Dynamic Risk Avoidance Method for Firefighting Robots Based on Adaptive Threshold of Thermal Field Risk Confidence

Abstract

This invention relates to the field of fire-fighting robot avoidance control technology, and discloses a dynamic avoidance method for fire-fighting robots with an adaptive threshold for thermal field risk confidence. The method includes acquiring temperature data, thermal radiation data, smoke concentration data, airflow velocity data, and wall thermal reflection data from a fire environment, constructing a heat flow direction field, generating heat flow convergence state and thermal field closure trend parameters based on the heat flow direction vector set, and simultaneously generating thermal inertia retention parameters by combining temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period. This generates a thermal field risk confidence level, and a dynamic risk threshold is generated based on the environment's escape capability to control the fire-fighting robot to perform dynamic avoidance actions. This invention, by constructing a heat flow direction field, enables the fire-fighting robot to determine risk based on changes in the heat flow structure, thereby reducing the risk identification lag under a fixed temperature threshold approach.

Description

Technical Field

[0001] This invention relates to the field of fire-fighting robot avoidance and control technology, and in particular to a dynamic avoidance method for fire-fighting robots with an adaptive threshold for thermal field risk confidence. Background Technology

[0002] Firefighting robots are used for fire detection, area inspection, and operation in hazardous areas. In environments with high temperatures, dense smoke, and complex obstacles, the movement path needs to be dynamically adjusted according to the heat distribution of the fire scene to reduce the chance of the robot entering high-risk areas.

[0003] In existing technologies, firefighting robots typically acquire temperature data in the fire area using temperature sensors, infrared thermal imaging equipment, or smoke detection devices, and then identify the risk of the current area based on preset temperature thresholds. When the temperature value in the detected area exceeds the preset temperature threshold, the system controls the firefighting robot to stop moving or adjust its movement path. Some technical solutions also combine the regional heat radiation intensity or smoke concentration to classify the fire area into different hazard levels and generate corresponding evacuation paths based on different hazard levels. In some fire path planning schemes, high-temperature areas in the fire area are marked by constructing a regional heat distribution map, and the robot's movement path is generated based on the heat distribution results.

[0004] However, the inventors of this application discovered in the process of realizing the technical solution of this application that the prior art usually assesses the risk of a fire area based on the regional temperature value or a fixed temperature threshold. However, the heat diffusion process in a fire involves heat flow accumulation, heat flow recirculation, and local heat retention. When the temperature in some areas has not yet exceeded the preset temperature threshold, the heat flow structure in the local area has already formed a continuous accumulation state, which leads to the problem of delayed risk identification in the prior art. Summary of the Invention

[0005] To overcome the above shortcomings, this invention provides a dynamic risk avoidance method for fire-fighting robots with an adaptive threshold for thermal risk confidence. This method aims to improve the problem of risk identification lag in existing technologies when the temperature in some areas has not yet exceeded the preset temperature threshold.

[0006] In a first aspect, the present invention provides the following technical solution: a dynamic risk avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence, comprising the following steps:

[0007] S1. Obtain thermal field state data of the fire scene environment where the fire robot is located. The thermal field state data includes temperature data, thermal radiation data, smoke concentration data, airflow velocity data, and wall thermal reflection data.

[0008] S2. Construct a heat flow direction field based on the thermal field state data, and generate a set of heat flow direction vectors within the corresponding region;

[0009] S3. Calculate the heat flow convergence state in the local area based on the heat flow direction vector set, and generate thermal field closure trend parameters according to the heat flow direction change relationship;

[0010] S4. Generate thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period.

[0011] S5. Generate a thermal field risk confidence level based on the heat flow convergence state, the thermal field closure trend parameters, and the thermal inertia retention parameters;

[0012] S6. Generate a dynamic risk threshold based on the thermal field risk confidence level and environmental escape capability, and determine the risk of the current area based on the dynamic risk threshold.

[0013] S7. When the risk triggering conditions are met in the current area, a safe escape direction is generated according to the heat flow direction field, and the fire-fighting robot is controlled to perform dynamic risk avoidance actions.

[0014] Preferably, in step S2, the step of generating the set of heat flow direction vectors within the corresponding region includes:

[0015] Divide the fire area into regional grids;

[0016] Obtain the temperature change status in each grid region;

[0017] Thermal gradient information is generated based on the temperature change relationship between adjacent grid regions;

[0018] The direction of heat diffusion is determined based on the thermal gradient information;

[0019] Generate a heat flow direction vector in the corresponding grid region based on the heat diffusion direction;

[0020] The heat flow direction field is constructed based on the heat flow direction vectors in multiple grid regions.

[0021] Preferably, in step S3, the step of calculating the heat flow convergence state within a local area based on the heat flow direction vector set includes:

[0022] Extract multiple heat flow direction vectors in a local region;

[0023] Calculate the directional consistency among multiple heat flow direction vectors;

[0024] The heat flow convergence state is generated based on the directional convergence relationship between multiple heat flow direction vectors;

[0025] The heat flow convergence intensity is generated based on the degree of directional convergence between multiple heat flow direction vectors.

[0026] Preferably, in step S3, the step of generating the thermal field closure trend parameters based on the heat flow direction change relationship includes:

[0027] Detecting the reversal of heat flow direction in a localized area; detecting the circulation of heat flow in a localized area; detecting the obstruction of heat flow diffusion in a localized area;

[0028] The heat flow accumulation region is determined based on the heat flow direction reversal state, heat flow circulation state, and heat flow diffusion obstruction state; the thermal field closure trend parameter is generated based on the heat flow accumulation region.

[0029] Preferably, in step S4, the step of generating thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period includes:

[0030] Obtain the regional temperature change status over a continuous time period;

[0031] Detect the change in heat flux diffusion rate over a continuous time period;

[0032] Obtain the changes in wall thermal reflection over a continuous time period;

[0033] The degree of heat retention in the region is calculated based on the continuous temperature change, the slowed heat diffusion, and the enhanced heat reflection from the wall surface.

[0034] Thermal inertia retention parameters are generated based on the degree of heat retention in the region.

[0035] Preferably, in step S5, the step of generating the thermal field risk confidence score includes:

[0036] Risk mapping processing is performed on the aforementioned heat flow convergence state;

[0037] Structural risk mapping is performed on the thermal field closure trend parameters;

[0038] The thermal inertia retention parameters are subjected to thermal accumulation risk mapping processing;

[0039] Multiple risk parameters after mapping are fused to generate a thermal field risk confidence score.

[0040] Preferably, in step S6, the step of generating the dynamic risk threshold includes:

[0041] Obtain the status of passable paths and obstacle distribution in the current area;

[0042] Environmental escape capability is generated based on the number of passable paths and the density of obstacle distribution.

[0043] Obtain the basic risk threshold;

[0044] The basic risk threshold is dynamically adjusted based on the aforementioned thermal field risk confidence level;

[0045] A dynamic risk threshold is generated based on the adjusted risk threshold and the environmental escape capability.

[0046] Preferably, the step of dynamically adjusting the basic risk threshold based on the thermal field risk confidence level includes:

[0047] Lower the baseline risk threshold when the confidence level of thermal field risk increases; raise the baseline risk threshold when heat flow diffusion is enhanced; lower the baseline risk threshold when the tendency of thermal field closure is enhanced.

[0048] Lower the baseline risk threshold when the ability to escape from the environment decreases.

[0049] Preferably, in step S7, the step of generating a safe escape direction based on the heat flow direction field and controlling the firefighting robot to perform dynamic avoidance actions includes:

[0050] Identify the thermal risk zone; determine the heat flow accumulation direction based on the heat flow direction vector in the thermal risk zone; generate a safe escape direction based on the opposite direction of the heat flow accumulation direction; control the fire-fighting robot to perform evacuation actions along the safe escape direction.

[0051] Secondly, this invention provides the following technical solution: a dynamic avoidance system for fire-fighting robots with an adaptive threshold for thermal risk confidence, comprising:

[0052] The thermal field status data acquisition module acquires thermal field status data in the fire scene environment where the firefighting robot is located.

[0053] The heat flow direction field construction module constructs a heat flow direction field based on the heat field state data and generates a set of heat flow direction vectors within the corresponding region;

[0054] The heat flow convergence analysis module calculates the heat flow convergence state within a local area based on the heat flow direction vector set;

[0055] The thermal field closure trend analysis module generates thermal field closure trend parameters based on the relationship between heat flow direction changes.

[0056] The thermal inertia retention analysis module generates thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period.

[0057] The thermal field risk confidence generation module generates thermal field risk confidence based on heat flow convergence state, thermal field closure trend parameters, and thermal inertia retention parameters.

[0058] The dynamic risk threshold generation module generates dynamic risk thresholds based on thermal field risk confidence and environmental escape capability.

[0059] The dynamic risk avoidance control module generates a safe escape direction based on the heat flow direction field when the risk triggering conditions are met in the current area, and controls the fire-fighting robot to perform dynamic risk avoidance actions.

[0060] The present invention has the following beneficial effects:

[0061] 1. This invention constructs a heat flow direction field and generates heat flow convergence state and thermal field closure trend parameters based on the heat flow direction vector set, enabling fire-fighting robots to make risk judgments based on changes in heat flow structure, thereby reducing the risk identification lag under the fixed temperature threshold method.

[0062] 2. This invention generates thermal inertia retention parameters based on temperature change, heat flow diffusion, and wall thermal reflection, and combines these parameters with heat flow convergence to generate thermal field risk confidence, enabling firefighting robots to identify areas where heat is retained, thereby reducing path misjudgment.

[0063] 3. This invention generates a dynamic risk threshold based on the confidence level of the thermal field risk and the ability to escape from the environment, and generates a safe escape direction based on the dynamic risk threshold, so that the fire-fighting robot can adjust its avoidance path according to changes in the fire environment, thereby reducing the chance of entering high-risk areas. Attached Figure Description

[0064] Figure 1 This is a flowchart of the dynamic risk avoidance method for fire-fighting robots with adaptive threshold for thermal field risk confidence proposed in this invention.

[0065] Figure 2 This is a diagram of the dynamic risk avoidance system architecture for firefighting robots with adaptive threshold for thermal field risk confidence, as proposed in this invention. Detailed Implementation

[0066] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] Example 1

[0068] Reference Figure 1 In the first embodiment of the present invention, the present invention provides a dynamic risk avoidance method for firefighting robots with an adaptive threshold for thermal field risk confidence, comprising the following steps:

[0069] S1. Acquire thermal field state data of the fire scene environment where the fire robot is located. The thermal field state data includes temperature data, thermal radiation data, smoke concentration data, airflow velocity data, and wall thermal reflection data.

[0070] Specifically, after the firefighting robot enters the fire scene, it simultaneously collects data on the thermal field conditions using thermal imaging, infrared thermography, smoke concentration, airflow detection, and wall reflection detection units mounted on the robot. The thermal imaging unit acquires the surface heat distribution in the fire area; the infrared thermography unit acquires the area temperature; the smoke concentration unit acquires the concentration of smoke particles in the air; the airflow detection unit acquires the airflow velocity and direction; and the wall reflection detection unit acquires the thermal radiation reflection from walls or obstacles in the fire scene. To ensure spatial correspondence between different types of thermal field data, the system first divides the area where the firefighting robot is currently located into a spatial grid and establishes a regional grid coordinate system based on the robot's current position. Let the regional grid set be:

[0071] ;

[0072] in: Represents the grid set of the fire area; Indicates the first Each area grid; Indicates the number of grid cells in the vertical region; This indicates the number of grid cells in the horizontal region.

[0073] After completing the regional grid division, the data acquired by each acquisition unit is mapped to the corresponding regional grid to form the thermal field state vector of the corresponding region. The thermal field state vector in each regional grid is represented as follows:

[0074] ;

[0075] in: Indicates the first Thermal field state vectors in each region grid; Indicates the temperature value of the area; Indicates the intensity of thermal radiation; Indicates smoke concentration; Indicates airflow velocity; This indicates the intensity of heat reflection from the wall surface.

[0076] To mitigate the impact of sampling time differences between different sensors on the thermal field analysis process, a unified time stamp is added to the data acquired by each acquisition unit during the data mapping process. Data from the same moment are then combined according to a preset sampling period to form regional thermal field state data for that specific moment. For regions with missing data, historical sampling data from adjacent regions are used to supplement the missing data, ensuring regional continuity in the subsequent generation of the heat flow direction field.

[0077] This step involves uniformly collecting temperature data, thermal radiation data, smoke concentration data, airflow velocity data, and wall thermal reflection data from the fire environment, and establishing a regional grid correspondence. This allows different thermal field state data to form unified regional thermal field state data, thus providing a regionally corresponding data foundation for subsequent heat flow direction field generation and thermal field risk analysis.

[0078] S2. Construct a heat flow direction field based on the thermal field state data, and generate a set of heat flow direction vectors within the corresponding region;

[0079] Preferably, in step S2, the step of generating the set of heat flow direction vectors within the corresponding region includes:

[0080] Divide the fire area into regional grids;

[0081] Obtain the temperature change status in each grid region;

[0082] Thermal gradient information is generated based on the temperature change relationship between adjacent grid regions;

[0083] Determine the direction of heat diffusion based on thermal gradient information;

[0084] Generate heat flow direction vectors in the corresponding grid regions based on the heat diffusion direction;

[0085] The heat flow direction field is constructed based on the heat flow direction vectors in multiple grid regions.

[0086] Specifically, after acquiring the regional thermal field state data, the system analyzes the temperature change status of each region based on the fire area grid set and generates thermal gradient information according to the temperature differences between adjacent region grids. Since the heat diffusion process in the fire area is directional, the system does not directly use the regional temperature value as the basis for risk analysis, but instead determines the direction of heat diffusion through the temperature change relationship between regions. For any region grid, the system calculates the temperature change status in both the lateral and longitudinal directions, where the lateral thermal gradient is expressed as:

[0087] ;

[0088] The longitudinal thermal gradient is expressed as:

[0089] ;

[0090] in: Indicates the first Lateral thermal gradients in each region grid; Indicates the first Longitudinal thermal gradient in each region grid; This indicates the temperature values ​​in adjacent horizontal regions; This represents the temperature values ​​in adjacent longitudinal regions; Indicates the horizontal grid spacing; Indicates the vertical grid spacing.

[0091] The system determines the direction of heat diffusion in the current region based on the lateral and longitudinal thermal gradients, and generates a heat flow direction vector for the corresponding region. The heat flow direction vector is represented as:

[0092] ;

[0093] in: Indicates the first Heat flow direction vectors in each region grid; Indicates the component of the lateral heat diffusion direction; This represents the component of longitudinal heat diffusion.

[0094] Due to smoke disturbance, obstruction, and localized heat reflection in the fire environment, the system further corrects the heat flow direction by combining airflow velocity data and wall heat reflection data after generating the heat flow direction vector. When the airflow direction in the area deviates from the heat diffusion direction, the system compensates for the deviation of the heat flow direction vector according to the airflow direction; when the wall heat reflection intensity in the area exceeds a preset range, the system attenuates the heat flow direction vector in the corresponding area to reduce the impact of wall heat reflection on the heat flow direction analysis process.

[0095] After generating the heat flow direction vectors in each region, the system combines the heat flow direction vectors from multiple regions to form a heat flow direction field according to the spatial connection relationship between the region grids, and constructs the regional heat diffusion distribution relationship based on the direction change state in the heat flow direction field. For regions with continuous heat flow direction changes, the system maintains the continuous connection state between heat flow directions; for regions where the heat flow direction changes abruptly, the system marks them as local thermal disturbance regions for use in subsequent heat flow convergence state analysis processes.

[0096] This step generates thermal gradient information based on the regional temperature change relationship, and combines it with airflow velocity data and wall thermal reflection data to generate a heat flow direction field. This enables the system to establish the regional heat flow distribution relationship based on the heat diffusion direction in the fire field, thus providing a basis for the subsequent heat flow convergence state analysis and thermal field closure trend analysis process.

[0097] S3. Calculate the heat flow convergence state in the local area based on the heat flow direction vector set, and generate thermal field closure trend parameters according to the heat flow direction change relationship;

[0098] Preferably, in step S3, the step of calculating the heat flow convergence state within a local region based on the heat flow direction vector set includes:

[0099] Extract multiple heat flow direction vectors in a local region;

[0100] Calculate the directional consistency among multiple heat flow direction vectors;

[0101] The heat flow convergence state is generated based on the directional convergence relationship between multiple heat flow direction vectors;

[0102] The heat flow convergence intensity is generated based on the degree of directional convergence between multiple heat flow direction vectors.

[0103] Preferably, in step S3, the step of generating the thermal field closure trend parameters based on the heat flow direction change relationship includes:

[0104] Detecting the reversal of heat flow direction in a localized area; detecting the circulation of heat flow in a localized area; detecting the obstruction of heat flow diffusion in a localized area;

[0105] The heat flow accumulation region is determined based on the heat flow direction reversal state, heat flow circulation state, and heat flow diffusion obstruction state; the thermal field closure trend parameter is generated based on the heat flow accumulation region.

[0106] Specifically, after constructing the heat flow direction field, the system extracts the heat flow direction vectors in local regions based on the spatial adjacency relationships between regional grids, and analyzes the regional heat accumulation state through the changes in heat flow direction in local regions. Since the heat diffusion process in a fire involves local convergence and local rollback phenomena, the system constructs a local analysis window centered on the current regional grid and extracts multiple heat flow direction vectors within this window. The local region window is represented as follows:

[0107] ;

[0108] in: Indicates the first Local area windows corresponding to each region; This represents the direction vector of heat flow in a local region; Indicates the radius of the local region window.

[0109] The system performs directional consistency analysis on multiple heat flow direction vectors within a local region window and generates a heat flow convergence state based on the directional convergence relationship between these vectors. When the directions of multiple heat flow direction vectors gradually converge towards the interior of the local region, the system determines that there is a heat accumulation trend in the current region. The heat flow convergence state is represented as follows:

[0110] ;

[0111] in: Indicates the first The state of heat flow convergence in each region; This represents the result of superimposing heat flow direction vectors in a local region; This represents the magnitude of the heat flow direction vector.

[0112] The heat flow convergence state increases when the heat flow direction in a local area gradually converges towards a single area; conversely, the heat flow convergence state decreases when the heat flow direction disperses and diffuses. The system further generates a heat flow convergence intensity based on the degree of directional convergence between multiple heat flow direction vectors, and uses this intensity in subsequent thermal field risk analysis.

[0113] After generating the heat flow convergence state, the system further analyzes the heat flow direction changes in local areas to identify the thermal field closure trend in the fire scene. Specifically, the system detects the heat flow direction change sequence in local areas. When the heat flow direction in a local area reverses, the system determines that there is a heat flow rewind phenomenon in the area; when multiple heat flow directions in a local area form a cyclic flow relationship, the system determines that there is a heat flow cyclic flow state in the area; when the heat flow direction in a local area continuously accumulates towards the obstacle area or wall area, the system determines that there is a state of obstructed heat flow diffusion in the area.

[0114] ;

[0115] in: Indicates the first Thermal field closure trend parameters in each region; This represents the number of heat flow direction vectors in a local region; Indicates the first The direction angle of the heat flow direction vector; This indicates the average heat flow direction angle in a local region.

[0116] When multiple heat flow directions in a local area gradually form a closed relationship, the thermal field closure tendency parameter increases; when the heat flow directions remain divergent, the thermal field closure tendency parameter decreases. The system jointly analyzes the thermal field closure tendency parameter and the heat flow convergence state to determine the heat accumulation state in the fire area.

[0117] This step involves performing directional consistency analysis and directional closure relationship analysis on the heat flow direction vector in a local area. This enables the system to identify heat flow accumulation areas in the fire scene based on the heat accumulation state during the heat flow diffusion process, thus providing a thermal structure analysis basis for subsequent thermal inertia retention analysis and the generation of thermal field risk confidence.

[0118] S4. Generate thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period.

[0119] Preferably, in step S4, the step of generating thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period includes:

[0120] Obtain the regional temperature change status over a continuous time period;

[0121] Detect the change in heat flux diffusion rate over a continuous time period;

[0122] Obtain the changes in wall thermal reflection over a continuous time period;

[0123] The degree of heat retention in the region is calculated based on the continuous temperature change, the slowed heat diffusion, and the enhanced heat reflection from the wall surface.

[0124] Thermal inertia retention parameters are generated based on the degree of heat retention in the region.

[0125] Specifically, after generating parameters for heat flow convergence and thermal field closure trends, the system further analyzes the continuous heat retention state in the fire area. Since local areas in the fire may experience heat accumulation that cannot dissipate in time, the system generates thermal inertia retention parameters based on the regional temperature change, heat flow diffusion, and wall thermal reflection states over a continuous time period. The system first continuously samples the thermal field state data of the same area at preset time intervals and establishes a time-series sampling window for the corresponding area. Within the continuous time period, the system acquires the temperature change state in the area and calculates the regional temperature change rate based on the temperature change relationship between adjacent sampling times. The regional temperature change rate is expressed as:

[0126] ;

[0127] in: Indicates the first Temperature change rate in each region; Indicates the first Each region at time The temperature value of the area below.

[0128] The system analyzes the continuous change of heat in a region based on the rate of temperature change over a continuous time period. When the rate of temperature change in a region remains low and the region temperature remains high, the system determines that there is a continuous tendency for heat to stagnate in the current region.

[0129] After acquiring the temperature change status of the area, the system further analyzes the change status of heat diffusion rate over a continuous time period. The system calculates the heat diffusion rate in the area based on the change relationship of the heat flow direction vector over the continuous time period. When the amplitude of the heat flow direction vector in the area continuously decreases, the system determines that the heat diffusion capacity of the current area is reduced. Simultaneously, the system acquires the change status of wall heat reflection over a continuous time period. When the intensity of wall heat reflection in the area continuously increases, the system determines that heat accumulation due to wall reflection exists in the local area.

[0130] After analyzing the continuous temperature change, slowed heat diffusion, and enhanced wall thermal reflection states, the system calculates the degree of heat retention in the current area. The thermal inertia retention parameter is expressed as:

[0131] ;

[0132] in: Indicates the first Thermal inertia retention parameters in each region; Indicates the number of consecutive samples;

[0133] Indicates the first The rate of temperature change at the next sampling time; Indicates the first Next sampling time.

[0134] When the rate of temperature change in a region continuously decreases, the thermal inertia retention parameter increases. When the heat diffusion rate in a region decreases and wall thermal reflection increases, the system further increases the thermal inertia retention parameter in the corresponding region to characterize the current state of heat retention. Furthermore, for regions with heat flow accumulation and a continuously increasing thermal field closure trend parameter, the system increases the update frequency of the thermal inertia retention parameter in the corresponding region to enhance the detection capability of localized heat accumulation. When the heat flow direction in a region undergoes re-diffusion, the system decreases the thermal inertia retention parameter in the corresponding region to reflect changes in the region's heat diffusion state.

[0135] This step involves a joint analysis of regional temperature changes, heat flow diffusion, and wall thermal reflection over a continuous time period. This enables the system to identify the continuous heat retention state in the fire area, thus providing a basis for the analysis of continuous heat accumulation in the subsequent thermal risk confidence generation process.

[0136] S5. Generate the confidence level of thermal field risk based on the heat flow convergence state, thermal field closure trend parameters, and thermal inertia retention parameters;

[0137] Preferably, in step S5, the step of generating the thermal field risk confidence score includes:

[0138] Risk mapping is performed on the heat flow convergence state;

[0139] Structural risk mapping is performed on the thermal field closure trend parameters;

[0140] Thermal accumulation risk mapping is performed on the thermal inertia retention parameters;

[0141] Multiple risk parameters after mapping are fused to generate a thermal field risk confidence score.

[0142] Specifically, after generating parameters for heat flow convergence, thermal field closure trend, and thermal inertia retention, the system further integrates and analyzes multiple risk parameters to generate the thermal field risk confidence level for the corresponding region. Since the risk formation process in a fire is related not only to the temperature state in a local area but also to the heat accumulation state, thermal structure closure state, and continuous heat retention state, the system employs a multi-risk parameter fusion approach to comprehensively analyze the thermal risk state of the current region. The system first performs risk mapping processing on the heat flow convergence state. When the heat flow direction in a region gradually converges towards the interior of the local area, the system increases the heat flow convergence risk value in the corresponding region; for regions where the heat flow direction continuously diverges, the system decreases the heat flow convergence risk value in the corresponding region.

[0143] After completing the heat flow convergence state mapping process, the system further performs structural risk mapping on the thermal field closure trend parameters. When the heat flow direction in a region exhibits rewinding, circulating flow, or obstructed heat diffusion, the system determines that the thermal structure in the current region has a closure trend and increases the structural risk value of the corresponding region. For regions where the heat flow direction continues to diffuse outward and no heat flow convergence region has formed, the system decreases the structural risk value of the corresponding region.

[0144] Subsequently, the system performs heat accumulation risk mapping on the thermal inertia retention parameters. When the heat retention time in a region increases and the heat flow diffusion rate continues to decrease, the system increases the heat accumulation risk value in the corresponding region; for regions where heat diffusion recovers or heat flow redisperses, the system decreases the heat accumulation risk value in the corresponding region.

[0145] After mapping multiple risk parameters, the system fuses these parameters to generate the thermal field risk confidence score for the corresponding region. The thermal field risk confidence score is expressed as:

[0146] ;

[0147] in: This represents the confidence level of thermal field risk in the i, ji, ji, j region; Indicates the state of heat flow convergence; Parameters indicating the tendency of thermal field closure; Indicates the thermal inertia retention parameter; , , This represents the weighting coefficient of the corresponding risk parameter.

[0148] The system dynamically adjusts the weighting coefficients of corresponding risk parameters based on the heat flow accumulation state, thermal structure closure state, and heat retention state in different regions. When the heat flow accumulation state in a region continuously increases, the weighting coefficient corresponding to the heat flow accumulation state is increased; when the thermal field closure trend parameter in a region continuously increases, the weighting coefficient corresponding to the thermal field closure trend parameter is increased; when the thermal inertia retention parameter in a region continuously increases, the weighting coefficient corresponding to the thermal inertia retention parameter is increased, thereby enhancing the system's ability to identify risks in areas with continuous local heat accumulation.

[0149] Furthermore, in cases where multiple adjacent heat flow accumulation areas exist, the system performs a joint analysis of the thermal risk confidence levels in these adjacent areas. When the thermal risk confidence levels in adjacent areas increase simultaneously, the system raises the risk association level of the corresponding area and marks it as a continuous thermal risk area for use in the subsequent dynamic risk threshold generation process.

[0150] This step integrates and analyzes the heat flow convergence state, thermal field closure trend parameters, and thermal inertia retention parameters, enabling the system to generate regional thermal field risk confidence based on the heat accumulation state, thermal structure change state, and continuous heat retention state. This provides a regional risk analysis basis for subsequent dynamic risk threshold generation and dynamic risk avoidance control processes.

[0151] S6. Generate dynamic risk thresholds based on thermal risk confidence and environmental escape capability, and determine the risk of the current area based on the dynamic risk thresholds.

[0152] Preferably, in step S6, the step of generating the dynamic risk threshold includes:

[0153] Obtain the status of passable paths and obstacle distribution in the current area;

[0154] Environmental escape capability is generated based on the number of passable paths and the density of obstacle distribution.

[0155] Obtain the basic risk threshold;

[0156] The basic risk threshold is dynamically adjusted based on the confidence level of the thermal field risk;

[0157] Dynamic risk thresholds are generated based on the adjusted risk thresholds and the ability to escape from the environment.

[0158] Preferably, the steps for dynamically adjusting the basic risk threshold based on the thermal field risk confidence level include:

[0159] Lower the baseline risk threshold when the confidence level of thermal field risk increases; raise the baseline risk threshold when heat flow diffusion is enhanced; lower the baseline risk threshold when the tendency of thermal field closure is enhanced.

[0160] Lower the baseline risk threshold when the ability to escape from the environment decreases.

[0161] Specifically, after generating the thermal risk confidence level, the system further dynamically adjusts the risk threshold based on the environmental escape capability of the current area to generate a dynamic risk threshold for the corresponding area. Since the thermal risk state in the fire environment is related to the area's escape conditions, the system does not use a fixed risk threshold to determine the area's risk state, but rather dynamically adjusts the risk threshold based on both the area's thermal risk state and its escape status. The system first obtains the status of traversable paths and the distribution of obstacles in the current area. The traversable path status represents the number of evacuation directions in the current area, and the obstacle distribution status represents the degree of obstacle aggregation in the current area. The system generates the environmental escape capability based on the number of traversable paths and the obstacle density in the area. The environmental escape capability is expressed as:

[0162] ;

[0163] in: Indicates the first Environmental escape capacity in each area; This indicates the number of passable paths in the current area; This indicates the density of obstacle distribution in the current area.

[0164] The environmental escape capability increases when the number of traversable paths in a region increases; conversely, the environmental escape capability decreases when the density of obstacles in a region increases. The system determines the risk tolerance status of the current region based on its environmental escape capability.

[0165] After generating environmental escape capability, the system acquires a basic risk threshold and dynamically adjusts it based on the thermal field risk confidence level. When the thermal field risk confidence level in a region continuously increases, the system lowers the basic risk threshold in the corresponding region to improve the sensitivity of regional risk detection. When heat flow diffusion in a region intensifies and the heat flow direction diverges again, the system raises the basic risk threshold in the corresponding region to reduce the possibility of false alarms. When the thermal field closure trend parameter in a region continuously increases, the system lowers the basic risk threshold in the corresponding region to enhance the system's ability to identify risks in areas where heat continues to accumulate. When the environmental escape capability in a region decreases, the system further lowers the basic risk threshold in the corresponding region to improve the robot's response speed to avoidance in areas with low escape capability.

[0166] The system generates a dynamic risk threshold based on the adjusted risk threshold and the environmental escape capability. The dynamic risk threshold is expressed as:

[0167] ;

[0168] in: Indicates the first Dynamic risk thresholds in each region; Indicates the basic risk threshold; Indicates the confidence level of thermal field risk; Indicates the ability to escape from the environment; This represents the risk adjustment coefficient; This represents the escape compensation coefficient.

[0169] After generating a dynamic risk threshold, the system assesses the risk of the current area based on the relationship between the thermal risk confidence level and the dynamic risk threshold. When the thermal risk confidence level exceeds the dynamic risk threshold, the system determines that the current area meets the risk triggering conditions; when the thermal risk confidence level is lower than the dynamic risk threshold, the system determines that the current area is passable. Furthermore, for multiple adjacent areas within a continuous thermal risk area, the system further adjusts the dynamic risk threshold in a coordinated manner based on the changing trends of thermal risk between the areas. When the thermal risk confidence level in adjacent areas increases simultaneously, the system synchronously lowers the dynamic risk threshold in the corresponding area to enhance the system's response capability to the regional thermal risk diffusion process.

[0170] This step dynamically adjusts the basic risk threshold based on the confidence level of the thermal field risk and the environmental escape capability, enabling the system to generate dynamic risk thresholds based on the regional thermal risk status and regional escape status, thereby providing a basis for regional risk determination in the subsequent dynamic risk avoidance and control process.

[0171] S7. When the risk triggering conditions are met in the current area, generate a safe escape direction based on the heat flow direction field, and control the fire-fighting robot to perform dynamic risk avoidance actions.

[0172] Preferably, in step S7, the steps of generating a safe escape direction based on the heat flow direction field and controlling the firefighting robot to perform dynamic avoidance actions include:

[0173] Identify the thermal risk zone; determine the heat flow accumulation direction based on the heat flow direction vector in the thermal risk zone; generate a safe escape direction based on the opposite direction of the heat flow accumulation direction; control the fire-fighting robot to perform evacuation actions along the safe escape direction.

[0174] Specifically, after generating the dynamic risk threshold, the system further assesses the risk of the current area based on the relationship between the thermal risk confidence level and the dynamic risk threshold. When the thermal risk confidence level in an area exceeds the dynamic risk threshold in the corresponding area, the system determines that the current area meets the risk triggering condition and marks the corresponding area as a thermal risk area. The system analyzes the heat accumulation state of the area based on the heat flow direction field in the thermal risk area and determines the heat accumulation direction based on multiple heat flow direction vectors in the area. Since there is a correspondence between the heat diffusion direction in the fire and the expansion direction of the dangerous area, the system determines the heat accumulation trend in the current area based on the directional change relationship of the heat flow direction vectors.

[0175] The system first extracts multiple heat flow direction vectors from the thermal risk region, and then generates the regional heat flow accumulation direction based on the superposition of these vectors. The regional heat flow accumulation direction is represented as:

[0176] ;

[0177] in: Indicates the first The direction of heat flow concentration in each region; This represents the set of thermal risk areas;

[0178] This represents the direction vector of heat flow in the thermal risk zone.

[0179] The system generates a safe escape direction based on the opposite direction of heat flow accumulation. The safe escape direction is represented as:

[0180] ;

[0181] in: Indicates the direction of safe escape; Indicates the direction of heat flow concentration; This indicates the magnitude of the direction of heat flow concentration.

[0182] The system adjusts the motion control direction of the firefighting robot based on the safe escape direction and controls the robot to perform evacuation actions along that direction. During the evacuation, the system continuously acquires thermal field state data in the current area and regenerates the thermal flow direction field, thermal field risk confidence level, and dynamic risk threshold to perform real-time risk assessment of the robot's current position. When the thermal field risk confidence level in the corresponding area continuously decreases during the robot's movement, the system maintains the current evacuation direction; when the thermal field risk confidence level in the area ahead of the robot increases again during its movement, the system regenerates the safe escape direction to adjust the robot's evacuation path.

[0183] Furthermore, when multiple thermal risk areas are simultaneously clustered, the system generates a regional risk diffusion trend based on the changing heat flow directions between these areas. When multiple thermal risk areas show a convergence trend, the system increases the evacuation priority in the corresponding areas and prioritizes directions away from areas of heat flow accumulation as safe escape routes. For areas with obstacles or narrow passages, the system modifies the safe escape direction based on environmental escape capabilities to reduce the likelihood of firefighting robots entering areas with low escape capabilities.

[0184] This step generates a safe escape direction based on the heat flow direction in the hot zone of the fire, and controls the firefighting robot to perform dynamic avoidance actions based on the safe escape direction. This allows the system to dynamically adjust the robot's evacuation path according to the direction of heat diffusion in the fire, thereby reducing the likelihood of the firefighting robot entering an area where heat continues to accumulate.

[0185] Example 2:

[0186] Reference Figure 2 In a second embodiment of the present invention, the present invention provides a dynamic risk avoidance system for firefighting robots with an adaptive threshold for thermal field risk confidence, comprising:

[0187] The thermal field status data acquisition module acquires thermal field status data in the fire scene environment where the firefighting robot is located.

[0188] The heat flow direction field construction module constructs the heat flow direction field based on the heat field state data and generates a set of heat flow direction vectors within the corresponding region;

[0189] The heat flow convergence analysis module calculates the heat flow convergence state within a local area based on the heat flow direction vector set;

[0190] The thermal field closure trend analysis module generates thermal field closure trend parameters based on the relationship between heat flow direction changes.

[0191] The thermal inertia retention analysis module generates thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period.

[0192] The thermal field risk confidence generation module generates thermal field risk confidence based on heat flow convergence state, thermal field closure trend parameters, and thermal inertia retention parameters.

[0193] The dynamic risk threshold generation module generates dynamic risk thresholds based on thermal field risk confidence and environmental escape capability.

[0194] The dynamic risk avoidance control module generates a safe escape direction based on the heat flow direction field when the risk triggering conditions are met in the current area, and controls the fire-fighting robot to perform dynamic risk avoidance actions.

[0195] Specifically, the thermal field state data acquisition module is used to acquire temperature data, thermal radiation data, smoke concentration data, airflow velocity data, and wall thermal reflection data in the fire area, and performs regional grid mapping processing on each type of thermal field state data. The heat flow direction field construction module is used to generate thermal gradient information based on the temperature change relationship between adjacent areas, and generate heat flow direction vectors based on the thermal gradient information to construct the heat flow direction field in the corresponding area.

[0196] The heat flow convergence analysis module generates the heat flow convergence state based on multiple heat flow direction vectors in a local area, and generates the heat flow convergence intensity based on the degree of directional convergence between these vectors. The thermal field closure trend analysis module detects the heat flow direction reversal state, heat flow circulation state, and heat flow diffusion obstruction state in a local area, and generates thermal field closure trend parameters based on the heat flow convergence area.

[0197] The thermal inertia retention analysis module generates thermal inertia retention parameters based on the regional temperature change, heat flow diffusion, and wall thermal reflection over a continuous time period. The thermal field risk confidence generation module fuses the heat flow convergence state, thermal field closure trend parameters, and thermal inertia retention parameters to generate a thermal field risk confidence score.

[0198] The dynamic risk threshold generation module generates environmental escape capability based on the number of passable paths and obstacle distribution density in the current area, and generates a dynamic risk threshold based on the thermal field risk confidence level and environmental escape capability. The dynamic avoidance control module generates a safe escape direction based on the heat flow direction vector in the thermal field risk area when the risk triggering conditions are met in the current area, and controls the fire-fighting robot to perform dynamic avoidance actions along the safe escape direction.

[0199] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A dynamic risk avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence, characterized in that, Includes the following steps: S1. Obtain thermal field state data of the fire scene environment where the fire robot is located. The thermal field state data includes temperature data, thermal radiation data, smoke concentration data, airflow velocity data, and wall thermal reflection data. S2. Construct a heat flow direction field based on the thermal field state data, and generate a set of heat flow direction vectors within the corresponding region; S3. Calculate the heat flow convergence state in the local area based on the heat flow direction vector set, and generate thermal field closure trend parameters according to the heat flow direction change relationship; S4. Generate thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period. S5. Generate a thermal field risk confidence level based on the heat flow convergence state, the thermal field closure trend parameters, and the thermal inertia retention parameters; S6. Generate a dynamic risk threshold based on the thermal field risk confidence level and environmental escape capability, and determine the risk of the current area based on the dynamic risk threshold. S7. When the risk triggering conditions are met in the current area, a safe escape direction is generated according to the heat flow direction field, and the fire-fighting robot is controlled to perform dynamic risk avoidance actions.

2. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S2, the step of generating the set of heat flow direction vectors within the corresponding region includes: Divide the fire area into regional grids; Obtain the temperature change status in each grid region; Thermal gradient information is generated based on the temperature change relationship between adjacent grid regions; The direction of heat diffusion is determined based on the thermal gradient information; Generate a heat flow direction vector in the corresponding grid region based on the heat diffusion direction; The heat flow direction field is constructed based on the heat flow direction vectors in multiple grid regions.

3. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S3, the step of calculating the heat flow convergence state within a local area based on the heat flow direction vector set includes: Extract multiple heat flow direction vectors in a local region; Calculate the directional consistency among multiple heat flow direction vectors; The heat flow convergence state is generated based on the directional convergence relationship between multiple heat flow direction vectors; The heat flow convergence intensity is generated based on the degree of directional convergence between multiple heat flow direction vectors.

4. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S3, the step of generating the thermal field closure trend parameters based on the heat flow direction change relationship includes: Detecting the reversal of heat flow direction in a localized area; detecting the circulation of heat flow in a localized area; detecting the obstruction of heat flow diffusion in a localized area; The heat flow accumulation region is determined based on the heat flow direction reversal state, heat flow circulation state, and heat flow diffusion obstruction state; the thermal field closure trend parameter is generated based on the heat flow accumulation region.

5. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S4, the step of generating thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period includes: Obtain the regional temperature change status over a continuous time period; Detect the change in heat flux diffusion rate over a continuous time period; Obtain the changes in wall thermal reflection over a continuous time period; The degree of heat retention in the region is calculated based on the continuous temperature change, the slowed heat diffusion, and the enhanced heat reflection from the wall surface. Thermal inertia retention parameters are generated based on the degree of heat retention in the region.

6. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S5, the step of generating the thermal field risk confidence score includes: Risk mapping processing is performed on the aforementioned heat flow convergence state; Structural risk mapping is performed on the thermal field closure trend parameters; The thermal inertia retention parameters are subjected to thermal accumulation risk mapping processing; Multiple risk parameters after mapping are fused to generate a thermal field risk confidence score.

7. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S6, the step of generating the dynamic risk threshold includes: Obtain the status of passable paths and obstacle distribution in the current area; Environmental escape capability is generated based on the number of passable paths and the density of obstacle distribution. Obtain the basic risk threshold; The basic risk threshold is dynamically adjusted based on the aforementioned thermal field risk confidence level; A dynamic risk threshold is generated based on the adjusted risk threshold and the environmental escape capability.

8. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 7, characterized in that, The steps for dynamically adjusting the basic risk threshold based on the thermal field risk confidence level include: Lower the baseline risk threshold when the confidence level of thermal field risk increases; raise the baseline risk threshold when heat flow diffusion is enhanced; lower the baseline risk threshold when the tendency of thermal field closure is enhanced. Lower the baseline risk threshold when the ability to escape from the environment decreases.

9. The dynamic hazard avoidance method for firefighting robots with adaptive threshold for thermal field risk confidence as described in claim 1, characterized in that, In step S7, the step of generating a safe escape direction based on the heat flow direction field and controlling the firefighting robot to perform dynamic avoidance actions includes: Identify the thermal risk zone; determine the heat flow accumulation direction based on the heat flow direction vector in the thermal risk zone; generate a safe escape direction based on the opposite direction of the heat flow accumulation direction; control the fire-fighting robot to perform evacuation actions along the safe escape direction.

10. A dynamic risk avoidance system for firefighting robots with adaptive threshold for thermal risk confidence, characterized in that, A dynamic hazard avoidance method for fire-fighting robots applied to the adaptive threshold of thermal field risk confidence as described in any one of claims 1-9, comprising: The thermal field status data acquisition module acquires thermal field status data in the fire scene environment where the firefighting robot is located. The heat flow direction field construction module constructs a heat flow direction field based on the heat field state data and generates a set of heat flow direction vectors within the corresponding region; The heat flow convergence analysis module calculates the heat flow convergence state within a local area based on the heat flow direction vector set; The thermal field closure trend analysis module generates thermal field closure trend parameters based on the relationship between heat flow direction changes. The thermal inertia retention analysis module generates thermal inertia retention parameters based on the temperature change state, heat flow diffusion state, and wall thermal reflection state over a continuous time period. The thermal field risk confidence generation module generates thermal field risk confidence based on heat flow convergence state, thermal field closure trend parameters, and thermal inertia retention parameters. The dynamic risk threshold generation module generates dynamic risk thresholds based on thermal field risk confidence and environmental escape capability. The dynamic risk avoidance control module generates a safe escape direction based on the heat flow direction field when the risk triggering conditions are met in the current area, and controls the fire-fighting robot to perform dynamic risk avoidance actions.

Images