this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of systems and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0017] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0018] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] Chemical plants are characterized by their densely packed production facilities and the flammable and explosive nature of their materials. Fires in these plants are characterized by their suddenness, rapid spread, and immense destructive power, easily causing significant casualties, property damage, and environmental pollution. Current fire rescue operations in chemical plants heavily rely on manual labor, lacking real-time awareness of the plant environment and the location of those in distress, resulting in low rescue efficiency and high risks. Specifically, manual fire rescue solutions for chemical plants suffer from the following major problems: First, a lack of intelligent and collaborative rescue equipment systems; unmanned platforms have limited functionality, failing to fully leverage the collaborative advantages of heterogeneous multi-agent systems, resulting in limited coverage and blind spots; Second, an inability to accurately assess the development of the fire and the movement trends of those in distress, leading to low evacuation efficiency; Third, existing positioning technologies in rescue systems largely rely on fixed base stations or simple navigation devices, which are susceptible to multipath effects and non-line-of-sight interference in complex and unknown fire environments, resulting in poor positioning accuracy and stability; Fourth, the inability to update dynamic information within the fire scene in a timely manner, leaving firefighters without effective rescue navigation support, leading to low search and rescue efficiency and high safety risks.

[0021] In view of this, this application provides a heterogeneous multi-agent rescue system for chemical plant fires. This solution proposes a heterogeneous collaborative mode between unmanned vehicles (UAVs) and drones. First, the UAV subsystem senses the first environmental data inside the chemical plant where the fire is occurring in real time. Then, the drone subsystem conducts coverage detection of fire-affected areas within the chemical plant, including areas with obstacles or where UAVs cannot pass, to obtain second environmental data and the location information of each person in distress. Finally, the mobile terminal subsystem obtains the location information of each rescuer within the chemical plant. Next, the host computer subsystem merges the first environmental data and the second environmental data of the fire-affected area to construct a new internal map of the chemical plant. Combining the location information of each person in distress and each rescuer, path planning is performed to determine the optimal rescue path for each rescuer. Subsequently, the mobile terminal subsystem provides real-time rescue path navigation to each rescuer. This system enables comprehensive perception and assessment of the fire situation inside the chemical plant, improving the intelligence level and response speed of emergency rescue in chemical plant fires, reducing the safety risks for rescuers entering the chemical plant, and increasing the success rate of rescuing those in distress.

[0022] Please see Figure 1 , Figure 1 This is an optional schematic diagram of a heterogeneous multi-agent rescue system for chemical plant fires provided in this application embodiment. The system includes a host computer subsystem and connected to it an unmanned vehicle subsystem, an unmanned aerial vehicle subsystem, and a mobile terminal subsystem.

[0023] In practical applications, the unmanned vehicle subsystem perceives the primary environmental data inside the chemical plant where the fire is occurring in real time and sends it to the host computer subsystem. The unmanned vehicle can also act as a base station, providing positioning support for drones and mobile terminals in complex indoor environments. The drone subsystem covers and probes the fire area within the chemical plant where there are obstacles or where the unmanned vehicle cannot pass, acquiring detection data which is then sent to the host computer subsystem. This detection data includes secondary environmental data of the fire area and the location information of each person in distress. The mobile terminal subsystem acquires the location information of each rescuer within the chemical plant and sends it to the host computer subsystem. The host computer subsystem merges the primary environmental data and the secondary environmental data of the fire area to construct a new internal map of the chemical plant. Based on this map, the location information of each person in distress, and the location information of each rescuer, path planning is performed to determine the optimal rescue path for each rescuer, which is then sent to the mobile terminal subsystem. The mobile terminal subsystem provides real-time rescue path navigation to each rescuer based on this optimal rescue path information.

[0024] In some embodiments, the unmanned vehicle subsystem includes multiple unmanned vehicles with identical internal configurations and application functions. The first environmental data within the chemical plant includes multiple first environmental sub-data points detected by the corresponding unmanned vehicles. For each unmanned vehicle: the unmanned vehicle is equipped with a first UWB positioning module, a first environmental perception module, and a first communication module. The unmanned vehicle also has a built-in first control module, which is connected to the first UWB positioning module, the first environmental perception module, and the first communication module, respectively. See [link to documentation]. Figure 2 As shown, the first communication module preferably adopts a WIFI communication module. In specific implementation, the first UWB positioning module provides a positioning reference for the UAV subsystem and the mobile terminal subsystem, and can also determine the location information of the unmanned vehicle and send it to the first control module; the first environmental perception module monitors the local environment of the area where the unmanned vehicle is located in real time when the unmanned vehicle is inside the chemical enterprise, obtains the first environmental sub-data, and sends it to the first control module; data transmission with the host computer subsystem is performed through the first communication module, which can be done by the first control module sending the location information of the unmanned vehicle and the first environmental sub-data detected by the unmanned vehicle to the first communication module, and then the first communication module sending the location information of the unmanned vehicle and the first environmental sub-data detected by the unmanned vehicle to the host computer subsystem; the first control module receives the optimal deployment path issued by the host computer subsystem through the first communication module to control the unmanned vehicle to avoid obstacles and move and park within the chemical enterprise.

[0025] In some embodiments, the first environmental perception module includes a first lidar device, a first gas sensor, a first temperature sensor, and a first smoke sensor, wherein: the first lidar device is used to acquire point cloud data of the local environment of the area where the unmanned vehicle is located, which includes information such as the location, number, shape, and outline of obstacles detected by the unmanned vehicle; the first gas sensor is used to monitor the gas concentration in the area where the unmanned vehicle is located, which can be at least one of CO concentration, CH4 concentration, and VOCs concentration; the first temperature sensor is used to monitor the temperature in the area where the unmanned vehicle is located; and the first smoke sensor is used to monitor the smoke concentration in the area where the unmanned vehicle is located. It is understood that the first environmental sub-data detected by the unmanned vehicle includes point cloud data of the local environment of the area where the unmanned vehicle is located, as well as the gas concentration, temperature, and smoke concentration in the area where the unmanned vehicle is located.

[0026] In some embodiments, the unmanned vehicle subsystem adjusts the unmanned vehicle movement path and mobile base station deployment when the environment inside the chemical enterprise changes.

[0027] The method for adjusting the movement path of the unmanned vehicle is explained as follows: The update of the unmanned vehicle's path does not rely on a globally preset path planning algorithm, but adopts a strategy that combines real-time reactive navigation based on local perception with group collaboration. This can be understood as introducing an artificial potential field method, in which the unmanned vehicle calculates a desired direction and speed of movement based on the synthetic potential field it perceives locally. This synthetic potential field is the vector sum of a repulsive potential field and an attractive potential field. The repulsive potential field is generated when there are obstacles in the area where the unmanned vehicle is located and / or when the area where the unmanned vehicle is located is a high-risk environment. This high-risk environment meets at least one of the following three conditions: excessive temperature, excessive concentration of toxic gases, and excessive smoke concentration. The attractive potential field is generated by the task area, which includes areas where communication coverage needs to be enhanced and areas where the unmanned vehicle needs to be recharged.

[0028] The deployment adjustment method for mobile base stations is explained as follows: First, the host computer subsystem, based on the internal map of the chemical enterprise, performs multi-objective optimization calculations with the primary optimization objectives of maximizing effective positioning coverage and maintaining global network connectivity, while taking into account the hard constraint of avoiding all environmental threats. This yields the optimal deployment locations for multiple unmanned vehicles (UAVs) and distributes the results to them accordingly. Then, the multiple UAVs perform autonomous motion control and collaborative obstacle avoidance to ultimately reach or approach their respective optimal deployment locations. Maximizing effective positioning coverage can be understood as minimizing positioning blind spots or areas with insufficient base station deployment caused by obstacles or excessive distance when multiple UAVs act as mobile base stations. This ensures that any target to be located within the chemical enterprise is within the line-of-sight range formed by the multiple UAVs and meets the requirements of trilateration. This target could be rescue personnel or drones. Maintaining global network connectivity refers to maintaining the connectivity of the communication network composed of multiple UAVs. Avoiding all environmental threats refers to controlling the UAVs to avoid high-risk environmental areas.

[0029] In some embodiments, the unmanned aerial vehicle (UAV) subsystem includes multiple UAVs, which have identical internal configurations and application functions. The second environmental data of the fire area includes multiple second environmental sub-data detected by the corresponding UAVs. For each UAV: ​​the UAV is equipped with a second UWB positioning module, a second environmental perception module, an image acquisition module, and a second communication module. The UAV has a built-in onboard processing module, which is connected to the second UWB positioning module, the second environmental perception module, the image acquisition module, and the second communication module, respectively. See [link to documentation]. Figure 3As shown, the second communication module preferably adopts a WIFI communication module. In specific implementation, the second UWB positioning module determines the drone's location information and sends it to the onboard processing module; the second environmental perception module monitors the local environment of the area where the drone is located when it is in the fire zone to obtain second environmental sub-data and sends it to the onboard processing module; the image acquisition module collects image data of the area where the drone is located and sends it to the onboard processing module; the second communication module transmits data with the host computer subsystem, which can be done by the onboard processing module sending the drone's location information and the second environmental sub-data detected by the drone to the second communication module, and then the second communication module sending the drone's location information and the second environmental sub-data detected by the drone to the host computer subsystem; the onboard processing module receives the search path issued by the host computer subsystem to control the drone to detect in the fire zone and identify the image data of the area where the drone is located. When a person in distress is identified in the image data of the area where the drone is located, the location information of the person in distress is determined based on the image data of the area where the drone is located and the drone's location information, and then the location information of the person in distress is sent to the host computer subsystem via the second communication module.

[0030] The identification of the image data of the area where the drone is located can be achieved by loading a pre-trained personnel identification model. This personnel identification model is trained on the YOLO series model based on a pre-collected training dataset. The training dataset contains several fire scene images, and each fire scene image carries marker information indicating whether there are personnel in the fire scene. The YOLO series model can be either the YOLOv3 model or the YOLOv5 model.

[0031] The image data of the area where the drone is located can reflect the spread of the fire in that area. The image data of the area where the drone is located can be sent to the host computer subsystem via the second communication module through the onboard processing module, providing information support for the command and decision-making of the command center and facilitating effective guidance for on-site rescue personnel to adjust rescue strategies.

[0032] In some embodiments, the second environmental perception module includes a second lidar device, a second gas sensor, a second temperature sensor, and a second smoke sensor. The second lidar device acquires point cloud data of the local environment in the area where the drone is located, including information such as the location, quantity, shape, and outline of obstacles detected by the drone. The second gas sensor monitors the gas concentration in the area where the drone is located, which can be at least one of CO, CH4, and VOCs concentrations. The second temperature sensor monitors the temperature in the area where the drone is located. The second smoke sensor monitors the smoke concentration in the area where the drone is located. It is understood that the second environmental sub-data detected by the drone includes point cloud data of the local environment in the area where the drone is located, as well as the gas concentration, temperature, and smoke concentration in the area where the drone is located.

[0033] In some embodiments, the onboard processing module can also identify and judge the second environmental sub-data detected by the UAV, and then autonomously adjust the search path when it is determined that the temperature and / or toxic gas concentration in the area where the UAV is located exceeds the standard, and report to the host computer subsystem via the second communication module.

[0034] In some embodiments, there is a one-to-one correspondence between multiple unmanned vehicles and multiple drones. In an environment without obstacles and where the unmanned vehicles can pass, the drones carried by the unmanned vehicles move on the ground to conduct environmental detection, that is, the unmanned vehicles actually perform the detection task. In an environment with obstacles and / or where the unmanned vehicles cannot pass, the drones fly away from the unmanned vehicles and conduct detection in the air, that is, the drones actually perform the detection task. Specifically, the unmanned vehicles are equipped with dedicated drone nests for storing drones. The dedicated drone nests are opened remotely by the host computer subsystem so that the drones can take off.

[0035] In some embodiments, the chemical enterprise has multiple key areas, including production unit areas, storage tank areas and pipeline corridor areas. Multiple unmanned vehicles are dispersed in these key areas, with at least three unmanned vehicles and three drones carried by them in each key area for environmental detection.

[0036] In some embodiments, for each unmanned vehicle (UAV): the UAV is further equipped with a charging module, which is generally located inside a dedicated nest of the UAV. The charging module is connected to a first control module of the UAV and provides charging functionality for the UAV subsystem. Specifically, when any UAV is conducting reconnaissance, if the onboard processing module of the UAV detects that the current battery level of the UAV is lower than a preset battery threshold, a charging signal is generated and sent to the host computer subsystem via the second communication module of the UAV. The host computer subsystem then selects the target UAV closest to the UAV from among the multiple UAVs based on the location information of multiple UAVs and the location information of the UAV, generates a charging control signal, and sends it to the first control module of the target UAV via the first communication module of the target UAV. This causes the target UAV to open its dedicated nest to await the UAV's landing and charging. The host computer subsystem also generates target path information for the UAV to fly and land on the target UAV based on the location information of the target UAV and the location information of the UAV, and sends it to the onboard processing module of the UAV via the second communication module of the UAV.

[0037] In some embodiments, the mobile terminal subsystem includes a smart safety helmet with AR display function, configured for each rescuer. This smart safety helmet protects the rescuer's safety. The smart safety helmet includes a third UWB positioning module, a third communication module, and a display module integrated into the protective face shield. The third communication module is connected to both the third UWB positioning module and the display module. (See also...) Figure 4 As shown in the diagram. In specific implementation, the location information of the rescuer is determined by the third UWB positioning module and sent to the third communication module. The third communication module then transmits data with the host computer subsystem. This data transmission can occur either by the third communication module sending the rescuer's location information to the host computer subsystem or by the third module receiving the optimal rescue route information corresponding to the rescuer from the host computer subsystem. The optimal rescue route information corresponding to the rescuer is then projected and displayed by the display module, thereby providing the rescuer with autonomous rescue navigation services.

[0038] In some embodiments, the physiological status information of each rescuer is acquired through the mobile terminal subsystem and sent to the host computer subsystem. The physiological status information of the rescuer includes the rescuer's heart rate and blood oxygen saturation. Specifically, the smart safety helmet also integrates a PPG (photoplethysmography) sensor. The PPG sensor can use the rescuer's earlobe as the main measurement location to monitor the rescuer's heart rate and blood oxygen saturation, so that the command center can determine in real time whether the rescuer's blood oxygen has decreased due to smoke or hypoxia, and prevent the rescuer from suffocation or hypoxia injury.

[0039] In some embodiments, the method of fusing first environmental data of the chemical plant's interior and second environmental data of the fire area to construct a new internal map of the chemical plant may include, but is not limited to, using SLAM (Simultaneous Localization and Intervention)... Mapping (Simultaneous Localization and Mapping) technology matches and merges point cloud data collected by multiple unmanned vehicles and drones at different locations to gradually construct a 3D map framework for the chemical plant. Based on gas concentration, temperature, and smoke concentration data collected by multiple unmanned vehicles and drones at different locations, data association and environmental attribute labeling are performed on this 3D map framework. For example, if an unmanned vehicle collects a CO concentration value at a certain location, that CO concentration value is marked at that location on the 3D map framework as the corresponding environmental attribute data. All unobstructed spaces within the labeled 3D map framework are identified as passable areas. Then, based on the labeled environmental attribute data, all passable areas are classified for safety. Passable areas without environmental threat labels or with low environmental threat levels are marked as safe passageways or rescue passageways, while passable areas with high environmental threat levels are marked as danger zones or restricted areas, thus forming the final internal map of the chemical plant.

[0040] The section on data association and environmental attribute labeling on the 3D map framework based on gas concentration, temperature, and smoke concentration data collected by multiple unmanned vehicles and drones at different locations can be understood as follows: The gas concentration, temperature, and smoke concentration data collected by multiple unmanned vehicles and drones at different locations are collectively referred to as the current environmental measurement values. First, historical environmental measurement values ​​collected by multiple unmanned vehicles and drones at different locations are obtained. Then, a Kalman filter algorithm is used to fuse the historical and current environmental measurement values ​​at the same location to obtain the estimated current environmental value for that location. This filters out instantaneous sensor noise and predicts short-term trends in environmental parameters. Subsequently, data association and environmental attribute labeling are performed on the 3D map framework based on the current environmental estimate. It should be noted that if only the current environmental measurement value is available for a certain location, then that current environmental measurement value is directly used as the estimated current environmental value for that location.

[0041] In some embodiments, when the host computer subsystem performs path planning based on the internal map of the chemical plant, the location information of each person in distress, and the location information of each rescuer, path safety and efficiency optimization can be taken as common objectives and implemented using a preset path planning algorithm, preferably Dijkstra's algorithm. Achieving path safety can be understood as ensuring the planned path is as far away as possible from all hazardous or prohibited areas marked on the internal map of the chemical plant. Achieving efficiency optimization can be understood as determining the shortest path from the rescuers to the people in distress based on the location information of both rescuers and people in distress. It should be noted that when defining the objective function used in the algorithm, the weight of the path safety term is generally set greater than the weight of the efficiency optimization term.

[0042] In some embodiments, the host computer subsystem marks and displays the locations of each person in distress and each rescuer on a newly constructed internal map of the chemical enterprise, allowing the command center to intuitively grasp the current rescue progress. Furthermore, the host computer subsystem can also mark and display the locations of multiple unmanned vehicles and multiple drones on a newly constructed internal map of the chemical enterprise, allowing the command center to intuitively grasp the coordinated rescue efforts of heterogeneous multi-agent systems.

[0043] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0044] The terms "first," "second," "third," "fourth," etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0045] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0046] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A heterogeneous multi-agent rescue system for chemical plant fires, characterized in that, include: The unmanned vehicle subsystem is used to perceive the first environmental data inside the chemical enterprise where a fire has occurred in real time and send it to the host computer subsystem. It also serves as a base station to provide positioning support for drones and mobile terminals in complex indoor environments. The unmanned aerial vehicle (UAV) subsystem is used to conduct coverage detection of fire areas within the chemical enterprise where there are obstacles or where unmanned vehicles cannot pass, in order to obtain detection data and send it to the host computer subsystem; wherein, the detection data includes second environmental data of the fire area and location information of each person in distress; The mobile terminal subsystem is used to acquire the location information of each rescuer inside the chemical enterprise and send it to the host computer subsystem. The host computer subsystem is used to fuse the first environmental data and the second environmental data to construct the latest internal map of the chemical enterprise, and then perform path planning based on the internal map of the chemical enterprise, the location information of each person in distress and the location information of each rescuer to determine the optimal rescue path information corresponding to each rescuer and send it to the mobile terminal subsystem. The mobile terminal subsystem is also used to provide real-time rescue route navigation to each rescuer based on the optimal rescue route information corresponding to each rescuer.

2. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 1, characterized in that, The unmanned vehicle subsystem is also used to adjust the unmanned vehicle movement path and mobile base station deployment when the environment inside the chemical enterprise changes.

3. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 1, characterized in that, The unmanned vehicle subsystem includes multiple unmanned vehicles; Each of the unmanned vehicles is equipped with a first UWB positioning module, a first environmental perception module, and a first communication module. The first UWB positioning module is used to provide a positioning reference for the unmanned vehicle subsystem and the mobile terminal subsystem. The first environmental perception module is used to monitor the local environment of the area where the unmanned vehicle is located in real time when the unmanned vehicle is inside the chemical enterprise. The first communication module is used to transmit data with the host computer subsystem. Each of the unmanned vehicles is equipped with a first control module, which is used to receive the optimal deployment path issued by the host computer subsystem in order to control the unmanned vehicle to avoid obstacles and move and park within the chemical enterprise.

4. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 3, characterized in that, The unmanned vehicle subsystem includes multiple unmanned vehicles, and there is a one-to-one correspondence between the multiple unmanned vehicles and the multiple unmanned vehicles. Each of the UAVs is equipped with a second UWB positioning module, a second environmental perception module, an image acquisition module, and a second communication module. The second UWB positioning module is used to determine the location information of the UAV. The second environmental perception module is used to monitor the local environment of the area where the UAV is located when the UAV is located in the fire area. The image acquisition module is used to acquire image data of the area where the UAV is located. The second communication module is used to transmit data with the host computer subsystem. Each of the aforementioned UAVs has a built-in onboard processing module. The onboard processing module is used to receive the search path issued by the host computer subsystem to control the UAV to conduct detection in the fire area. It is also used to identify the image data and, when it is identified that there are people in distress in the image data, to determine the location information of the people in distress based on the image data and the location information of the UAV.

5. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 3, characterized in that, The mobile terminal subsystem includes smart safety helmets configured for each rescuer; The smart safety helmet is equipped with a third UWB positioning module and a third communication module. The third UWB positioning module is used to determine the location information of the rescuer, and the third communication module is used to transmit data with the host computer subsystem. The smart safety helmet also has a display module integrated into the protective face shield, which is used to project and display the optimal rescue route information corresponding to the rescuer.

6. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 3, characterized in that, Each of the unmanned vehicles is also equipped with a charging module, which is used to provide charging functionality for the unmanned vehicle subsystem.

7. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 3, characterized in that, The first environmental perception module includes a first lidar device, a first gas sensor, a first temperature sensor, and a first smoke sensor. The first lidar device is used to acquire point cloud data of the local environment in the area where the unmanned vehicle is located. The first gas sensor is used to monitor the gas concentration in the area where the unmanned vehicle is located. The first temperature sensor is used to monitor the temperature in the area where the unmanned vehicle is located. The first smoke sensor is used to monitor the smoke concentration in the area where the unmanned vehicle is located.

8. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 4, characterized in that, The second environmental perception module includes a second lidar device, a second gas sensor, a second temperature sensor, and a second smoke sensor. The second lidar device is used to acquire point cloud data of the local environment in the area where the drone is located. The second gas sensor is used to monitor the gas concentration in the area where the drone is located. The second temperature sensor is used to monitor the temperature in the area where the drone is located. The second smoke sensor is used to monitor the smoke concentration in the area where the drone is located.

9. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 1, characterized in that, The mobile terminal subsystem is also used to acquire the physiological status information of each rescuer and send it to the host computer subsystem.

10. The heterogeneous multi-agent rescue system for chemical plant fires according to claim 1, characterized in that, The host computer subsystem is also used to mark and display the locations of the distressed personnel and the rescue personnel on the internal map of the chemical enterprise.

Images