Big data oriented intelligent traffic evacuation path fuzzy control system

Abstract

The current traffic evacuation path control system has high risk coefficient and path congestion, and low efficiency and system error coefficient. For this problem, a fuzzy control system of traffic evacuation path based on genetic method is proposed and designed in this paper. The data server, geographic information server, computing server, and application server are used to construct the system framework. The logical structure is divided into data source layer, data access layer, scheduling layer, computing model layer, and application interface layer. The function module is mainly composed of static data management module, emergency management module, dynamic data interface module, dynamic traffic assignment module, guidance information release module, and user management module. The system hardware is designed by using the logical structure in combination with the function module. In the system software, the coordinator-operator mode is introduced into the real-time computing operation mechanism. The interaction of the coordinator and the operator is to implement the user specified operational function. Traffic data is forecast by autoregressive model. It is substituted into the objective function of intelligent traffic evacuation and the genetic method is used to solve the objective function. At last, fuzzy control result of optimal traffic evacuation path is obtained. Experimental results show that the average risk coefficient in the evacuation process is about 0.27, the average time consuming is 0.3 h, and the congestion of the evacuation path is relatively low, so the fault tolerance coefficient of the system can be controlled within a reasonable range. The system has a good overall operation effect and is feasible.

Keywords

Big data intelligent transportation evacuation path fuzzy control system

1 Introduction

Intelligent traffic evacuation path control under the condition of big data is the focus of current research. The dynamic traffic assignment theory is the basis of the realization of the dynamic traffic guidance system, and the path selection is one of the important links of the dynamic traffic assignment. Therefore, the path selection model and its algorithm are the key and core technology of the dynamic traffic guidance system [1]. In the network system, the optimal path selection is to select the appropriate road resistance function, optimize the road network and select the appropriate algorithm to calculate the optimal path to meet certain conditions [2]. The path selection model and its algorithm can guide the driver to avoid traffic congestion through the fast solution of the optimal path from the departure place to the destination. It is of great significance to alleviate traffic congestion. Some outstanding achievements have been found in the research of traffic evacuation.

Achievement 1: A traffic evacuation system based on improved speed control strategy is proposed [3]. Dijkstra algorithm is used to plan multiple AGV paths. Through the improved speed control strategy, the speed of AGV at the collision node is controlled to solve the problem of traffic conflict. However, the system does not consider the security, and the risk coefficient of evacuation is high.

Achievement 2: An evacuation path identification system based on ant colony for congestion road traffic flow is proposed [4]. The dissipation rate of queuing vehicle and the moving speed of vehicle in the evacuation path are calculated. The evacuation path identification matrix of traffic flow under traffic accidents is built. The optimization identification of the evacuation path of traffic flow under traffic accidents is achieved. The stability of the system is good, but the efficiency of traffic evacuation is low.

Achievement 3: A traffic evacuation control system based on flow distribution is proposed [5]. The tunnel evacuation space is divided into different regions. The number of people to be evacuated is given for each region. The personnel in different regions are allocated to different liaison channels. The system is simple and easy to be realized, but the congestion degree of each evacuation path is high.

Achievement 4: A traffic evacuation control system based on macro traffic flow is proposed [6]. The corresponding discrete time flow evolution equation is established for the characteristics of the ground auxiliary route time made up of the lower ramp, the ground parallel road and the upper ramp. The equation of the auxiliary circuit system can be combined with the state evolution system of the existing expressway. The system can control the risk of traffic evacuation in a certain range, but the fault tolerance coefficient of the evacuation system is low.

The above existing system cannot achieve the effect of high security, high efficiency and low congestion. For this problem, a fuzzy control system for traffic evacuation path based on genetic method is proposed in this paper. The structure of this paper is as follows.

The system architecture is constructed. The system hardware logical structure is divided and the system function module is designed. Each function module improves the fault tolerance of the system, the efficiency and safety of the traffic evacuation, and reduces the congestion of the evacuationpath.

The system software is designed with real-time calculation operating mechanism in the background and calculation of optimal traffic evacuation path fuzzy control. The objective function is constructed and solved to achieve the optimal evacuation path fuzzy control.

Verification of fuzzy control system for traffic evacuation path based on genetic method.

Conclusions of this paper.

2 Material and methods

2.1 Fuzzy control system of intelligent traffic evacuation path

2.1.1 Hardware design of fuzzy control system for intelligent traffic evacuation path

In the proposed system, the hardware is designed as the modules of physical structure and external relation, physical structure, and functional part. With high fault tolerance, high efficiency and high safety, the low congestion of evacuation path for the design objective, the designed structure of physical structure and external relation is shown in Fig. 1.

Fig.1

Physical structure and external relation of fuzzy control system of intelligent traffic evacuation path.

The system framework is composed of 4 servers, which are data server, geo-information system (GIS) server, computing server, and application server. The basic form of the system framework for providing services is as follows. The user accesses the application server through various terminals and sends request to the application server through the various browser-oriented operations. Application server parses user requests into requests for themselves and three other servers. The corresponding return information is distributed in turn and received. The information is then processed to form the page information returned to the user.

According to the structure relation shown in Fig. 1, physical structure relationships among the various servers in the system are given by Fig. 2.

Fig.2

Physical structure of fuzzy control system of intelligent traffic evacuation path.

In Fig. 2, the main functions of four servers implementing the core functions of the system are as follows.

Data server: It mainly runs a comprehensive database that supports the framework of the real-time system. The functions include: The original traffic detection data is obtained from the existing data center of the emergency management department for data fusion to convert into the standardized basic data, which is then stored in the comprehensive database of the system; All the traffic flow feature data is to be stores in the system; Response to the requests of reading, writing, and updating of all data of the other three servers.

Computing server: It runs all kinds of computing models in real-time. The functions include: The input data is extracted from the database. Then the computations of dynamic traffic distribution, emergency path planning, emergency evacuation point distribution for evacuees, and dynamic equilibrium distribution of vehicle flow are implemented. The results are transferred to the data server for storage; Response to logical control instructions of various computing tasks from the Web application server; Response to the computing task instructions from the application server and execution of the corresponding computing task.

The main functions of the GIS server are: storage of the road network GIS data; Response to request from the application server. The request is parsed to the corresponding GIS data requirements and data requirements. The required data is obtained and integrates with the GIS data to obtain the visual information; Response to GIS information modification instructions from the application server.

Web application server: It is for response to web browsing requests from other terminals on the network. The user request is parsed into requirements for GIS data, traffic data, and computing. A request is sent to the other three servers and the corresponding return information is received. Page information is returned to browser users after synthesis processing [7].

The logical structure of the system is shown in Fig. 3.

Fig.3

The logical structure of fuzzy control system of intelligent traffic evacuation path.

In Fig. 3, it includes:

Data source: All kinds of traffic data collected and reported through traffic detection equipment (coil detector, microwave detector, video detector, etc.)

Data layer: Traffic synthesis database of the system for storage of static data, raw data, intermediate data, computing result data, and configuration data.

Data access layer: Other parts of the system perform tasks related to data through the data access layer, including data access, fusion, and cleaning.

Scheduling layer: Uninterrupted running coordinator module used to start the execution of various users’ specified computing tasks.

Computational model layer: Model modules for system decision support computing.

Application interface layer: It includes both the user-oriented web application interface and the external system oriented web service interface [8].

The design of the functional modules of fuzzy control system of intelligent traffic evacuation path is carried out and shown in Fig. 4.

Fig.4

Functional design of fuzzy control system of intelligent traffic evacuation path.

In Fig. 4, the functional modules include static data management module, emergency management module, dynamic data interface module, dynamic traffic assignment module, decision support module, guidance information release module, user management module, and log record and system maintenance module.

Static data management module. It is to manage all the static data required for the system running, including road network data, test equipment data, and guidance equipment data. It is for maintenance of road network traffic attribute data, management of guidance equipment, management of detection equipment [9], management of emergency facility, and generation of road network topology.

Emergency management module. It can realize the management of emergency. Add an emergency: Set the type, the time, the severity, and the scope of the impact of the emergency. Edit an emergency: Edit the attributes of the emergency [10]. Delete an emergency: Set the end time of an emergency and set it as a historical event.

Dynamic data interface module. It is used for various operations for dynamic data. Data acquisition: Read and store the collected traffic data. Data fusion and cleaning: Different traffic condition data fusion and normalization. Data access: The operations of reading and saving for dynamic traffic data.

Dynamic traffic assignment module. Dynamic traffic assignment parameter setting: Setting up the parameters required for the operation of a dynamic traffic distribution system. User equilibrium allocation: According to the principle of user equilibrium, the dynamic traffic assignment of OD data is carried out on the designated road network. Simulation of fixed path selection: A fixed vehicle path is used to simulate the driving behavior of the vehicle on the road network. Traffic estimation and prediction: The traffic status of road network is estimated and predicted based on dynamic traffic assignment according to real-time traffic collection data. The last is operation result output.

Guidance information release module. Release the guidance information and revise and delete the released information. Pedestrians are guided to walk or drive on the best road by guidance information.

User management module. It is for management of users and their system permissions. It includes the functions of user management, authority management, user registration, and user login and cancellation.

Log record and system maintenance module. Log record: record the usage of the system. Log archiving: archive and compress the history log with long time. Data archiving and cleaning: archive historical traffic data, emergency data, and so on. Clean up the intermediate data generated during the system running for saving the space [11]. System configuration: Database configuration, server network configuration, client network configuration, and system running parameter configuration.

The above functional modules basically have the performance of modifying, deleting, editing, and so on. This kind of performance can greatly improve the fault tolerance of the system.

2.1.2 Software design of fuzzy control system for intelligent traffic evacuation path

Based on the system hardware, through the software, the optimal evacuation results are obtained by controlling the system running efficiency, fault tolerance, security, and traffic congestion.

The software is composed of two parts. One is the real-time computing mechanism in the background, and the other is the best solution to solve the current congestion problem. The real-time computing operation mechanism in the background can be used for calculating any part of the system hardware. In order to ensure the stability of the system, the operation mode of the coordinator-operator is introduced in the framework of the real-time system, as shown in Fig. 5.

Fig.5

Running process of real-time running mechanism in the background.

With the operation mode in Fig. 5, the system implements the user-specified operational function through the interaction of the coordinator and the operator. As a system service, the coordinator runs uninterruptedly for 24 hours on the computing server. When a user calls a computing module to the computing server, the coordinator first checks whether the traffic data requirement for calculation is satisfied. If it is not satisfied, the coordinator will wait a while to try again. If the traffic data requirement is already satisfied, the coordinator will set the running parameters of the calculation module and call the data access module to read the input data. Then it starts an independent process to execute the computing module. After the successful execution of the computing module, the data access module is called to save the result data into the database. Because the computation module runs as an independent process, the exception of the program will only lead to the failure of the calculation task, without affecting the operation stability of the whole system.

The optimal traffic evacuation scheme is obtained by using the genetic method. It is necessary to predict the current period of traffic data before finding the optimal traffic evacuation scheme [12, 13]. The autoregressive model is used to predict the traffic data of the previous time period of the same road section, and the predicted value is regarded as the replacement data of the current time period. The prediction result is expressed as $α x = α_{1} x_{k} + α_{2} x_{k - 1} + \dots + α_{q} x_{k - q + 1}$ (1) where α_i is each order coefficient of the autoregressive model, which needs to be demarcated in advance according to the different sections of the section, x is the replacement data to be solved, if the current time period is the k + 1 th period, x_k, …, x_k-q+1 is the collected data at previous q time periods of the same road section, q is the used autoregressive order.

According to Equation (1), for the minimum risk coefficient min λ, the minimum congestion of each evacuation path min D, the highest evacuation efficiency min t and the highest fault tolerance coefficient of the system max C, the optimal objective function of intelligent traffic evacuation best (Z) is constructed and expressed as

$\begin{matrix} best (Z) & = & (\sum min λ min D min t \sum max C) \\ • \sum α x \end{matrix}$ (2)

The genetic method is used to solve the objective function. First, a set of initialized population is generated randomly. According to the number of elements in the objective function, the population is divided into a number of subpopulations, and the selection operation is performed on the subpopulations [14 –17]. The generated new subpopulations are merged into a complete population, and then crossover and mutation operations are performed for the newly generated complete population. The generated n solutions are put into the buffer pool. The optimal solution in the future iterative evolution process is replaced and stored. The fitness function is defined as

$\begin{matrix} fit (f (x)) \\ = {\begin{matrix} 1 - 0.5 \cdot [(f (x) - b) / a], | f (x) - b | < a \\ \frac{1}{1 + (f (x) - b) / a}, | f (x) - b | \geq a \end{matrix} \end{matrix}$ (3) where f (x) is the fitness function, fit (f (x)) is the optimal fitness function, a is the road section, b is the influence parameter of the fitness function. The selection operation is realized by roulette method. Crossover operation and mutation operation are realized by adaptive crossover rate p_c and mutation rate p_m. $\begin{matrix} p_{c} = {\begin{matrix} k_{1} (f_{max} - f^{1}) / (f_{max} - f_{avg}), & f^{1} \geq f_{avg} \\ k_{2} & f^{1} < f_{avg} \end{matrix} \end{matrix}$ (4) $\begin{matrix} p_{m} = {\begin{matrix} k_{3} (f_{max} - f_{i}) / (f_{max} - f_{avg}), & f_{i} \geq f_{avg} \\ k_{4} & f_{i} < f_{avg} \end{matrix} \end{matrix}$ (5) where f_max is the largest fitness function of the population, f_avg is the average fitness, f¹ is the larger fitness of two crossover individuals, f_i is the fitness of an individual to be mutated, k₁, k₂, k₃, and k₄ are the constants with the value range of [0, 1].

By using the above calculation, it is judged whether or not the termination condition is satisfied. If the maximum number of iterations is reached, the iteration is terminated and the optimal solution is output.

$\begin{matrix} best (Z) \\ = \frac{(\sum min λ min D min t \sum max C) • \sum α x}{fit (f (x)) • p_{m} • p_{c}} \end{matrix}$ (6)

3 Results

To verify the effectiveness of the proposed system, the experiment is carried out with matlab2017. The experimental data are derived from the city road network in a city. Assume the congestion occurs before 7:40 a.m. and the evacuation start at 7:40 a.m., and all the sections are two-way. The road resistance information of travel time and intersection delay is updated every 10 min. The following indexes of the proposed system are verified.

Traffic evacuation risk; Congestion degree of traffic evacuation path; Traffic evacuation efficiency; Fault tolerance of traffic evacuation path control system.

The experimental results are as follows.

From Fig. 6, it can be seen that, the evacuation risk coefficient curve of evacuation system based on improved speed control strategy is increased higher. There is a downward trend in 0.8 h, but the average risk coefficient is still up to 0.53. The evacuation risk coefficient curve of the evacuation route fuzzy control system based on genetic method is more stable. With the extension of the evacuation time, the risk coefficient has an upward trend. The average risk coefficient is about 0.27. The comparison results show that the proposed system has lower risk and high reliability.

Fig.6

Comparison of traffic evacuation risk coefficient between different systems.

From Fig. 7, it can be seen that, the degree of the system based on flow distribution is increased with the increase of the evacuees. It is in the state of low controllability. The traffic evacuation congestion curve of the genetic-based system has a rising trend in the initial stage of the system operation, but with the increasing number of evacuees, there is no uncontrollable situation. When the number of evacuees is 3000, the congestion curve tends to be slow. The superiority of the proposed system is verified.

Fig.7

Comparison of evacuation path congestion degree between different systems.

Form Fig. 8, it can be seen that, the traffic evacuation path recognition system based on ant colony has long time consuming, low efficiency, and the average time of evacuation is 0.59 h. The traffic evacuation path fuzzy control system based on genetic method has little change in evacuation time, and the average time is 0.3 h.

Fig.8

Comparison of evacuation efficiency between different systems.

From Fig. 9, the fault tolerance coefficient of the traffic evacuation control system based on macro traffic flow is low, which indicates that once the system is wrong, it is difficult to recover. The fuzzy control system of traffic evacuation path based on genetic method has high fault tolerance coefficient, which indicates that the system can quickly recover.

Fig.9

Comparison of fault tolerance between different systems.

4 Conclusions

For the problems existing in the current traffic evacuation path control system, a fuzzy control system of traffic evacuation path based on genetic method is proposed in this paper. In the proposed system, for the problem of high risk of traffic evacuation, the end time of emergency event is set up in the emergency management module. The parameters of the traffic flow in the road section of the road network can be quickly recovered to improve the efficiency of the traffic evacuation. For the problem of high traffic congestion degree, it is reduced through the dynamic traffic assignment module, which includes dynamic traffic assignment parameter setting, user equilibrium assignment, traffic estimation and prediction, and operation result output. The guidance information release module reduces the risk of traffic evacuation by guiding guide pedestrians to walk or drive on the best road. The function modules of the system all have the functions of modification and editing, which can greatly improve the fault tolerance of the system. Some suggestions are given for the further research: the evacuation of the traffic path should be considered more about the psychological conditions of the evacuees, so that the evacuation results can be more scientific.

Footnotes

Acknowledgments

1. Henan province 2018 scientific and technological breakthrough project.

Research on parallel algorithm for intelligent traffic path optimization in large data environment (182102210139).

2. Henan province 2018 scientific and technological breakthrough project.

Research on key technology of large data analysis platform for developing smart farmers under distributed heterogeneous environment (182102110277).

References

Lim

G.J.

, Baharnemati

M.R.

and Kim

S.J.

, An optimization approach for real time evacuation reroute planning, Annals of Operations Research238(1-2) (2016), 375–388.

Zhou

, Dong

, Wen

, et al., Modeling of crowd evacuation with assailants via a fuzzy logic approach, IEEE Transactions on Intelligent Transportation Systems17(9) (2016), 2395–2407.

Zhang

S.Y.

, Yang

Y.S.

, Liang

C.J.

, et al., Optimal control of multiple AGV path conflict in automated terminals, Journal of Transportation Systems Engineering and Information Technology17(2) (2017), 83–89.

Zhang

, Study on optimization of Urban traffic congestion evacuation route, Computer Simulation34(3) (2017), 200–203.

, Wang

R.C.

and Qiu

Q.Q.

, Passenger assignment model for emergency evacuation in metro rail tunnels, Journal of the China Railway Society38(4) (2016), 8–14.

S.X.

, Freeway network optimal control during emergency evacuation with buffering of ground bypass, Journal of Transportation Systems Engineering and Information Technology17(3) (2017), 47–53.

Ding

, Yang

, Chi

, et al., Enabling smart transportation systems: A parallel spatio-temporal database approach, IEEE Transactions on Computers65(5) (2016), 1377–1391.

Kalamaras

, Zamichos

, Salamanis

, et al., An interactive visual analytics platform for smart intelligent transportation systems management, IEEE Transactions on Intelligent Transportation Systems (99) (2017), 1–10.

Zhu

, Li

, Chen

, et al., Parallel transportation management and control system and its applications in building smart cities, IEEE Transactions on Intelligent Transportation Systems17(6) (2016), 1576–1585.

10.

Hathout

, Bekar

and Yayli

, N-Legendre and N-Slant curves in the unit tangent bundle of surfaces, Kuwait Journal of Science44(3) (2017), 28–33.

11.

Tian

, Yan

, Zhao

and Xiao

, Empirical study of value effect of enterprises space MA in the new normal background in China, Journal of Interdisciplinary Mathematics20(1) (2017), 219–230.

12.

Zhang

and Qiang

, Modeling of complex information system based on hierarchical decision-making theory, Journal of Discrete Mathematical Sciences and Cryptography20(1) (2017), 137–148.

13.

Fatima

, Mahomed

F.M.

and Khalique

C.M.

, Conditional symmetries of nonlinear third-order ordinary differential equations, Discrete and Continuous Dynamical Systems-Series S11(4SI) (2018), 655–666.

14.

and Liu

, Research on the phenomenon of chinese residents’ spiritual contagion for the reuse of recycled water based on SC-IAT, Water9(84611) (2017).

15.

Peng

, Ge

, Liu

and Furuta

, Adsorption characteristics of sulfur powder by bamboo charcoal to restrain sulfur allergies, Saudi Journal of Biological Sciences24(1) (2017), 103–107.

16.

Pyskunov

S.O.

, Maksimyk

Y.V.

and Valer

V.V.

, Finite element analysis of influence of non-homogenous temperature field on designed lifetime of spatial structural elements under creep conditions, Mathematics and Nonlinear Sciences1 (2016), 253–262.

17.

Awadalla

N.S.

, Hanna

M.A.

, Ismail

, Hassan

I.A.

and Elkhamisy

M.A.

, Period variation study and light curve analysis of the eclipsing binary gsc 02013-00288, Mathematics and Nonlinear Sciences1 (2016), 321–334.