CN105547310B - A kind of path planning apparatus and method based on the trip of PM2.5 health - Google Patents

Abstract

The invention discloses a path planning device and method based on PM2.5 healthy travel, including three parts: vehicle-mounted PM2.5 detection equipment, an information transmission module and a server terminal, wherein the information transmission module includes a GPS module and a GPRS module, and the server terminal The terminal includes a path planning module. The PM2.5 detection equipment is installed on the floating vehicle and connected to the information transmission module. The information transmission module is connected to the server terminal. The user's mobile device can easily access the data in the server terminal only with the help of a browser. . The invention adopts the improved Dijkstra algorithm, mainly takes the comprehensive concentration of PM2.5 as the impedance, and takes the minimum total concentration as the goal, and the health path planning follows the optimization function of the optimal path goal of PM2.5.

Description

A route planning device and method based on PM2.5 healthy travel

technical field

The invention belongs to a route planning and navigation system based on big data, and in particular is a device and method capable of detecting PM2.5 value and realizing healthy travel route planning.

Background technique

With the gradual improvement of the road network, the number of cars is increasing, and the high-density urbanization process has made the traffic situation increasingly tense and the air pollution is becoming more and more serious. Due to the complex composition of atmospheric particulate matter, the heavy metal elements and cyclohydrocarbon organic substances carried by it enter the human body along the respiratory tract, seriously endangering human health. Among them, PM2.5 is the most harmful. Its particle size is small and its specific surface area is large, so it is easier to accumulate toxic substances. According to research, PM2.5 pollution is positively correlated with the disease rate of adult respiratory system.

Therefore, when people travel, how to accurately obtain a large number of PM2.5 concentration values, and plan a convenient and healthy route based on the overall concentration value is to avoid direct inhalation of a large amount of harmful gases when the existing PM2.5 value is too high , a problem that should be considered in the health travel with the least degree of injury.

At present, in domestic and foreign research, path planning is mainly based on the optimal path planning based on traffic information according to user needs. The criteria for the optimal path include: the lowest cost, road section avoidance, and passing through specific areas. The navigation process from the start point to the end point at the cost of the lowest cost, distance, time or cost, does not cover the health needs; in terms of dealing with air pollution, it is mainly the people who wear masks spontaneously or go out less when the PM2.5 value exceeds the table. coping behaviors. Therefore, it is very necessary to invent a healthy travel navigation system and route planning method for PM2.5 value.

Contents of the invention

The purpose of the present invention is to put the PM2.5 value into big data, provide a new navigation method for travel, and make the travel process more convenient and healthy.

The technical solution that realizes the object of the present invention is:

A path planning device based on PM2.5 healthy travel, including three parts: vehicle-mounted PM2.5 detection equipment, information transmission module and server terminal, wherein the information transmission module includes a GPS module and GPRS module, and the server terminal includes a path planning module. The PM2.5 detection equipment is installed on the floating vehicle and connected to the information transmission module, which is connected to the server terminal; the PM2.5 detection equipment can directly collect the PM2.5 value of the path passed by the floating vehicle, and the GPS module can obtain the geographical location of the floating vehicle For location information, the GPRS module will obtain the PM2.5 value and geographic location information and transmit them to the path planning module of the server terminal, and the path planning module realizes the final path planning. The information transmission module transmits the detected PM2.5 value, geographic coordinates and detection time; the geographic coordinates record the passing position of the floating vehicle that detects PM2.5, using standard latitude and longitude coordinates; the path planning module uses the improved Dijkstra algorithm to plan healthy travel routes , based on the original road network structure to locate the start and end points, form a weighted network topology map according to the PM2.5 detection value and simplify it into a directed weighted connected graph, and traverse nodes and regions with the goal of optimal PM2.5 concentration Narrow down according to the search restriction conditions. When the search area satisfies the rectangular restriction search conditions, the search work is carried out in units of nodes. When the distance is greater than a certain value R, the search level can be increased until reaching the highest level of the road network; when the search area When the search conditions of the rectangular limit are not satisfied, change the search area to a rectangular search, and repeat the shortest path search process with the goal of the minimum PM2.5 concentration value, and gradually improve the planned path as the search level increases; among them, the vehicle-mounted PM2.5 The PM2.5 detection values obtained by the detection equipment at the same place at different times need to be recorded repeatedly.

The path planning method based on PM2.5 healthy travel, the specific process is:

Step 1: Determine the sample size of floating cars in a single section

During the statistical period T _p , the estimated average PM2.5 value is for:

In the formula, P _i is the PM2.5 value of the i-th floating car; i is the serial number of the floating car; N _p is the total number of floating cars passing the road section within the statistical time period.

The distribution frequency of the PM2.5 daily average concentration measured by the floating car is close to the logarithmic normal distribution, so that M is the logarithm of the PM2.5 concentration value P, that is, M=In(P), and M is standardized to is the upper α quantile point of the standard distribution, then the calculation model for the number of floating cars on a single road section is:

In the formula, is the upper α quantile; σ _p is the standard deviation; ε _p is the allowable concentration error value.

Considering the influence of line length on the number of floating cars, _Np is revised to get:

In the formula, l is the length of the target section; T _p is the statistical time.

Step 2: Determine the sample size of floating vehicles in the road network

Since the floating car route is not fixed, in order to ensure the target accuracy, the traffic flow density can be used to describe the frequency. The probability P _i of the floating car appearing on any road is:

In the formula, a is the number of road grades, which is a constant _; N _i is the total number of roads of type _{i ;}

The probability P _i,j of the floating car appearing on the jth section r _i,j of the type i road is:

Considering the stop time and the number of errors of the floating car itself, the revised P _i,j is:

P′ _i,j ＝P _i,j (1-P _s )(1-P _c )

In the formula, P _s is the suspension rate of the floating car; P _c is the error rate of the floating car.

The calculation model for the number of floating cars under the entire road network is:

In the formula, N _z,i is the number of floating vehicles required for the i-type road, which can be obtained by step 1; Z _i is the influencing factor of different types of roads, and the calculation method is as follows:

Step 3: Route planning according to the sequence of PM2.5 concentration values

After the start point (x ₁ , y ₁ ) and end point (x ₂ , y ₂ ) are determined, establish an elliptical area to limit the search:

In the formula,

Calculate the partial derivatives for x and y, and you can get the extreme values x _min , x _max , y _min , y _max of x and y, and the pole coordinates (x _max , y _max ), (x _min ,y _min ), (x _max , y _min ), (x _min ,y _max ) four points constitute a rectangular area to limit the search.

in,

The improved Dijkstra algorithm based on PM2.5 concentration value path planning is applied in the restricted search rectangular area, that is, the environment optimal path objective optimization function:

In the formula, g _ij (t) is the PM2.5 concentration value sequence from node i to j at time t; f _i (t) is the optimal PM2.5 concentration value sequence from i to end point at time t, and N is the end point node.

Compared with the prior art, the present invention has the remarkable advantages of:

(1) Obtain PM2.5 concentration value, health path planning

In terms of route planning, most of the existing navigation systems search for the route with the shortest time, the shortest distance or the shortest cost, and there are few travel route design schemes that consider the health status of travelers. The present invention uses the vehicle-mounted PM2.5 detection equipment installed on the floating car to detect the air quality of each road section on the road network, and forms a network topology map with PM2.5 as the weight according to the PM2.5 concentration values obtained by each floating car. Thereby providing a healthy travel route for travelers.

(2) Hierarchical road network, efficient path planning

In order to pursue the best route, the current path planning methods often have complex algorithms, which greatly affects the efficiency of the entire path planning process. The present invention proposes the idea of road network layering in the aspect of path planning algorithm, adopts the improved version of Dijkstra's algorithm of region-limited search to combine preprocessing and layered search, and converts the original road network plane into a multi-layer road network. Continuous conversion search, this method greatly saves the search space and improves the efficiency of line search.

(3) Real-time data processing, dynamic path planning

Most of the current path planning algorithms are based on the shortest path algorithm based on fixed road edge weights, that is, the static shortest path algorithm. Due to the large scale of the traffic road network, the efficiency of the static path solving process is low, and the service response speed is slow. The invention utilizes the GPS module and the GPRS module for real-time data acquisition and transmission, realizes dynamic path planning, and improves the efficiency and accuracy of the entire path planning process.

(4) Combined with user needs, multi-path planning

Compared with other patents and existing navigation equipment, the present invention adds the function of "healthy travel" route planning on the basis of navigation. If healthy travel is combined with factors such as time, distance, and cost, it can achieve optimal comprehensive factors. path planning.

Description of drawings

Fig. 1 is a flow chart of the path planning algorithm of the present invention.

Fig. 2 is a schematic diagram of a matrix area for a restricted search.

Fig. 3 is a correlation diagram of various components of the present invention.

Detailed ways

The present invention will be further described below in conjunction with accompanying drawing and specific embodiment:

As shown in Figure 1, the route planning device based on PM2.5 healthy travel of the present invention obtains the PM2.5 concentration value through the vehicle-mounted PM2.5 detection equipment and establishes a traffic database in combination with traffic flow parameters. When the user enters the starting and ending information for navigation, the traffic data is imported into the GIS database, and the improved Dijkstra algorithm is used for line matching, and the matched line is transmitted to the web server through the cloud, and finally displayed on the user terminal computer or mobile phone .

As shown in Figure 2, the path planning method based on PM2.5 healthy travel in the present invention is an improved Dijkstra algorithm. After the user inputs the start and end points, the network topology map is loaded, and the complex topology map is hierarchically operated, that is, the road network Carry out hierarchical division to obtain multiple road network levels with different densities and number of road sections, and simplify the data structure of complex road networks based on different levels of sparseness. The specific operation steps are as follows:

Step 1: Split the road network into a different levels of road network according to the road level.

Step 2: Initialize the road network to simplify the complex road network. Use the data structure (S, T, L, P) to record the road network information of two nodes, where S is node 1, T is node 2, L is the shortest path between the two nodes, and P is the shortest between the two nodes under the optimal concentration of PM2.5 road.

Step 3: Traversing the road network nodes, if L is less than a fixed threshold (X), two nodes are similar by default, one of them is eliminated, and a virtual road section is generated to connect two nodes close to the eliminated point, and a new simplified structure of the road network is generated by updating, virtual The edge creation process is shown in Figure 2 below.

Step 4: Aiming at the same node, merge the split road networks of a different levels.

According to the positions of the starting point and the ending point in the topological map, an adjacency matrix M is established, and the adjacency matrix M is reordered according to the edge weight (PM2.5 concentration value of each road section) to construct an adjacency table T. If T satisfies the rectangular limit search condition Lmax, that is L≤Lmax, then directly use the Dijkstra algorithm to obtain the healthy travel path with the smallest PM2.5 concentration value. The specific Dijkstra algorithm is:

In the formula, g _ij (t) is the PM2.5 concentration value sequence from node i to j at time t; f _i (t) is the optimal PM2.5 concentration value sequence from i to the end point at time t.

If L<Lmax, first construct the ellipse equation with the starting point (x ₁ , y ₁ ) and end point (x ₂ , y ₂ ):

In the formula,

Then calculate the partial derivative for x and y, and obtain the restricted search rectangular area with pole coordinates (x _max , y _max ), (x _min , y _min ), (x _max , y _min ), (x _min , y _max ), As shown in Figure 3.

in,

Finally, in the rectangular area, the optimal path is calculated using the improved Dijkstra algorithm based on PM2.5 concentration value path planning, namely:

In the formula, g _ij (t) is the PM2.5 concentration value sequence from node i to j at time t; f _i (t) is the optimal PM2.5 concentration value sequence from i to end point at time t, and N is the end point node.

Claims (1)

1. A path planning method based on PM2.5 healthy trips is characterized by comprising the following specific steps:

the method comprises the following steps: determining single-section floating car sample size

Statistical time period T_pInner, mean PM2.5 estimate

In the formula, P_iPM2.5 value of the ith floating car; i is the floating car serial number; n is a radical of_pCounting the total number of floating cars passing through a road section in a time period;

let M be the logarithm of the PM2.5 concentration value P, i.e., M ═ In (P), and M is normalized tothe calculation model of the number of the floating cars on the single road section is as follows:

in the formula,is the upper α quantile and sigma_pIs the standard deviation; epsilon_pTo allow for concentration error values;

to N_pThe revision is carried out as follows:

in the formula, l is the length of the target road section; t is_pCounting time;

step two: determining road network floating car sample volume

Probability P of floating car appearing on any road_iComprises the following steps:

in the formula, a is the number of road grades and is a constant; n is a radical of_iThe total number of the roads is type i; rho_iI road traffic flow density; l_i,jIs the jth section r of the type i road_i,jLength of (d);

the floating car is on the j section r of the type i road_i,jProbability of occurrence P_i,jComprises the following steps:

revision P_i,jComprises the following steps:

P′_i,j＝P_i,j(1-P_s)(1-P_c)

in the formula, P_sThe floating car stop rate is obtained; p_cIn order to achieve the error rate of the floating car,

the number calculation model of the floating cars under the whole road network is as follows:

in the formula, N_z,iThe number of floating cars required by the i-shaped road can be obtained by the first step; z_iFor the influence factors of different types of roads, the calculation method comprises the following steps:

step three: planning according to PM2.5 concentration value sequence path

At the starting point (x)₁,y₁) Endpoint (x)₂,y₂) After determination, a rectangular area for limiting search is established:

in the formula,

partial derivatives are calculated for x and y to obtain extreme values x of x and y_min、x_max、y_min、y_maxPolar coordinate (x)_max,y_max)、(x_min,y_min)、(x_max,y_min)、(x_min,y_max) Four points form a rectangular area for limiting search;

wherein,

traversing network topology nodes by taking PM2.5 optimal as a target in a limited search rectangular area to obtain an optimal path, namely constructing an environment optimal path target optimization function:

in the formula, g_ij(t) the value of PM2.5 concentration from node i to j at time t; f. of_i(t) is the optimal PM2.5 concentration value from i to the end point at time t, and N is the node at the end point.