A method and terminal for optimizing traffic operation status based on the city brain

Abstract

This invention provides a traffic operation state optimization method and terminal based on a city brain. By constructing a cloud reflection arc for the city brain, it is used to identify traffic conditions and control traffic signal timing to adjust traffic conditions. It can react in real time to changes in traffic conditions of various road segments. In this cloud reflection arc, cloud sensors are responsible for acquiring the real-time traffic conditions of the current road, and then transmitting them to the cloud neural center to calculate the entropy value of the real-time traffic conditions. The cloud neural center then makes decisions based on the entropy value, and finally, the cloud effector adjusts the traffic operation state of the current road in real time based on the decision results to optimize traffic conditions. The entropy value of the real-time traffic conditions is calculated using traffic spatiotemporal entropy, which enables a comprehensive assessment of traffic conditions. This not only effectively reduces the processing pressure of the city brain on a large amount of traffic data, but also enables real-time control, adjustment and optimization of local traffic conditions.

Description

Technical Field

[0001] This invention relates to the field of urban traffic management and optimization technology, and in particular to a method and terminal for optimizing traffic operation status based on a city brain. Background Technology

[0002] With the rapid development of technology, urban management is entering a new intelligent era. At the forefront of this era, the city brain has become a highly anticipated concept. The city brain is no longer merely the lifeblood of a city, but rather an intelligent system based on information technology that provides unprecedented management and service capabilities to the city through the aggregation, analysis, and application of data.

[0003] For a city's transportation network, since the traffic conditions of each road are different, if the city brain only makes a global optimization strategy for the entire city's transportation network system, it may not be able to alleviate the traffic conditions of some roads. If the traffic conditions of each road are processed and analyzed by the city brain to generate a corresponding optimization decision, this will create a lot of unnecessary load on the city brain. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a traffic operation status optimization method and terminal based on the city brain, which can effectively reduce the processing pressure of the city brain on a large number of traffic data, and at the same time realize the real-time optimization of local traffic conditions.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A method for optimizing traffic operation status based on a city brain includes the following steps:

[0007] S1. Construct a cloud reflex arc for the city brain based on traffic signal timing adjustment, wherein the cloud reflex arc includes a cloud sensor, a cloud neural center, and a cloud effector;

[0008] S2. Obtain the real-time traffic conditions of the current road through the cloud sensor;

[0009] S3. Based on the traffic spatiotemporal entropy calculation formula, the entropy value of the real-time traffic road conditions is calculated through the cloud neural center;

[0010] S4. The cloud nerve center makes a decision based on the entropy value and obtains the decision result;

[0011] S5. Based on the decision results, the traffic operation status of the current road is adjusted in real time through the cloud effector.

[0012] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0013] A traffic operation status optimization terminal based on a city brain includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:

[0014] S1. Construct a cloud reflex arc for the city brain based on traffic signal timing adjustment, wherein the cloud reflex arc includes a cloud sensor, a cloud neural center, and a cloud effector;

[0015] S2. Obtain the real-time traffic conditions of the current road through the cloud sensor;

[0016] S3. Based on the traffic spatiotemporal entropy calculation formula, the entropy value of the real-time traffic road conditions is calculated through the cloud neural center;

[0017] S4. The cloud nerve center makes a decision based on the entropy value and obtains the decision result;

[0018] S5. Based on the decision results, the traffic operation status of the current road is adjusted in real time through the cloud effector.

[0019] The beneficial effects of this invention are as follows: It provides a traffic operation state optimization method and terminal based on the city brain. By constructing a cloud reflection arc for the city brain, it is used to identify traffic road conditions and control traffic signal timing to adjust traffic road conditions. It can react in real time to changes in traffic operation conditions of various road segments. In this cloud reflection arc, the cloud sensor is responsible for acquiring the real-time traffic road conditions of the current road, and then transmitting it to the cloud neural center to calculate the entropy value of the real-time traffic road conditions. The cloud neural center then makes a decision based on the entropy value. Finally, the cloud effector adjusts the real-time timing of the traffic operation state of the current road based on the decision result to optimize the traffic road conditions. The entropy value of the real-time traffic road conditions is calculated using traffic spatiotemporal entropy, which can achieve a comprehensive assessment of traffic conditions. This not only effectively reduces the processing pressure of the city brain on a large amount of traffic data, but also enables real-time control, adjustment and optimization of local traffic road conditions. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the main process of a traffic operation status optimization method based on a city brain in an embodiment of the present invention.

[0021] Figure 2 This is a flowchart illustrating a traffic operation status optimization method based on a city brain, as described in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the real-time traffic conditions of the current road in an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of the structure of a traffic operation status optimization terminal based on the city brain in an embodiment of the present invention;

[0024] Label Explanation:

[0025] 1. A traffic operation status optimization terminal based on the city brain; 2. Memory; 3. Processor. Detailed Implementation

[0026] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0027] Please refer to Figures 1 to 3 A traffic operation status optimization method based on urban brain includes the following steps:

[0028] S1. Construct a cloud reflex arc for the city brain based on traffic signal timing adjustment, wherein the cloud reflex arc includes a cloud sensor, a cloud neural center, and a cloud effector;

[0029] S2. Obtain the real-time traffic conditions of the current road through the cloud sensor;

[0030] S3. Based on the traffic spatiotemporal entropy calculation formula, the entropy value of the real-time traffic road conditions is calculated through the cloud neural center;

[0031] S4. The cloud nerve center makes a decision based on the entropy value and obtains the decision result;

[0032] S5. Based on the decision results, the traffic operation status of the current road is adjusted in real time through the cloud effector.

[0033] As can be seen from the above description, the beneficial effects of the present invention are as follows: It provides a traffic operation state optimization method and terminal based on the city brain. By constructing a cloud reflection arc for the city brain, it is used to identify traffic road conditions and control traffic signal timing to adjust traffic road conditions. It can react in real time to changes in traffic operation conditions of various road segments. In this cloud reflection arc, the cloud sensor is responsible for acquiring the real-time traffic road conditions of the current road, and then transmitting it to the cloud neural center to calculate the entropy value of the real-time traffic road conditions. The cloud neural center then makes a decision based on the entropy value. Finally, the cloud effector adjusts the real-time timing of the traffic operation state of the current road based on the decision result to optimize the traffic road conditions. The entropy value of the real-time traffic road conditions is calculated using traffic spatiotemporal entropy, which can achieve a comprehensive assessment of traffic conditions. This not only effectively reduces the processing pressure of the city brain on a large amount of traffic data, but also enables real-time control, adjustment and optimization of local traffic road conditions.

[0034] Further, step S3 specifically includes:

[0035] S31. The real-time traffic conditions collected by the cloud sensors are transmitted to the cloud nerve center via the cloud afferent nerve. The cloud nerve center then calculates the entropy value H of the real-time traffic conditions according to the traffic spatiotemporal entropy calculation formula, which is:

[0036] H = -∑ i P(s i ,t,l)γ(ρ)logP(s i ,t,l) (1);

[0037] Where s represents the set of all traffic states, s i Let P(s) represent the i-th traffic state in the set s. i (t, l) represents the state s that occurs at time t and spatial location l. i The probability, where ρ represents vehicle density, is used as an adjustment factor, and its formula is:

[0038] γ(ρ)=a+bρ (2);

[0039] Where a and b are the learnable weight parameters of the adjustment factor ρ;

[0040] S32. The change of the traffic spatiotemporal entropy is calculated according to formula (3):

[0041] ΔH (t,l) =H (t+Δt,l) -H (t,l) (3);

[0042] The evolution of traffic conditions is assessed by the change in the entropy value of traffic conditions within a time interval Δt.

[0043] As described above, entropy is an important concept in information theory, used to measure the degree of uncertainty or disorder in information. In the urban traffic management of this invention, changes in real-time traffic conditions can be considered as the entropy of the current road. By acquiring various information about real-time traffic conditions collected by cloud sensors, such as the speed, density, and flow of traffic at intersections, the entropy value of the current road is calculated. Based on the entropy value, the timing of traffic lights can be adjusted using corresponding cloud reflection arcs to achieve effective, timely, and accurate control and optimization of local traffic conditions. The entropy value is determined by considering the probability distribution of various traffic states s. Traffic conditions provide comprehensive information; meanwhile, to adapt to dynamic changes, a time-state-based vehicle density ρ is introduced as an adjustment factor in the traffic spatiotemporal entropy calculation formula. This factor adjusts the calculation of spatiotemporal entropy according to changes in vehicle density, thus more sensitively reflecting changes in traffic conditions. Specifically, by adding the adjustment factor, the change in the entropy value of the traffic spatiotemporal entropy formula becomes more pronounced when vehicle density increases, and it more sensitively reflects the transition from smooth traffic to congestion, enabling a more accurate capture of dynamic changes in traffic conditions. In addition, the evolution of traffic conditions is further assessed by calculating the changes in traffic state entropy values ​​over a certain time interval.

[0044] Further, step S4 specifically includes:

[0045] S41. Determine whether the entropy value has a preset threshold, using the following formula:

[0046]

[0047] Wherein, E is the value obtained after determining whether the entropy value exceeds the preset threshold;

[0048] S42. If the entropy value exceeds the preset threshold, proceed to step S43; otherwise, return to step S2 to continue obtaining the real-time traffic road conditions after the current road.

[0049] S43. Determine whether the cloud reflection arc has formed based on the spontaneous response of the traffic lights on the current road. If the traffic lights on the current road can spontaneously adjust according to the value E after the entropy value exceeds the preset threshold, then determine that the cloud reflection arc has been formed and proceed to step S45; otherwise, proceed to step S44.

[0050] S44. The entropy value E after exceeding the preset threshold is transmitted to the cloud central nervous system through the cloud afferent nerve for learning to obtain the optimal decision to solve the current traffic congestion, and the construction of the cloud reflex arc is improved based on the real-time traffic road conditions and the optimal decision.

[0051] S45. The entropy value E after exceeding the preset threshold is transmitted to the cloud effector through the cloud efferent nerve.

[0052] As described above, this invention adds the construction of a cloud reflection arc. The formation of the cloud reflection arc depends on whether the system can spontaneously and autonomously process and resolve the current traffic congestion situation when the entropy value exceeds a preset threshold in the real-time traffic conditions. In other words, the key indicator of whether the cloud reflection arc has been formed is whether the system can immediately and spontaneously adjust when the traffic conditions change without human intervention. Therefore, if the cloud reflection arc has not yet been fully formed, the system can learn the optimal strategy for resolving the current traffic congestion based on the obtained real-time traffic conditions using existing commonly used algorithms such as neural network algorithms. This will generate the corresponding cloud reflection arc, enabling the execution of the optimal solution strategy in the future. Furthermore, it can avoid repeatedly overloading the computing power of the city's brain when encountering the same congestion situation again.

[0053] Furthermore, the cloud effector is a traffic light, and step S5 specifically involves:

[0054] The cloud effector analyzes the value E to obtain the decision result, and adjusts the traffic light timing of the current road in real time according to the decision result.

[0055] As described above, the value E after the entropy exceeds the preset threshold is the decision result obtained by the cloud neural center based on the real-time traffic conditions of the current road. Therefore, the cloud effector adjusts the timing of its traffic lights according to the value of E. That is, when the entropy is low, it means that the traffic on the current road is relatively orderly, and a shorter red and green light duration can be used; when the entropy is high, it means that the traffic on the current road is relatively chaotic, and a longer red and green light duration can be used to organize the traffic flow, so as to adapt to the changes in different time periods and traffic flow.

[0056] Furthermore, the cloud sensor is a sensor and camera installed on the current road to sense and collect traffic flow data information on the current road.

[0057] As described above, by placing various sensors and cameras at the locations on roads where traffic operation status optimization is needed, the real-time traffic conditions of the roads can be accurately obtained, which can then be used for cloud reflection arc construction and decision calculation.

[0058] Please refer to Figure 4 A traffic operation status optimization terminal based on a city brain includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:

[0059] S1. Construct a cloud reflex arc for the city brain based on traffic signal timing adjustment, wherein the cloud reflex arc includes a cloud sensor, a cloud neural center, and a cloud effector;

[0060] S2. Obtain the real-time traffic conditions of the current road through the cloud sensor;

[0061] S3. Based on the traffic spatiotemporal entropy calculation formula, the entropy value of the real-time traffic road conditions is calculated through the cloud neural center;

[0062] S4. The cloud nerve center makes a decision based on the entropy value and obtains the decision result;

[0063] S5. Based on the decision results, the traffic operation status of the current road is adjusted in real time through the cloud effector.

[0064] As described above, the beneficial effects of this invention are as follows: Based on the same technical concept, and in conjunction with the aforementioned method for optimizing traffic operation status based on a city brain, a terminal for optimizing traffic operation status based on a city brain is provided. By constructing a cloud reflection arc for the city brain, it is used to identify traffic conditions and control traffic signal timing to adjust traffic conditions. It can react instantly to changes in traffic conditions on various road segments. In this cloud reflection arc, the cloud sensor is responsible for acquiring the real-time traffic conditions of the current road, and then transmitting it to the cloud neural center to calculate the entropy value of the real-time traffic conditions. The cloud neural center then makes a decision based on the entropy value, and finally the cloud effector adjusts the real-time timing of the traffic operation status of the current road based on the decision result to optimize the traffic conditions. The entropy value of the real-time traffic conditions is calculated using traffic spatiotemporal entropy, which enables a comprehensive assessment of traffic conditions. This not only effectively reduces the processing pressure of a large amount of traffic data on the city brain, but also enables real-time control, adjustment, and optimization of local traffic conditions.

[0065] Further, step S3 specifically includes:

[0066] S31. The real-time traffic conditions collected by the cloud sensors are transmitted to the cloud nerve center via the cloud afferent nerve. The cloud nerve center then calculates the entropy value H of the real-time traffic conditions according to the traffic spatiotemporal entropy calculation formula, which is:

[0067] H = -∑ i P(s i ,t,l)γ(ρ)logP(s i ,t,l) (1);

[0068] Where s represents the set of all traffic states, s i Let P(s) represent the i-th traffic state in the set s.i (t, l) represents the state s that occurs at time t and spatial location l. i The probability, where ρ represents vehicle density, is used as an adjustment factor, and its formula is:

[0069] γ(ρ)=a+bρ (2);

[0070] Where a and b are the learnable weight parameters of the adjustment factor ρ;

[0071] S32. The change of the traffic spatiotemporal entropy is calculated according to formula (3):

[0072] ΔH (t,l) =H (t+Δt,l) -H (t,l) (3);

[0073] The evolution of traffic conditions is assessed by the change in the entropy value of traffic conditions within a time interval Δt.

[0074] As described above, entropy is an important concept in information theory, used to measure the degree of uncertainty or disorder in information. In the urban traffic management of this invention, changes in real-time traffic conditions can be considered as the entropy of the current road. By acquiring various information about real-time traffic conditions collected by cloud sensors, such as the speed, density, and flow of traffic at intersections, the entropy value of the current road is calculated. Based on the entropy value, the timing of traffic lights can be adjusted using corresponding cloud reflection arcs to achieve effective, timely, and accurate control and optimization of local traffic conditions. The entropy value is determined by considering the probability distribution of various traffic states s. Traffic conditions provide comprehensive information; meanwhile, to adapt to dynamic changes, a time-state-based vehicle density ρ is introduced as an adjustment factor in the traffic spatiotemporal entropy calculation formula. This factor adjusts the calculation of spatiotemporal entropy according to changes in vehicle density, thus more sensitively reflecting changes in traffic conditions. Specifically, by adding the adjustment factor, the change in the entropy value of the traffic spatiotemporal entropy formula becomes more pronounced when vehicle density increases, and it more sensitively reflects the transition from smooth traffic to congestion, enabling a more accurate capture of dynamic changes in traffic conditions. In addition, the evolution of traffic conditions is further assessed by calculating the changes in traffic state entropy values ​​over a certain time interval.

[0075] Further, step S4 specifically includes:

[0076] S41. Determine whether the entropy value has a preset threshold, using the following formula:

[0077]

[0078] Wherein, E is the value obtained after determining whether the entropy value exceeds the preset threshold;

[0079] S42. If the entropy value exceeds the preset threshold, proceed to step S43; otherwise, return to step S2 to continue obtaining the real-time traffic road conditions after the current road.

[0080] S43. Determine whether the cloud reflection arc has formed based on the spontaneous response of the traffic lights on the current road. If the traffic lights on the current road can spontaneously adjust according to the value E after the entropy value exceeds the preset threshold, then determine that the cloud reflection arc has been formed and proceed to step S45; otherwise, proceed to step S44.

[0081] S44. The entropy value E after exceeding the preset threshold is transmitted to the cloud central nervous system through the cloud afferent nerve for learning to obtain the optimal decision to solve the current traffic congestion, and the construction of the cloud reflex arc is improved based on the real-time traffic road conditions and the optimal decision.

[0082] S45. The entropy value E after exceeding the preset threshold is transmitted to the cloud effector through the cloud efferent nerve.

[0083] As described above, this invention adds the construction of a cloud reflection arc. The formation of the cloud reflection arc depends on whether the system can spontaneously and autonomously process and resolve the current traffic congestion situation when the entropy value exceeds a preset threshold in the real-time traffic conditions. In other words, the key indicator of whether the cloud reflection arc has been formed is whether the system can immediately and spontaneously adjust when the traffic conditions change without human intervention. Therefore, if the cloud reflection arc has not yet been fully formed, the system can learn the optimal strategy for resolving the current traffic congestion based on the obtained real-time traffic conditions using existing commonly used algorithms such as neural network algorithms. This will generate the corresponding cloud reflection arc, enabling the execution of the optimal solution strategy in the future. Furthermore, it can avoid repeatedly overloading the computing power of the city's brain when encountering the same congestion situation again.

[0084] Furthermore, the cloud effector is a traffic light, and step S5 specifically involves:

[0085] The cloud effector analyzes the value E to obtain the decision result, and adjusts the traffic light timing of the current road in real time according to the decision result.

[0086] As described above, the value E after the entropy exceeds the preset threshold is the decision result obtained by the cloud neural center based on the real-time traffic conditions of the current road. Therefore, the cloud effector adjusts the timing of its traffic lights according to the value of E. That is, when the entropy is low, it means that the traffic on the current road is relatively orderly, and a shorter red and green light duration can be used; when the entropy is high, it means that the traffic on the current road is relatively chaotic, and a longer red and green light duration can be used to organize the traffic flow, so as to adapt to the changes in different time periods and traffic flow.

[0087] Furthermore, the cloud sensor is a sensor and camera installed at the current intersection to sense and collect traffic flow data information at the current intersection.

[0088] As described above, by placing various sensors and cameras at the locations on roads where traffic operation status optimization is needed, the real-time traffic conditions of the roads can be accurately obtained, which can then be used for cloud reflection arc construction and decision calculation.

[0089] This invention provides a traffic operation status optimization method and terminal based on a city brain, which is mainly applied in urban traffic management and control scenarios. The following is a detailed description with reference to specific embodiments:

[0090] Please refer to Figure 1 and Figure 3 Embodiment 1 of the present invention is as follows:

[0091] A traffic operation status optimization method based on the city brain, such as Figure 1 As shown, the steps include:

[0092] S1. Construct a city brain cloud reflex arc based on traffic signal timing adjustment.

[0093] In this embodiment, the cloud reflex arc includes a cloud receptor, a cloud nerve center, and a cloud effector.

[0094] S2. Obtain real-time traffic conditions of the current road through cloud sensors.

[0095] In this embodiment, the cloud sensor can be a sensor or camera installed on the current road. For example... Figure 3 The road sensors shown are used to sense and collect current road traffic data to accurately obtain real-time traffic conditions on the road, which can then be used for cloud reflection arc construction and decision-making calculations.

[0096] S3. Based on the traffic spatiotemporal entropy calculation formula, the entropy value of real-time traffic road conditions is calculated through the cloud neural center.

[0097] S4. The cloud nerve center makes decisions based on entropy values ​​and obtains the decision results.

[0098] S5. Based on the decision results, adjust the traffic operation status of the current road in real time through the cloud effector.

[0099] In this embodiment, a cloud reflex arc is constructed for the city brain to identify traffic conditions and control traffic signal timing to adjust traffic conditions. This allows for real-time responses to changes in traffic conditions on various road segments. In this cloud reflex arc, cloud sensors acquire real-time traffic conditions and transmit them to the cloud neural center to calculate the entropy value of the real-time traffic conditions. The cloud neural center then makes decisions based on the entropy value, and finally, cloud effectors adjust the real-time timing of traffic conditions based on the decision results to optimize traffic conditions. This effectively reduces the processing pressure of large amounts of traffic data on the city brain and enables real-time control, adjustment, and optimization of local traffic conditions.

[0100] like Figure 2 and Figure 3 As shown, Embodiment 2 of the present invention is as follows:

[0101] A traffic operation status optimization method based on the city brain, in this embodiment, step S3 specifically includes:

[0102] S31. The real-time traffic and road conditions collected by the cloud receptors are transmitted to the cloud nerve center via the cloud afferent nerve. The cloud nerve center then calculates the entropy value H of the real-time traffic and road conditions according to the traffic spatiotemporal entropy calculation formula. The traffic spatiotemporal entropy calculation formula is as follows:

[0103] H = -∑ i P(s i ,t,l)γ(ρ)logP(s i ,t,l) (1);

[0104] Where s represents the set of all traffic states, s i Let P(s) represent the i-th traffic state in the set s. i (t, l) represents the state s that occurs at time t and spatial location l. i The probability, where ρ represents vehicle density, is used as an adjustment factor, and its formula is:

[0105] γ(ρ)=a+bρ (2);

[0106] Where a and b are the learnable weight parameters of the adjustment factor ρ.

[0107] S32. The changes in traffic spatiotemporal entropy are calculated according to formula (3):

[0108] ΔH (t,l) =H (t+Δt,l)-H (t,l) (3);

[0109] The evolution of traffic conditions can be assessed by measuring the change in the entropy of traffic conditions over a time interval Δt.

[0110] In this embodiment, since entropy is an important concept in information theory, used to measure the degree of uncertainty or disorder in information, in the urban traffic management of this invention, changes in real-time traffic conditions can be regarded as the entropy of the current road. By acquiring various information about real-time traffic conditions collected by cloud sensors, such as the speed, density, and flow of traffic at intersections, the entropy value of the current road is calculated. Therefore, the timing of traffic lights can be adjusted using the corresponding cloud reflection arc based on the entropy value, achieving effective, timely, and accurate control and optimization of local traffic conditions. The entropy value is determined by considering the probability distribution of various traffic states s. Traffic conditions provide comprehensive information; meanwhile, to adapt to dynamic changes, a vehicle density ρ based on time state is introduced as an adjustment factor in the traffic spatiotemporal entropy calculation formula. This factor adjusts the calculation of spatiotemporal entropy according to changes in vehicle density, thus more sensitively reflecting changes in traffic conditions. Specifically, by adding the adjustment factor, the change in the entropy value of the traffic spatiotemporal entropy formula becomes more obvious when vehicle density increases, and it more sensitively reflects the transition from smooth traffic to congestion, enabling a more accurate capture of dynamic changes in traffic conditions. In addition, the evolution of traffic conditions is further assessed by calculating the change in traffic state entropy values ​​over a certain time interval.

[0111] At the same time, such as Figure 2 As shown, in this embodiment, step S4 specifically includes:

[0112] S41. Determine whether the entropy value has a preset threshold. The formula for this determination is:

[0113]

[0114] Where E is the value taken after determining whether the entropy value exceeds the preset threshold.

[0115] S42. If the entropy value exceeds the preset threshold, proceed to step S43; otherwise, return to step S2 to continue obtaining the real-time traffic conditions of the current road.

[0116] S43. Determine whether a cloud reflection arc has formed based on the spontaneous response of the traffic lights on the current road. If the traffic lights on the current road can spontaneously adjust according to the value E after the entropy value exceeds the preset threshold, then it is determined that a cloud reflection arc has been formed and proceed to step S45; otherwise, proceed to step S44.

[0117] S44. The entropy value E after exceeding the preset threshold is transmitted to the cloud central nervous system through the cloud afferent nerve for learning to obtain the optimal decision to solve the current traffic congestion, and the construction of the cloud reflex arc is improved based on the real-time traffic conditions and the optimal decision.

[0118] In this embodiment, the learning method that transmits the value E to the cloud central nervous system for learning to output the optimal decision can be implemented using DQN (i.e., Deep Q-Network, a Q-Learning algorithm based on deep learning). In this embodiment, some parameters of DQN are set as follows:

[0119] 1. State setting: In this embodiment, four states can be set, namely vehicle status information, signal phase scheme information, red light duration, and green light duration.

[0120] 2. Control Scheme Action Settings: In this embodiment, a four-phase control scheme is adopted, and the phase switching order and the duration of each phase are not fixed, thereby increasing the flexibility of traffic signal control. At each simulation step, the traffic light controller will execute the phase scheme automatically selected by intelligent AI. However, if the current action differs from the action at the next moment, a 3-second yellow light transition period is required to ensure traffic safety at the intersection.

[0121] 3. Reward settings: In this embodiment, rewards can be set for four components: cumulative vehicle delay time, vehicle throughput, green light duration, and red light duration.

[0122] 4. Specific operation process settings: In this embodiment, a simulation environment can be created. In the simulation environment, the simulation duration is set to T. At each practice step t, the traffic status s of the current intersection is collected through the cloud sensor. t The cloud sensor randomly selects an action a from the action space with probability ε. t , indicating exploration; the action corresponding to the maximum Q value in the neural network is selected with a probability of (1-ε) as a. t This indicates the use of previously learned knowledge.

[0123] Then execute the selected action a. t This refers to adjusting the phase and duration of traffic lights at intersections.

[0124] Then select action a t When applied to a road network environment, the system simulates vehicles driving according to a new traffic light control strategy and obtains the new road network traffic state. (t+1) and the corresponding reward value r t The reward value can be calculated based on the cumulative vehicle delay time, vehicle throughput, green light duration, and red light duration indicators.

[0125] Finally, the empirical tuple (s) t ,a t ,r t ,s (t+1) The learned knowledge is stored in an experience pool for subsequent experience replay. A batch of experiences is randomly sampled from the experience pool to train the DQN network and calculate the target Q value. Generally, the update rule of Q-Learning is used to update the weights of the DQN network to minimize the error between the predicted Q value and the target Q value. In this process, the probability of exploration can be reduced by gradually decreasing the ε value, thereby increasing the utilization of the learned knowledge.

[0126] Repeat the above steps until the simulation duration T ends.

[0127] S45. The entropy value E after exceeding the preset threshold is transmitted to the cloud effector through the cloud efferent nerve.

[0128] Because this invention adds the construction of a cloud reflection arc, the formation of a cloud reflection arc depends on whether the system can spontaneously and autonomously handle and resolve the current traffic congestion situation when the entropy value exceeds a preset threshold in the real-time traffic conditions. In other words, the key indicator of whether the cloud reflection arc has been formed is whether the system can immediately and spontaneously adjust when the traffic conditions change without human intervention. Therefore, if the cloud reflection arc has not yet been fully formed, the optimal strategy for resolving the current traffic congestion can be learned based on the obtained real-time traffic conditions using existing commonly used algorithms such as neural network algorithms. This will generate the corresponding cloud reflection arc, enabling the execution of the optimal solution strategy in the future. Furthermore, it can avoid repeatedly overloading the computing power of the city's brain when encountering the same congestion situation again.

[0129] like Figure 3 As shown, the cloud effector is a traffic light. In this embodiment, step S5 specifically involves:

[0130] The cloud effector analyzes the value E to obtain the decision result, and adjusts the timing of the traffic lights on the current road in real time based on the decision result.

[0131] The value E after the entropy exceeds the preset threshold is the decision result obtained by the cloud neural center based on the real-time traffic conditions of the current road. Therefore, the cloud effector adjusts the timing of its traffic lights according to the value of E. When the entropy is low, it means that the traffic on the current road is relatively orderly, and a shorter red and green light duration can be used; when the entropy is high, it means that the traffic on the current road is relatively chaotic, and a longer red and green light duration can be used to organize the traffic flow, so as to adapt to the changes in different time periods and traffic flow.

[0132] It is also worth noting that the real-time traffic and road conditions acquired each time can be used for the continuous learning and construction of the cloud reflection arc to improve the local optimization strategy for traffic and roads.

[0133] like Figure 4 As shown, Embodiment 3 of the present invention is as follows:

[0134] A traffic operation status optimization terminal 1 based on urban brain includes a memory 2, a processor 3, and a computer program stored on the memory 2 and run on the processor 3. When the processor 3 executes the computer program, it completes the steps in the traffic operation status optimization method based on urban brain in the above embodiment one or embodiment two.

[0135] In summary, the present invention provides a traffic operation state optimization method and terminal based on the city brain. By constructing a cloud reflection arc for the city brain, it is used to identify traffic conditions and control traffic signal timing to adjust traffic conditions. It can react in real time to changes in traffic operation conditions of various road segments. In this cloud reflection arc, the cloud sensor is responsible for acquiring the real-time traffic conditions of the current road, and then transmitting it to the cloud neural center to calculate the entropy value of the real-time traffic conditions. The cloud neural center then makes decisions based on the entropy value, and finally the cloud effector adjusts the traffic operation state of the current road in real time according to the decision results to optimize traffic conditions. This effectively reduces the processing pressure of the city brain on a large amount of traffic data, and also enables real-time control, adjustment and optimization of local traffic conditions.

[0136] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for optimizing traffic operation status based on a city brain, characterized in that: Including the following steps: S1. Construct a city brain cloud reflex arc based on traffic signal timing adjustment. The cloud reflex arc includes a cloud sensor, a cloud neural center, and a cloud effector. The cloud effector is a traffic light, and the cloud sensor is a sensor and a camera installed on the current road. S2. Obtain the real-time traffic conditions of the current road through the cloud sensor; S3. Based on the traffic spatiotemporal entropy calculation formula, the entropy value of the real-time traffic road conditions is calculated through the cloud neural center; S4. The cloud nerve center makes a decision based on the entropy value and obtains the decision result; S5. Based on the decision results, the traffic operation status of the current road is adjusted in real time using the cloud effector; Step S3 specifically involves: S31. The real-time traffic conditions collected by the cloud sensors are transmitted to the cloud nerve center via the cloud afferent nerve, and the cloud nerve center calculates the entropy value of the real-time traffic conditions according to the traffic spatiotemporal entropy calculation formula. H The formula for calculating the spatiotemporal entropy of traffic is: （1）； in, s This represents the set of all traffic conditions. s i Indicates a collection s The first in i Traffic conditions, P ( s i , t , l ) indicates that at time t Time and spatial location are l State occurs in the following circumstances s i The probability, ρ Vehicle density, used as an adjustment factor, is expressed by the following formula: （2）； in, a and b Adjustment factor ρ Learnable weight parameters; S32. The change of the traffic spatiotemporal entropy is calculated according to formula (3): （3）； By time interval Δ t The change in the entropy value of the internal traffic condition assesses the evolution of traffic conditions; Step S4 specifically involves: S41. Determine whether the entropy value has a preset threshold, using the following formula: （4）； in, E The value is determined after determining whether the entropy value exceeds the preset threshold. S42. If the entropy value exceeds the preset threshold, proceed to step S43; otherwise, return to step S2 to continue obtaining the real-time traffic road conditions after the current road. S43. Determine whether the cloud reflection arc has formed based on the spontaneous response of the traffic lights on the current road. If the traffic lights on the current road can determine the value after the entropy value exceeds the preset threshold... E If the adjustment is made spontaneously, it is determined that the cloud reflection arc has been formed and proceeds to step S45; otherwise, proceeds to step S44. S44. The value obtained after the entropy value exceeds the preset threshold. E The cloud afferent nerve transmits the information to the cloud central nervous system for learning, and then obtains the optimal decision to solve the current traffic congestion. Based on the real-time traffic conditions and the optimal decision, the construction of the cloud reflex arc is improved. S45. The value obtained after the entropy value exceeds the preset threshold. E The signal is transmitted to the cloud effector via the cloud efferent nerve.

2. The traffic operation status optimization method based on urban brain according to claim 1, characterized in that, Step S5 specifically involves: The cloud effector analyzes the value. E Once the decision results are obtained, the traffic light timings for the current road are adjusted in real time based on these results.

3. The traffic operation status optimization method based on urban brain according to claim 1, characterized in that, The cloud sensor is used to sense and collect current traffic flow data on the road.

4. A traffic operation status optimization terminal based on the city brain, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the various steps of the traffic operation state optimization method based on any one of claims 1 to 3.

Images