Emergency control system

Abstract

The invention relates to an emergency control system comprising the following phases: a detection and sensing phase in which data is collected in real time by means of a network of sensors and operatives in the field, a processing and predictive simulation phase in which future scenarios are predicted on the basis of the collected data and historical data, a visualisation phase for visualising the previous phase in a 3D model, a progressive scenario generation phase, and a phase of interaction with operatives in the field in which an automatic response to the emergency is managed in real time and on the basis of the data obtained in the previous phases.

Description

[0001] Emergency control system.

[0002] DESCRIPTION

[0003] Emergency control system characterized by comprising the following phases: a detection and sensing phase in which data is collected in real time through a network of sensors and field personnel, a processing and predictive simulation phase in which future scenarios are predicted based on the collected and historical data, a visualization phase of the previous phase in a three-dimensional model, a phase of generating progressive scenarios, and a phase of interaction with field personnel in which an automatic response to the emergency is managed in real time and based on the data obtained in the previous phases.

[0004] BACKGROUND OF THE INVENTION

[0005] For some time now, the European Union has been facing a series of challenges within the framework of unitary cooperation.

[0006] Despite these strategic advances, the EU faces a structural challenge: the lack of real-time coordination and access to accurate data visibility, both essential for managing complex situations such as civil emergencies, humanitarian crises, cross-border fires, natural disasters and critical infrastructure operations.

[0007] This is even more critical and complex with the collaboration of entities from different agents and countries, which complicates decision-making in a unified manner, since different points of view appear when making decisions.

[0008] These gaps considerably hinder decision-making by response teams, delaying their ability to optimally assess and respond to a given problem.

[0009] BRIEF DESCRIPTION OF THE INVENTION The present invention falls within the technical field of inventions that manage the coordination of emergency situations that require the coordinated intervention of multiple institutions.

[0010] The solution of the present invention constitutes an Expert Decision Network for Extended Emergency and Security Response in the civil sphere, and improves the ability to perceive and interpret elements of the environment, identify potential hazards and accurately anticipate future events.

[0011] Response time in an event is reduced to almost instantaneous moments, compared to the times that are being adopted today.

[0012] In other words, the system significantly reduces the response time from the moment the detection occurs until the operational action is carried out, allowing very short reaction times, compared to conventional reaction times.

[0013] An object of the present invention is an emergency control system characterized in that it comprises the following phases: a detection and sensing phase in which data is collected in real time by means of a network of sensors and field personnel, a processing and predictive simulation phase in which future scenarios are predicted based on the collected data and historical data, a visualization phase of the previous phase in a three-dimensional model, a phase of generation of progressive scenarios, and a phase of interaction with field personnel in which an automatic response to the emergency is managed in real time and based on the data obtained in the previous phases.

[0014] CONCRETE REALIZATION OF THE INVENTION

[0015] Thus, the emergency control system that is the subject of this invention comprises a series of phases, which are described below. Initially, it comprises a detection and sensing phase in which, through a network of sensors and also through field operations, data is collected in real time.

[0016] The sensors may include, but are not limited to, temperature sensors, humidity sensors, wind speed sensors (anemometers), smoke detectors, thermal cameras, seismic sensors, and any other environmental sensor that could be considered relevant for emergency detection.

[0017] In a subsequent phase, known as the predictive processing and simulation phase, future scenarios are predicted based on the collected data and historical data.

[0018] Thus, in this phase, machine learning algorithms are developed and implemented to analyze historical data and predict the behavior of certain catastrophes.

[0019] Next, within this same phase, predictive models based on future learning (also known as Deep Learning) are built to estimate the magnitude and spread of emergencies in real time, using the risk levels provided by the competent authority (for example, in Spain the State Meteorological Agency, known as AEMET) together with the risk data of a given area obtained through GIS (Geographic Information Systems) maps where the risks of the terrain are represented.

[0020] Next comes the visualization phase, in which the information processed in the previous phase is represented in a three-dimensional model. The goal of this phase is to provide operators with an immersive understanding of the situation they are facing and to visually grasp the emergency they will be confronting.

[0021] Subsequently, the progressive scenario generation phase involves implementing optimization algorithms that dynamically allocate evacuation routes and resources based on various variables. Simulations based on historical scenarios are designed to validate the effectiveness of the emergency plans.

[0022] Finally, in the phase of interaction with field operations, an automatic response to the emergency is managed in real time and based on the data obtained in the previous phases.

[0023] A real-time monitoring system is being developed to evaluate the performance of the equipment during drills.

[0024] Subsequently, automatic feedback modules are created to adjust the difficulty of the scenarios based on staff performance.

[0025] Next, algorithms are optimized to generate dynamic simulations based on user feedback.

[0026] Finally, the effectiveness of training in simulated environments with different levels of complexity is validated.

[0027] In the event of automation of the execution of emergency and rapid response plans, algorithms would be developed for the automatic activation of protocols after the detection of emergencies.

[0028] Next, automated systems for distributing customized orders to teams based on their location and function would be implemented.

[0029] Thus, an algorithm would be developed that suggests when a communication should be made to the population and selects the population group to launch the communication using the Mcx communications standard through mobile or desktop applications

[0030] Real-time monitoring models would also be developed or implemented to adjust orders based on live data (for example, if an evacuation route becomes inaccessible or if a high-risk area is identified).

[0031] Finally, algorithms for the autonomous management of unmanned aerial vehicles (UAVs) in reconnaissance tasks and in terrain reconnaissance and modeling tasks using Gaussian Splatting techniques (it is a rendering technique that represents 3D scenes using Gaussian primitives) would be investigated and developed.

[0032] Optionally, the system is activated autonomously, based on certain predefined parameters, some action protocols.

[0033] It could also be configured, as an option, so that the system, in the phase of interaction with field operators, includes an event-based control engine, which monitors in real time the data from the previous phases.

[0034] Another implementation option could be to configure a distributed modular architecture, composed of several interrelated subsystems that allow the aforementioned real-time data processing, prediction generation, and coordinated execution of actions.

[0035] If the modular architecture described in the previous paragraph were chosen, it would be possible to configure the aforementioned modules to be interconnected via a low-latency internal messaging bus. This modular architecture ensures system scalability and allows for the integration of new data sources or functionalities without altering the core components.

[0036] For a hypothetical emergency plan, the state of the art of generative modeling technologies and optimization algorithms applied to emergency management is analyzed.

[0037] For example, broadly speaking, in the event of a fire, the system would operate as follows: In the detection phase, a fire would be detected by sensors and drones set up on the ground, which could detect smoke or a sudden increase in temperature at a specific point, for example, above 60°C.

[0038] In this phase, for example, a data acquisition module would be available, which would be configured to acquire data. This module would be configured to receive information from environmental sensors, Internet of Things (IoT) devices, fixed or unmanned aerial vehicle (UAV) cameras, local weather stations, geographic information systems (GIS), communication networks, and external third-party platforms.

[0039] Next, a fusion and normalization module unifies the heterogeneous data and generates a coherent operational state of the environment through filtering, interpolation, and automatic validation techniques of the received information.

[0040] The system incorporates the ability to receive real-time video and telemetry streams from unmanned aerial vehicles (UAVs) deployed over the emergency zone. From these sources, the system generates updated three-dimensional terrain models using reconstruction techniques based on Gaussian splatting or other equivalent methods.

[0041] The resulting models are directly integrated into the system's predictive engine (detailed below), allowing for the recalculation of emergency propagation, the updating of safe routes, and the redefinition of risk areas with greater spatial and temporal accuracy. This integration reduces latency between capturing the environment and updating the system's operational status.

[0042] In the predictive simulation phase, based on data provided by sensors and drones, evacuation routes would be generated based on weather forecasts, wind speed and direction, the existence of firebreaks, etc. In other words, it is calculated that, given the existing wind conditions (with a certain speed and direction), the existing fuel load of vegetation, the dryness of the terrain, and the ambient humidity, the fire could spread, burning a certain area in a certain period of time.

[0043] In this phase, a predictive simulation engine can be used. This engine combines physical models, historical data, and machine learning algorithms to estimate the future evolution of the emergency and calculate possible trajectories or alternative scenarios.

[0044] In the visualization phase, it will be possible to observe in real time the fire fronts, the deployed ground units and the available air resources.

[0045] For the current phase, a 3D and XR visualization interface could be used, which would be configured to represent in an immersive or conventional way the current state of the emergency, its predictions and alternative scenarios, facilitating the spatial interpretation of the environment.

[0046] The system could include an immersive visualization interface based on extended reality (XR), which represents on a georeferenced three-dimensional model the current state of the emergency, the position of the deployed equipment and the predictions generated by the simulation engine.

[0047] This three-dimensional representation improves the spatial interpretation of the operational scenario and reduces the possibility of human error in decision-making, especially in situations of low visibility or high geographical complexity.

[0048] This representation would allow us to observe the current position of the fire and its fronts, the burned vegetation areas and those that present a greater risk (due to the spread, the difficulty of the terrain or their environmental value), the existing firebreaks, the populations at risk, etc.

[0049] In the next phase, three progressive scenarios are simulated: one in which the fire is under control, another with medium intensity, and a final scenario in which the situation is very serious. This prepares the scenarios so that, should one of them occur, an immediate response can be implemented. For this purpose, a module can also be used—in this case, a progressive scenario generation module—which develops a series of operational hypotheses and proposes corresponding courses of action based on different combinations of variables, available resources, and environmental conditions.

[0050] For example, it could assign a certain number of brigades to a fire front, calculate the optimal routes avoiding the areas of greatest fire risk for emergency teams with route updates based on the collected data, etc.

[0051] Finally, in the phase of interaction with field operations, different operational coordination centers are coordinated, for example, if it were in the Canary Islands, the Emergency and Security Operational Coordination Center (CECOES), the Island Operational Coordination Center (CECOPIN) and the local brigades, through a collaborative virtual environment.

[0052] This would employ a control and execution module, enabling real-time management of the operational response. This is achieved through the issuance of commands, dynamic resource allocation, route reconfiguration, and activation of predefined protocols.

[0053] If a system were used, it would incorporate an event-driven logic module that monitors real-time data from sensors, UAVs, weather sources, and GIS layers.

[0054] When certain combinations of parameters are detected, such as sudden temperature increases, adverse changes in wind direction, loss of route accessibility, the appearance of new hotspots, or exceedance of user-defined thresholds, the system automatically activates pre-configured action protocols.

[0055] These protocols may include issuing operational orders to personnel in the field, reassigning resources, updating evacuation routes, sending messages to the population through MCX systems, or activating internal security procedures.

[0056] Event-based automated execution reduces response time to critical changes in the situation and ensures operational consistency even in scenarios with high workload or limited availability of human operators.

[0057] The present invention describes a novel emergency control system. The examples mentioned herein are not limiting to the present invention, and therefore it may have different applications and / or adaptations.

Claims

CLAIMS 1.- Emergency control system characterized by comprising the following phases: - detection and sensing phase in which real-time data is collected through a network of sensors and field operations, - Processing and predictive simulation phase in which future scenarios are predicted based on the collected and historical data, - Visualization phase of the previous phase in a three-dimensional model, - phase of generating progressive scenarios, and - interaction phase with field operations in which an automatic response to the emergency is managed in real time and based on the data obtained in the previous phases. 2.- System, according to claim 1, characterized in that it comprises a fusion and normalization phase following the detection and sensing phase, unifying data and generating an operational state of the environment. 3.- System, according to claim 1 or 2, characterized in that the system is configured to autonomously activate, based on the detection of conditions that exceed certain predefined parameters, automatic action protocols. 4.- System, according to any of the preceding claims, characterized in that it comprises in the phase of interaction with field operators an event-based control engine, configured to monitor in real time the data from the previous phases and activate automatic responses based on the events detected. 5.- System according to claims 3 and 4, characterized in that it comprises a distributed modular architecture, composed of several interconnected subsystems that allow the aforementioned real-time data processing, generation of predictions and execution of actions in a coordinated manner. 6.- System, according to claim 5, characterized in that the aforementioned modules are interconnected by means of a low latency internal messaging bus.