Fire rescue path planning method based on artificial intelligence

Abstract

The invention relates to the technical field of fire rescue path planning, and discloses a fire rescue path planning method based on artificial intelligence, and the method comprises the steps: 1, building a dynamic environment model through the modeling of the state of a firefighter, enabling the state of the firefighter to be represented by variables of multiple dimensions, forming a state vector, and enabling the state vector to be updated according to the change of time, the current state of the firefighter is dynamically reflected; step 2, constructing an optimization objective function; data decentralized sharing in fire rescue path planning is realized by adopting a block chain technology, the bottleneck of information transmission limitation in a traditional centralized system is avoided, firemen, command centers and participants can be ensured to seamlessly share real-time path planning, fire dynamic and building structure data, a decentralized data transmission mechanism is adopted, and the data transmission efficiency is improved. The cooperative combat efficiency of all parties is improved, the high efficiency and instantaneity of data transmission are ensured, and the effects of quick response and decision support are obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of fire rescue path planning, and specifically provides a fire rescue path planning method based on artificial intelligence. Background Art

[0002] With the rapid development of modern cities, the frequency and complexity of fire disaster events are increasing day by day. Traditional fire rescue path planning methods rely on centralized information management and path optimization systems, which have problems such as limited data sharing, lagged information transmission, and single-point failures of the system. These problems cause firefighters to face the dilemmas of insufficient and untimely information during the rescue process, thereby affecting rescue efficiency and safety.

[0003] In dealing with fire disasters, the information transmission between firefighters and the command center is crucial. Traditional centralized systems are prone to data lag and coordination difficulties, making it impossible for all parties to share key fire dynamics and on-site information in a timely manner. In addition, there are significant trust risks in the current fire safety information system. The data is controlled by a few central nodes and is vulnerable to tampering or leakage threats, unable to guarantee the reliability of the information.

[0004] To improve the level of fire safety informatization management and ensure that firefighters can complete rescue tasks efficiently and safely, there is an urgent need for a new technical solution to solve the data transmission bottleneck, trust issues, and system failure risks in existing systems. In this context, blockchain technology, due to its decentralized and immutable characteristics, provides a new solution for fire rescue path planning. Through blockchain technology, decentralized sharing of data can be achieved, ensuring the efficiency and immediacy of information transmission, enhancing the trustworthiness of data, and guaranteeing the security of data and the high availability of the system through distributed ledger technology and redundant backup mechanisms.

[0005] Therefore, the present invention provides a fire rescue path planning method based on artificial intelligence by combining a path planning method based on artificial intelligence with blockchain technology to solve the problems mentioned above. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the present invention provides a fire rescue path planning method based on artificial intelligence to solve the problems mentioned in the above background art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions: A fire rescue path planning method based on artificial intelligence, including:

[0008] Step 1, by modeling the state of firefighters, a dynamic environment model is established. The state of the firefighters is represented by variables of multiple dimensions, forming a state vector, and the state vector is updated with the change of time to dynamically reflect the current state of the firefighters;

[0009] Step 2, construct an optimization objective function. The optimization objective function evaluates the quality of the path by comprehensively considering factors such as the time consumption, path length, control energy, and potential hazards of the firefighter's path. The design of the objective function is based on the dynamic distribution of the firefighter's current position, target position, and environmental factors;

[0010] Step 3, apply the optimal control theory to solve the optimal path. By setting control variables to adjust the movement mode of the firefighter, minimize the objective function, and then derive the control strategy according to the optimal control equation, so that the firefighter can move along the predetermined path and can respond to dynamic environmental changes in real time;

[0011] Step 4, based on the obtained path planning result, use deep reinforcement learning to dynamically optimize the path planning. At each moment, according to the environmental state and path planning feedback, the agent adjusts the path planning strategy so that the firefighter can adapt to environmental changes and select the optimal path;

[0012] Step 5, combined with the path planning result, use blockchain technology to achieve data sharing and security guarantee. Firefighters, command centers, and other participants share real-time location information, fire dynamics, and building structure data through the blockchain system. The data is verified and stored through encryption and consensus mechanisms;

[0013] Step 6, based on the distributed ledger technology of the blockchain, in the case of partial node failures and attacks, the remaining nodes can continue to provide complete path planning and environmental data to ensure the continuous operation of the firefighter's path planning system in a disaster environment.

[0014] Preferably, in the state vector in Step 1 where, x 1 (t) represents the position of the firefighter in two-dimensional and three-dimensional spaces, x 2 (t) represents the speed of the firefighter, x n (t) represents the attitude and acceleration dynamic factors, and the state vector is updated in real time according to time changes to dynamically reflect the state of the firefighter.

[0015] Preferably, the optimization objective function J in Step 2 is designed as:

[0016]

[0017] where, represents the speed of the firefighter, u(t) represents the control input, ||u(t)|| 2 is the control energy consumption, f h (x(t)) represents the influence of the dangerous area on the path, f f(x(t)) is the influence of the fire spread, and w 1 、w 2 、w 3 、w 4 are the corresponding weight coefficients, is the position change rate of the firefighter, J represents the overall cost of path planning, T is the total planning time, and dt represents the time step of time.

[0018] Preferably, in step 3, the optimal control theory is applied through the optimal control equation:

[0019]

[0020] wherein, represents the rate of change with respect to time t, H is the Hamiltonian, is the derivative of the state variable, is the partial derivative of the Hamiltonian with respect to the state change rate x, the partial derivative of the Hamiltonian with respect to the state variable ;

[0021] To solve the control strategy of the optimal path, the Hamiltonian H(x(t), u(t), λ(t)) is:

[0022]

[0023] wherein, H(x(t), u(t), λ(t)) is the Hamiltonian, which includes the state variable x(t), the control variable u(t) and the Lagrange multiplier λ(t), and L(x(t), u(t)) is the path optimization cost function, is the rate of change of the state variable.

[0024] Preferably, in step 4, deep reinforcement learning is used to optimize path planning, and the reward function R(S(t), A(t)) is designed as:

[0025] R(S(t), A(t)) = -γ·di(S(t), go) - λ·ha(S(t)) + μ·time e (A(t)),

[0026] wherein, R(S(t), A(t)) is the reward value under the current state S(t) and the action A(t) taken, γ is the discount factor,

[0027] di(S(t), go) is the distance from the current state S(t) to the target position go, λ is the weight coefficient,

[0028] ha(S(t)) is the danger level at the current state S(t), μ is the weight coefficient,

[0029] time e (A(t)) is the time efficiency of the current action A(t).

[0030] Preferably, the deep reinforcement learning in step 4 uses Q-learning to update the path planning strategy, and the update formula is:

[0031]

[0032] where Q(S t , A t ) is the Q value of taking action A t in the current state S t , α is the learning rate,

[0033] R t+1 is the immediate reward from the current moment t to the next moment t + 1,

[0034] γ is the discount factor, is the maximum Q value of taking action a t+1 in the next state S ′ ,

[0035] Q(S t+1 , a ′ ) is the Q value of taking action a t+1 in the next state S ′ .

[0036] Preferably, the blockchain technology is used in step 5 to achieve data sharing and security guarantee. Firefighters, the command center, and participants share real-time data through the blockchain, including the location information of firefighters, fire dynamics, and building structures, and data verification and storage are carried out through encryption and consensus mechanisms.

[0037] Preferably, the data is verified and stored through the smart contract of the blockchain to ensure the authenticity and immutability of the shared data. The smart contract automatically executes the data update, verification, and recording process according to the pre-set rules.

[0038] Preferably, the blockchain technology in step 5 ensures the decentralized sharing of fire safety information. Through the mechanism of the distributed ledger, real-time synchronization of data is achieved among firefighters, the command center, and participants, ensuring that all parties do not need to rely on a centralized server.

[0039] Preferably, the distributed ledger technology of the blockchain guarantees the redundancy and high availability of data. When nodes fail and are attacked, the remaining nodes can continue to provide path planning and environmental data, ensuring the continuous and stable operation of the path planning system for firefighters in a disaster environment.

[0040] The present invention provides a fire rescue path planning method based on artificial intelligence. It has the following beneficial effects:

[0041] 1. By adopting blockchain technology, the present invention realizes the decentralized sharing of data in fire rescue path planning, avoids the bottleneck of limited information transmission in traditional centralized systems, ensures that firefighters, command centers, and participants can seamlessly share real-time data on path planning, fire dynamics, and building structures, and the decentralized data transmission mechanism improves the efficiency of collaborative operations among all parties, ensures the high efficiency and immediacy of data transmission, and achieves the effects of rapid response and decision support.

[0042] 2. Through the immutability and consensus mechanism of the blockchain, the present invention solves the trust risk of fire safety information. The shared fire data is encrypted and stored and verified through the consensus mechanism to ensure the authenticity and consistency of the data. The records of data changes are all publicly traceable, greatly reducing the risk of information tampering, ensuring decisions based on trustworthy data, and achieving the effect of enhancing the reliability of fire rescue decisions.

[0043] 3. Through the distributed ledger technology and redundant backup mechanism of the blockchain, the present invention ensures the autonomy and multi-redundancy of data, avoids the risk of system collapse caused by single-point failures. Each node can store and update complete fire rescue path planning and environmental data. Even if some nodes fail and are attacked, the remaining nodes can continue to provide complete data, ensuring the high availability and data security of the system, and achieving the effect of continuous and stable operation in disaster scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] To enable those skilled in the art to understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0046] The present invention will be described in detail below with reference to the accompanying drawings:

[0047] Embodiment:

[0048] Please refer to the attached Figure 1 , the embodiment of the present invention provides a fire rescue path planning method based on artificial intelligence, including:

[0049] Step 1: Establish a dynamic environment model by modeling the state of firefighters. The state of firefighters is represented by variables in multiple dimensions, forming a state vector, which is updated with the change of time to dynamically reflect the current state of firefighters.

[0050] Step 2: Construct an optimization objective function. The optimization objective function evaluates the quality of a path by comprehensively considering factors such as the time consumption, path length, control energy, and potential hazards of the firefighter's path. The design of the objective function is based on the dynamic distribution of the current position of the firefighter, the target position, and environmental factors.

[0051] Step 3: Apply the optimal control theory to solve for the optimal path. By setting control variables to adjust the movement mode of the firefighter, the objective function is minimized, and then the control strategy is derived according to the optimal control equation, enabling the firefighter to move along the predetermined path and respond to dynamic environmental changes in real time.

[0052] Step 4: Based on the obtained path planning results, use deep reinforcement learning to dynamically optimize the path planning. At each moment, according to the environmental state and path planning feedback, the agent adjusts the path planning strategy so that the firefighter can adapt to environmental changes and select the optimal path.

[0053] Step 5: Combine the path planning results and use blockchain technology to achieve data sharing and security guarantee. Firefighters, the command center, and other participants share real-time location information, fire dynamics, and building structure data through the blockchain system. The data is verified and stored through encryption and consensus mechanisms.

[0054] Step 6: Based on the distributed ledger technology of blockchain, in the case of failures and attacks on some nodes, the remaining nodes can continue to provide complete path planning and environmental data, ensuring the continuous operation of the firefighter's path planning system in a disaster environment.

[0055] The advantage of Step 1 is that it can achieve real-time update and adaptive adjustment of path planning, ensuring that firefighters can respond to changes in the disaster environment in a timely manner.

[0056] The advantage of Step 2 is that it can effectively avoid dangerous areas and optimize energy consumption. Through the comprehensive optimization objective, it can ensure that the path selected by firefighters is both short and safe, effectively improving the rescue efficiency and safety.

[0057] The advantage of Step 3 is that it can provide the optimal movement plan for firefighters under different environmental conditions, ensuring that firefighters move along the most effective path in a dynamic environment and having the ability to adjust in real time to adapt to sudden environmental changes.

[0058] The advantage of Step 4 is that through continuous interaction with the environment, deep reinforcement learning can gradually improve the efficiency and reliability of the path planning strategy, enabling firefighters to make accurate and efficient path choices when facing complex and unpredictable disaster environments.

[0059] The advantage of Step 5 is that the decentralized feature of blockchain can solve the bottleneck problem of centralized data management, realizing instant sharing and high credibility of information. Encrypted storage and consensus mechanisms can effectively prevent data tampering and leakage, enhancing the security of the system.

[0060] The advantage of Step 6 is that through distributed storage and multi-redundancy mechanisms, high availability of data and stability of the system are ensured. Even in a disaster environment, the system can still operate stably, ensuring the smooth progress of rescue tasks.

[0061] In the state vector in Step 1 where, x 1 (t) represents the position of the firefighter in two-dimensional and three-dimensional spaces, x 2 (t) represents the speed of the firefighter, x n (t) represents attitude and acceleration dynamic factors, and the state vector is updated in real time according to time changes to dynamically reflect the state of the firefighter.

[0062] The state modeling method can provide real-time information on spatial positions and capture the motion state of the firefighter, ensuring that path planning takes into account the actual motion situation.

[0063] The state vector is updated in real time according to time changes, which can quickly respond to the state changes of the firefighter in a complex environment. This means that whether during the rescue process or when the fire and building structure change, the state information of the firefighter can be updated in a timely manner, providing the latest environmental data for subsequent path planning, enabling the path planning to adapt to the changing on-site environment.

[0064] Through the dynamically updated state vector, path planning can more accurately predict the next action of the firefighter and optimize path selection. Especially when sudden situations such as fire spreading and building collapse occur, timely state updates can improve the response speed of the system, ensuring that the firefighter selects the safest and most efficient path and reducing risks during the rescue process.

[0065] The optimization objective function J in Step 2 is designed as:

[0066]

[0067] where, represents the speed of the firefighter, u(t) represents the control input, ||u(t)|| 2 is the control energy consumption, f h(x(t)) represents the impact of the dangerous area on the path, and f f (x(t)) is the impact of the fire spread, and w 1 、w 2 、w 3 、w 4 are the corresponding weight coefficients, is the rate of change of the position of the firefighter, J represents the overall cost of path planning, T is the total planning time, and dt represents the time step of time.

[0068] The optimization objective function J in step 2 comprehensively considers multiple important factors of the firefighter's path. The multi-objective optimization method ensures that path planning focuses on the shortest path and can consider indicators such as path safety and energy conservation. Comprehensive consideration improves the scientific nature of path planning and can ensure that firefighters complete tasks in the safest way in the shortest time.

[0069] Adding factors related to path safety to the objective function has the advantage that when optimizing the path, it can avoid firefighters entering dangerous areas and near the fire source to the greatest extent, ensuring the safety of firefighters. By adjusting the weight coefficients, the system can flexibly balance path safety and efficiency, ensuring that the work efficiency and life safety of firefighters go hand in hand.

[0070] The optimization objective function includes a term for controlling energy consumption. By optimizing the control of the input u(t), it can effectively reduce the energy consumption of firefighters during the execution of tasks. This means that path planning focuses on speed and safety, can reduce unnecessary energy waste, improve the sustainability of task execution, and reduce the fatigue of personnel.

[0071] By adding a time factor, the objective function can optimize the time consumption of the path, ensuring that when performing tasks, firefighters can complete rescue tasks efficiently, shorten the response time, and extinguish the fire and rescue trapped personnel in a timely manner.

[0072] In step 3, the optimal control theory is applied through the optimal control equation:

[0073]

[0074] where, represents the rate of change with respect to time t, H is the Hamiltonian, is the derivative of the state variable, is the partial derivative of the Hamiltonian with respect to the state change rate x, The partial derivative of the Hamiltonian with respect to the state variable ;

[0075] To solve the control strategy for the optimal path, the Hamiltonian H(x(t), u(t), λ(t)) is:

[0076]

[0077] Among them, H(x(t), u(t), λ(t)) is the Hamiltonian, which includes the state variable x(t), the control variable u(t), and the Lagrange multiplier λ(t). L(x(t), u(t)) is the path optimization cost function. is the rate of change of the state variable.

[0078] Optimal control theory provides a rigorous mathematical framework for path planning problems. By applying the optimal control equations, it is possible to accurately solve how to optimize the path under given constraints and determine the optimal choice of control variables.

[0079] As the core in the optimal control problem, the Hamiltonian synthesizes the state variables, control inputs, and cost functions of the system, and is the key to measuring the optimality of the system during the path optimization process. By setting the Hamiltonian, from the perspective of control theory, it guides firefighters towards the optimal path, ensuring the path is efficient and safe, and avoiding the risks brought by blindly choosing paths.

[0080] In the optimal control equations, the state variable x(t) and the control variable u(t) are related to each other through the partial derivatives of the Hamiltonian. The real-time feedback mechanism can dynamically adjust the path planning according to the current state at each moment. When the environment changes, the optimal control theory can update the control strategy in real time to ensure that firefighters always move along the optimal path.

[0081] In step 4, deep reinforcement learning is used to optimize the path planning, and the reward function R(S(t), A(t)) is designed as:

[0082] R(S(t), A(t)) = -γ·di(S(t), go) - λ·ha(S(t)) + μ·time e (A(t)),

[0083] Among them, R(S(t), A(t)) is the reward value under the current state S(t) and the action A(t) taken. γ is the discount factor.

[0084] di(S(t), go) is the distance from the current state S(t) to the target position go. λ is the weight coefficient.

[0085] ha(S(t)) is the degree of danger at the current state S(t). μ is the weight coefficient.

[0086] time e (A(t)) is the time efficiency of the current action A(t).

[0087] The deep reinforcement learning in step 4 optimizes the path planning, and uses the reward function to comprehensively consider the factors of path safety, time efficiency and danger, providing an adaptive, flexible and continuously optimized path selection method. By dynamically adjusting the weight coefficient, the system can optimize the path selection according to the specific task requirements. At the same time, through the introduction of the discount factor, the intelligent body can weigh the long-term and short-term goals to ensure the comprehensiveness and efficiency of the path planning. Deep reinforcement learning improves the accuracy of path planning, and enhances the system's adaptability to environmental changes and ability to cope with complex situations, thereby improving the efficiency and safety of fire rescue tasks.

[0088] The deep reinforcement learning in step 4 uses Q-learning to update the path planning strategy. The update formula is:

[0089]

[0090] Among them, Q(S t ,A t ) is the current state S t Next take action A t The Q value, α is the learning rate,

[0091] R t+1 is the immediate reward from the current moment t to the next moment t+1,

[0092] γ is the discount factor, In the next state S t+1 Take action a ′ The maximum Q value,

[0093] Q(S t+1 ,a ′ ) is in the next state S t+1 Take action a ′ Q value.

[0094] By updating the path planning strategy through Qleaning, the system can continuously optimize the path selection based on the immediate feedback of each decision. The continuous learning mechanism enables the system to flexibly adapt to complex and dynamic environmental conditions, and balance short-term and long-term goals through discount factors to ensure the efficiency and safety of path planning. The introduction of the Qleaning algorithm enables path planning to quickly respond to the current environment and can be optimized according to potential changes in the future, providing an adaptive and reliable path planning solution, significantly improving the adaptability and decision-making quality in fire rescue missions.

[0095] In step 5, blockchain technology is used to achieve data sharing and security guarantee. Firefighters, the command center, and participants share real-time data through the blockchain, including the location information of firefighters, fire dynamics, and building structures, and the data is verified and stored through encryption and consensus mechanisms.

[0096] The data is verified and stored through the smart contract of the blockchain to ensure the authenticity and immutability of the shared data. The smart contract automatically executes the data update, verification, and recording process according to the pre-set rules.

[0097] The blockchain technology in step 5 ensures the decentralized sharing of fire safety information. Through the mechanism of the distributed ledger, instant synchronization of data is achieved among firefighters, the command center, and participants, ensuring that all parties do not need to rely on a centralized server.

[0098] The distributed ledger technology of the blockchain guarantees the redundancy and high availability of data. When a node fails or is attacked, the remaining nodes can continue to provide path planning and environmental data, ensuring the continuous and stable operation of the path planning system for firefighters in a disaster environment.

[0099] The blockchain technology in step 5 effectively solves the problems of data transmission bottlenecks, information tampering, and single-point system failures existing in traditional fire rescue path planning through decentralized data sharing, encryption verification, smart contracts, and distributed ledger mechanisms. Through decentralized data storage and synchronization, the system can share and verify data in real time, ensuring the accuracy and security of firefighters' path planning. In addition, the redundant backup and high-availability design of the blockchain ensure the continuous and stable operation of the system in a disaster environment, providing reliable and intelligent technical support for fire rescue tasks.

[0100] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A fire rescue path planning method based on artificial intelligence, characterized in that: include: Step 1, establishing a dynamic environment model by modeling the firefighter's state, wherein the firefighter's state is represented by variables of multiple dimensions to form a state vector, which is updated by the change of time to dynamically reflect the current state of the firefighter; Step 2, constructing an optimization objective function, wherein the optimization objective function evaluates the quality of the path by comprehensively considering the time consumption, path length, control energy and potential danger factors of the firefighter's path, and the objective function is designed based on the dynamic distribution of the firefighter's current position, target position and environmental factors; Step 3: Apply the optimal control theory to solve the optimal path, adjust the firefighter's movement mode by setting the control variables to minimize the objective function, and then derive the control strategy based on the optimal control equation, so that the firefighter can move according to the predetermined path and respond to dynamic environmental changes in real time; Step 4: Based on the obtained path planning results, deep reinforcement learning is used to dynamically optimize the path planning. At each moment, according to the environmental state and path planning feedback, the intelligent agent adjusts the path planning strategy so that the firefighters can adapt to environmental changes and choose the optimal path; Step 5: Combined with the path planning results, blockchain technology is used to achieve data sharing and security assurance. Firefighters, command centers, and other participants share real-time location information, fire dynamics, and building structure data through the blockchain system. The data is verified and stored through encryption and consensus mechanisms. Step 6: Based on the distributed ledger technology of blockchain, when some nodes fail or are attacked, the remaining nodes can continue to provide complete path planning and environmental data, ensuring that the firefighters' path planning system continues to operate in a disaster environment.

2. The fire rescue path planning method based on artificial intelligence according to claim 1 is characterized in that: The state vector in step 1 In the above example, x1(t) represents the position of the firefighter in two-dimensional and three-dimensional space, x2(t) represents the speed of the firefighter, and x n (t) represents the dynamic factors of posture and acceleration, and the state vector is updated in real time according to the change of time to dynamically reflect the state of the firefighter.

3. The fire rescue path planning method based on artificial intelligence according to claim 1 is characterized in that: The optimization objective function J in step 2 is designed as: in, represents the speed of the firefighter, u(t) represents the control input, ||u(t)|| 2 To control energy consumption, f h (x(t)) represents the impact of the dangerous area on the path, f f (x(t)) is the influence of fire spread, and w1, w2, w3, w4 are the corresponding weight coefficients, is the rate of change of the firefighter’s position, J represents the overall cost of path planning, T is the total time of planning, and dt represents the time step.

4. The fire rescue path planning method based on artificial intelligence according to claim 1 is characterized in that: In step 3, the optimal control theory is applied through the optimal control equation: in, represents the rate of change with respect to time t, H is the Hamiltonian, is the derivative of the state variable, is the partial derivative of the Hamiltonian with respect to the rate of change of state x, Hamiltonian versus state variables The partial derivative of To solve the control strategy of the optimal path, the Hamiltonian H(x(t),u(t),λ(t)) is: Among them, H(x(t),u(t),λ(t)) is the Hamiltonian, which contains the state variable x(t), the control variable u(t) and the Lagrange multiplier λ(t), L(x(t),u(t)) is the path optimization cost function, is the rate of change of the state variable.

5. The fire rescue path planning method based on artificial intelligence according to claim 1 is characterized in that: In step 4, deep reinforcement learning is used to optimize path planning, and the reward function R(S(t), A(t)) is designed as: R(S(t),A(t))=-γ·di(S(t),go)-λ·ha(S(t))+μ·time e (A(t)), Among them, R(S(t),A(t)) is the reward value under the current state S(t) and the action A(t), γ is the discount factor, di(S(t),go) is the distance from the current state S(t) to the target position go, λ is the weight coefficient, ha(S(t)) is the risk level of the current state S(t), μ is the weight coefficient, time e (A(t)) is the time efficiency of the current action A(t).

6. The fire rescue path planning method based on artificial intelligence according to claim 5 is characterized in that: The deep reinforcement learning in step 4 adopts Q-learning to update the path planning strategy, and the update formula is: Among them, Q(S t ,A t ) is the current state S t Next take action A t The Q value, α is the learning rate, R t+1 is the immediate reward from the current moment t to the next moment t+1, γ is the discount factor, In the next state S t+1 Take action a ′ The maximum Q value, Q(S t+1 ,a ′ ) is in the next state S t+1 Take action a ′ Q value.

7. The fire rescue path planning method based on artificial intelligence according to claim 6 is characterized in that: In step 5, blockchain technology is used to achieve data sharing and security assurance. Firefighters, command centers, and participants share real-time data through blockchain, including firefighters' location information, fire dynamics, and building structure, and data is verified and stored through encryption and consensus mechanisms.

8. The fire rescue path planning method based on artificial intelligence according to claim 7 is characterized in that: The data is verified and stored through the blockchain's smart contract to ensure the authenticity and non-tamperability of the shared data. The smart contract automatically executes the data update, verification and recording process according to pre-set rules.

9. The fire rescue path planning method based on artificial intelligence according to claim 7 is characterized in that: The blockchain technology in step 5 ensures the decentralized sharing of fire safety information. Through the mechanism of distributed ledger, the real-time synchronization of data between firefighters, command centers and participants is achieved, ensuring that all parties do not need to rely on centralized servers.

10. The fire rescue path planning method based on artificial intelligence according to claim 9, characterized in that: The distributed ledger technology of the blockchain ensures data redundancy and high availability. If a node fails or is attacked, the remaining nodes can continue to provide path planning and environmental data, ensuring that the firefighters' path planning system operates continuously and stably in a disaster environment.

Images