CN114550482B - Navigation method based on low-carbon target and parking lot navigation method - Google Patents

Abstract

The invention provides a navigation method based on a low-carbon target and a parking lot navigation method, which are carried out according to the following steps: step by stepStep 1: determining a starting point o and an end point d; step 2: obtain a set K, K= { K of all feasible paths among od in the whole city ₁ ，k ₂ ，k ₃ ，…，k _m Defining a road sections in all feasible paths among the od to obtain a road section set A= {1,2,3 … a }; solving MinZ (X) according to a road network carbon emission optimization principle, and calculating carbon emission of each path in the set K to obtain an optimal path K 'when the carbon emission is minimum, and navigating the vehicle to a destination according to the optimal path K'; according to the invention, the real-time road condition destination dynamic parking lot planning is used as a guided travel path optimization scheme, so that the optimization of the carbon dioxide emission of the whole road traffic network vehicle is realized, and the distance or time cost efficiency discussed by the traditional method is not optimal. Therefore, a new method is provided for realizing the goal of blocking and emission reduction in the urban infrastructure planning technology.

Description

A navigation method based on low-carbon goals and a parking lot navigation method

technical field

The invention belongs to the technical field of urban building planning, and in particular relates to a navigation method based on low-carbon targets and a parking lot navigation method.

Background technique

"Parking difficulties" and "normalization of congestion" have become important reasons that plague urban residents' travel and increase traffic carbon emissions. However, recent studies have shown that parking lots, as the starting and ending points of private car travel, can reduce congestion and reduce emissions by affecting the choice of driving routes. However, there are a series of problems in the practical application of this theory. First of all, the driving route planning with the parking lot as the destination lacks dynamic linkage with road conditions and parking lot usage information, and it is impossible to dynamically control the traffic flow distribution of the road network through parking navigation. Secondly, the current navigation system only recommends the driving route to the user based on the optimal time or distance. However, there is no linear relationship between the carbon emissions of vehicles and the driving time or distance. The individual driving route planning with the shortest time or distance cannot guarantee the lowest carbon emissions of vehicles in the road network as a whole. More importantly, the lack of linkage between dynamic parking information and driving planning causes drivers to often find that there are no vacant parking spaces when they arrive at their destination. Congestion carbon emissions. According to prior art records, in a busy traffic area, vehicles spend 3.5 to 14 minutes looking for a parking space, causing traffic flow to increase by 20-45%, causing serious congestion and high carbon emission problems. On a road with a traffic capacity of 1200veh/h, 18 minutes of illegal on-street parking will increase vehicle carbon emissions by 26.5%. Therefore, how to quickly and effectively provide drivers with parking lot navigation solutions that are beneficial to urban carbon emission reduction is an important issue in the field of low-carbon urban infrastructure planning technology.

With the advent of the 5G communication era, parking decisions and traffic guidance systems based on Internet big data have become a development trend. When drivers use mobile phones and vehicle-mounted devices for navigation, the vehicle becomes a terminal that generates, transmits, and receives data and executes navigation results. The dynamic data flow generated by car driving realizes two-way communication and calculation in the cloud through the Internet of Vehicles technology, which establishes the innovation of low-carbon route planning in the form of parking navigation.

Contents of the invention

In view of the deficiencies and defects of the above-mentioned prior art, the purpose of the present invention is to provide a navigation method based on low-carbon targets and a parking lot navigation method to solve the lack of a linkage between real-time road conditions and overall road network carbon emissions in the prior art. The problem of parking location and route guidance method.

In order to solve the above technical problems, the present invention adopts the following technical solutions to achieve:

A navigation method based on low-carbon goals, which is carried out according to the following steps:

Step 1: Determine the starting point o and the ending point d;

Step 2: Obtain the set K of all feasible paths k ₁ , k ₂ , k ₃ , ..., k _m between od in the whole city, K={k ₁ , k ₂ , k ₃ , ..., k _m }, define od All feasible paths between contain a road section, and the road section set A={1, 2, 3...a} is obtained;

According to the optimal principle of road network carbon emissions, solve MinZ(X), calculate the carbon emissions of each path in the set K, obtain the carbon emissions of each path in the set K, and obtain the minimum carbon emissions The optimal path k', k, k'∈K, the vehicle navigates to the destination according to the optimal path k';

The optimal principle of road network carbon emissions is calculated as follows:

X is the carbon emission of the road network;

Z is the overall carbon emission level of the traffic road network;

k represents the kth path in the set K;

a is the section number;

L _a is the length of road section a, the unit is m;

is the free flow time of road section a, the unit is s, that is, the zero flow impedance of road section a;

c _a is the traffic capacity of road section a, the unit is pcu/h;

α and β are retardation coefficients, which are 0.15 and 4 respectively;

is the traffic on the k-th path between the od points whose starting point is o and the ending point is d, and the unit is pcu;

It is a road segment-path variable, which is a 0 or 1 variable; if the road segment a belongs to the kth path between the od points whose starting point is o and the ending point is d, then /> otherwise />

is the design speed of road section a, in m/s;

f is a sixth-order function subject to speed-carbon emission rate conversion.

A parking lot navigation method based on a low-carbon target, comprising the following steps:

Step ①: Determine the starting point o and the ending point d;

Step ②: Based on the open source map platform, obtain real-time road condition information and parking facility usage information at the current moment, including the remaining number of parking spaces, average parking duration, and parking turnover rate; estimate the arrival time S according to the real-time road condition information, and determine the time to arrive at the destination d Whether there are free parking spaces in the construction parking lot, if there are vacant parking spaces in the equipped parking lot when reaching the end point d, execute step ③, if there is no free parking space in the equipped parking lot when reaching the end point d or there is no equipped parking lot in the end point d, execute Step ④,

For example: at 10:00 a.m., according to the real-time traffic information provided by the Baidu map platform (Table 2), the estimated arrival time is 10:07 a.m. According to the usage information of the auxiliary parking facilities, the hourly utilization rate of the parking spaces is 0.8. The auxiliary parking lot has 1000 parking spaces. At this time, the remaining parking spaces are 1000*(1-0.8)=200, and the hourly parking space turnover rate is 1. It is estimated that After 7 minutes, the vacant parking spaces in the equipped parking lot are: 200-1*(7/60)=200. It can be seen that d has a vacant parking lot when it arrives. Calculate according to formula (1) to get the optimal path, and navigate the vehicle to the parking lot d equipped according to the path planning scheme.

Step ③: Carry out path planning according to the optimal principle of road network carbon emissions, and navigate the vehicle to the end point d to build a parking lot according to the path planning scheme;

The optimal principle of road network carbon emissions is calculated as follows:

X is the carbon emission of the road network;

Z is the overall carbon emission level of the traffic road network;

b represents the bth path in the set B;

p is the section number;

L _p is the length of road section p, the unit is m;

is the free flow time of road segment p, the unit is s;

c _p is the traffic capacity of section p, the unit is pcu/h;

α and β are retardation coefficients, which are 0.15 and 4 respectively;

is the traffic on the bth path between the points od with the starting point o and the ending point d, the unit is pcu;

It is a link-path variable, which is a variable of 0 or 1; if the link p belongs to the bth path between the od points whose starting point is o and the ending point is d, then /> otherwise />

f is a sixth-order function subject to speed-carbon emission rate conversion.

If there is no parking space, go to step 4;

When the destination is a parking lot, the calculation process of the navigation path does not need to re-divide the road network structure, so the solution process will not affect the driving paths of other vehicles, and the path with the smallest carbon emission of this vehicle represents the smallest carbon emission of the road network calculation results. Since the carbon emission is a sixth-order function of the vehicle speed, the vehicle driving path is determined by the road flow. When building a carbon emission model, establish a relationship between vehicle speed and flow (Da), and express the flow with vehicle speed to establish a carbon emission model for solution.

The derivation process of formula one is as follows:

Analyze the objective formula (I) of the second principle of wardrop (system equilibrium).

In the formula, D _a is the flow rate of road section a (pcu/h);

t _a is the road resistance, which can be understood as the actual time (s) of driving through road section a;

t _a (D _a ) is a BPR function with the flow rate of road section a as an independent variable;

According to the second principle of wardrop, the solution of formula (I) is actually an optimization problem of the objective function of road flow Da. Since the relationship between carbon emissions and vehicle speed is a sixth-order function, the flow rate is expressed by vehicle speed, and a carbon emission model is established to solve it. There are mainly the following steps:

(1) Determine the relationship between travel time and actual flow, see formula (II):

In the formula, c _a is the traffic capacity of section a (pcu/h);

is the zero-flow impedance, that is, the time (s) for the vehicle to move freely through the road section when the road is empty, which can be considered as the quotient of the length of the road section and the design speed;

α and β are the retardation coefficients, which are 0.15 and 4 respectively in the allocation procedure of the US Federal Highway Administration.

(2) Determine the relationship between driving time and average vehicle speed:

In the formula, L _a represents the length (m) of road section a.

(3) Determine the relationship between vehicle speed and carbon emission rate:

Q _a =w(V _a )(V)

In the formula, Q _a represents the carbon emission per unit kilometer of a vehicle in road section a,

Obey the sixth-order function of speed-carbon emission rate conversion, see formula VI:

(3) Carbon emission target formula

The objective function is the minimum carbon emission of the system, and the objective formula is established according to the second balance principle: system optimal principle:

in, A sixth-order function that obeys the speed-to-carbon emission rate conversion.

(4) Construct the fitness function as:

in,

In the formula, X is the carbon emission of the road network; Z represents the overall carbon emission level of the traffic road network; D _a is the traffic flow of road section a; L _{a is} the length of road section _a ; The time spent in a; C _a is the road capacity of section a; is the zero-flow impedance of the road, which is taken as the free flow time passing through section a; α and β are the retardation coefficients, which are taken as 0.15 and 4 respectively; is the traffic on the kth path between the od point pair whose starting point is o and the ending point is d, the unit is pcu; /> It is taken for the link-path variable, when the link a is on the path k, it takes 1, otherwise it takes 0; /> is the design speed of road section a, the unit is m/s; send a request to the map platform every time t, and the calculation result update frequency is T+nt.

Therefore, formula (VII) is derived as (1):

Step ④: Obtain real-time traffic information and public parking lot usage information at the current moment, and predict whether there is an available public parking lot within 300 meters around when arriving at destination d;

If not, recommend the user to change the destination;

If so, obtain N public parking lots with vacant parking spaces in the city at the current moment, divide the Thiessen polygon with the centroids (N _x , N _y ) of these N public parking lots as feature points, and obtain the set of parking zones O= {1, 2,...i...j...N}(1≤i≤N, 1≤j≤N, i≠j), i is the Thiessen polygon partition starting from (i _x , i _y ), j is Thiessen polygon partition with (j _x , j _y ) as the end point;

Select N'(2≤N'≤N) parking lots in N public parking lots to form public parking lot combination alternatives {1, 2}; {1, 2}, {1, 3}, {2 , 3}, {1, 2, 3}; ...; {1, 2}, {1, 3} ... {1, N'}, {1, 2, 3} ... {1, 2, N'} ... {1, 2, 3...N'-1, N'} Total , divide the urban road network with the boundaries of the alternative public parking lot combination alternatives, and obtain the road section set R={1, 2, 3...r};

According to the optimal principle of road network carbon emissions, MinZ(X) is solved, and the carbon emissions of the urban road network divided by the combined alternatives of all public parking lots are calculated to obtain the optimal driving path S' of the vehicle;

According to the formula (2), program on the Matlab platform according to step 4, realize the dynamic solution of MinZ(X) through the single-parent genetic algorithm, calculate the public parking lot combination scheme M' under the condition of the minimum carbon emission of the overall road network, and obtain the starting point belonging to Tyson Polygon i', Thiessen polygon j' to which the end point belongs, and the vehicle calculated by the combined scheme M' navigates to d with the optimal driving path S'.

The specific formula is as follows:

X is the carbon emission of the road network;

Z is the overall carbon emission level of the traffic road network;

r is the section number;

ξ is the coefficient of the gravity model;

L _r is the length of road section r, the unit is m;

F is the proportion of travel by self-driving car in the Thiessen polygon partition;

D _rk is calculated according to the gravity model to obtain the flow of inter-cell traffic allocated to road section r; D _pr is the number of public parking lots (capacity) whose exit is on road section r;

D _er is the number of parking lots (saturation) equipped with the building equipped with the exit on the r-section road;

E _er is the berth turnover rate of the building's equipped parking lot in peak hours;

U _pr is the utilization rate of parking spaces in peak hours of public parking lots;

U _er is the peak hour berth utilization rate of the building's equipped parking lot;

E _pr is the turnover rate of berths in peak hours of public parking lots;

P _j represents the traffic volume generated by the Thiessen polygon partition starting from i;

B _j represents the traffic attraction generated by the Thiessen polygon partition with j as the end point;

t _r (D _r ) is the traffic impedance from the Thiessen polygon partition starting from i to the Thiessen polygon partition starting from j; take the shortest time estimated from i to j according to the current road conditions;

D _r is the traffic flow on road section r, the unit is pcu/h;

is the free flow time of road section r, the unit is s, that is, the zero flow impedance of road section r;

c _r is the traffic capacity of section r, the unit is pcu/h;

g is the number of jj pair, G is the number of jj pair;

S is the number of feasible paths, and H is the number of paths passing through r road segments in the feasible path;

e is the base of natural logarithm;

μ is the expected value, which is the shortest time most vehicles spend from i to j;

θ is the traffic conversion parameter, θ=3～3.5;

is any feasible path /> total travel time;

α and β are retardation coefficients, and α=0.15 and β=4 are respectively taken according to empirical values;

f is a sixth-order function subject to speed-carbon emission rate conversion.

Because the discussion at this time is the navigation of the public parking lot, the default parking lot is saturated

The derivation process of Formula 3 is as follows:

According to the second principle of wardrop, in the case of system equilibrium, the road network traffic flow should be allocated based on the average or minimum overall travel cost. Therefore, when navigating to a public parking lot for parking, the fitness function is constructed as:

Among them, r is the road segment number re-divided according to the Thiessen polygon boundary; D _r is the traffic flow on road segment r (pcu/h); L _r is the length of road segment r; Q _r is the carbon per unit kilometer of road segment r Emissions as a function of average velocity:

Q _r ＝f(V _r ) (XI)

The average velocity V _r can be expressed as:

Among them, L _r is the length of the road r; t _r is the time it takes for the vehicle to pass the road under the condition of D _r flow rate, t _r (D _r ) is the impedance function of the road segment r with the flow rate (D _r ) as the independent variable, and also Call the travel time function:

Formula (X) can be further expressed as:

Among them: C _r is the traffic capacity of the road section of the r section, the unit is (pcu/h), t ₀ is the zero-flow impedance of the road, that is, the time for the vehicle to drive freely through the road section under the state of the road is empty, and the unit is (s), which can be It is considered that t ₀ is the quotient of road length and road design speed, in the above formula, α, β are retardation coefficients, Indicates the traffic flow from i to j on road section r; D _rk is the total flow from i to j; D _r is the actual flow of road section r; C _r is the road capacity.

Since the Thiessen polygonal partition divides the road network into r segments, the actual flow of r segment is equal to the flow allocated to r segment minus the flow entering r segment and the public parking lot. According to the principle of proximity, it is considered that the amount of parking attraction generated by the community is firstly solved in the parking lot of the building, and the excess part is solved by the public parking lot. The formula (XIV) defaults that the parking lot of the road section is saturated.

therefore:

D _r ＝FD _rk -D _pr U _pr E _pr -D _er U _er E _er (XIV)

In the formula,

F is the proportion of travel by self-driving car in the Thiessen polygon partition;

D _r is the actual traffic flow of section r;

D _rk is calculated according to the gravity model to obtain the traffic flow between the sub-districts allocated to the r section;

D _rr is the parking number (capacity) of the public parking lot on the road where the exit is on section r;

D _er is the number of parking lots (saturation) equipped with the building equipped with the exit on the r-section road;

E _er is the berth turnover rate of the building's equipped parking lot in peak hours;

U _pr is the utilization rate of parking spaces in peak hours of public parking lots;

U _er is the peak hour berth utilization rate of the building's equipped parking lot;

E _pr is the turnover rate of berths in peak hours of public parking lots;

Therefore, the optimization problem of carbon emissions is transformed into the distribution problem of traffic flow among traffic areas.

According to the gravity model there are:

In the formula, D _ij is the traffic volume from cell i to cell j (i≠j), and the centroid of the Thiessen polygon where cell i and j are located is the feature point;

ξ is the gravity model parameter;

P _i is the traffic trip volume generated by i cell;

B _j is the traffic attraction generated by j cell;

t _r (D _r ) is the traffic impedance from the Thiessen polygon partition starting from i to the Thiessen polygon partition starting from j; take the shortest time from i to j.

D _r is the traffic flow on road section r (pcu/h)

P _i =∑ζ _h M _h (XVI)

In the formula:

h is the land use type of the plot;

ζ _h is the travel generation rate of type h land use;

M _h is the construction area of the h-type land.

The algorithm of B _j is similar to that of P _i ,

B _j ＝∑b _h M _h (XVII)

In the formula,

b _h is the traffic attraction coefficient of type h land use;

G is the number of ij pairs;

g is the ij pair number;

w _gr is the ratio of the gth ij pair (D _ij ) _g allocated to the road section r;

Take an od pair, assuming that there are S feasible paths from cell i to cell j, and among them, H paths pass through the Kth road segment, then choose the th road segment when traveling from cell i to cell j The probability of a path is:

u=t _r (D _r ) is the minimum traffic impedance from cell i to cell j, taking the shortest time from i to j.

In the formula, to select the /> probability of a path;

is the total travel time for each route;

θ is the traffic conversion parameter, taking 3.0~3.5.

Assume that among the feasible paths from i to j, the road with smaller impedance has a higher probability of being selected as the optimal path, and the distribution number is the largest. Therefore, the probability of choosing the optimal path conforms to the normal distribution curve, in which there are H paths that pass through r road sections.

Then the total ratio of the g-th ij pair allocated to the r segment is:

The steps to determine the feasible paths S and H are:

Step1 Determine the effective road section: if the road section node number is aa, bb, calculate the shortest travel time T(aa), T(bb) from the road section node to the starting point, T(aa)>T(bb) aa to bb is a valid road section;

step2 is composed of effective road sections to form an effective path;

In the step3 allocation, the number of times that the aa-bb road segment is retrieved to have traffic flow allocated to this road segment is H.

Substitute D _rk , w _gr , P _s into formula (XIV) to find Min Z(X),

In case of emergencies in step 4:

① If a traffic accident occurs along the optimal route b', return to step 2 to recalculate and judge.

If there are vacant parking spaces in the parking lot of the destination d, the path planning is carried out according to the formula (2), and the optimal path is obtained until the navigation ends at the end point.

If there is no available parking space in the equipped parking lot, step 4 is performed based on the road conditions at T+nt, and the vehicle driving path s _T+nt is navigated to d under the condition of minimum carbon emissions of the overall road network.

②If the public parking lot where the destination d is located is saturated (100% utilization rate), determine whether the current driving position of the vehicle is within 300 meters of the destination d;

If not, the Thiessen polygon grid is re-divided according to the centroid of the public parking lot with vacant parking spaces at T+nt as the feature points. Recalculate the location of the public parking lot and the optimal path s' _T+nt according to formula (3);

If it is, then navigate to the public parking lot where the road section where the vehicle is located is located in the Thiessen polygon division when T+nt is located;

Until the navigation to the destination, the end.

Since Equation (2) involves the dynamic optimization of road traffic induction, and there are many variable parameters, the genetic algorithm is used to solve it, and the centroid coordinates of the public parking lot location d' within 300 meters of the end point d are obtained when the overall carbon emissions of the road network are optimal. (xd', yd') and navigation path.

Since a request is sent to the map platform every time t during navigation, the update frequency of the formula calculation result is T+nt. Instructions can be sent to the map platform at intervals t to reacquire real-time road condition information and parking facility usage information at T+nt time, then Equation 1 and Equation 2 can be:

Compared with the prior art, the present invention has the following technical effects:

(I) The present invention is based on the destination dynamic parking lot (position) planning of the real-time road conditions as the oriented travel route optimization scheme, and realizes the optimization of the vehicle carbon dioxide emissions of the entire road traffic network, rather than the distance or time cost efficiency discussed in traditional methods best. Therefore, it provides a new method for urban infrastructure planning technology to achieve the goal of slowing congestion and reducing emissions.

(II) The present invention's dynamic public parking lot partition method based on Thiessen polygons has been developed and actually realized in an open source map navigation system. Utilize the geometric characteristics of Thiessen polygonal partitions, and the feedback from the open source data platform and dynamic traffic network system to achieve dynamic optimization, respond to urban road traffic emergencies, quickly select parking lots, simplify calculation steps to improve road traffic induction efficiency and carbon emissions Good practices for reducing emissions. This zoning method provides an accurate method for traffic allocation and parking demand estimation, which is not quadrangular or radial zoning, but more concerned with the time-varying characteristics of parking spaces and their impact on the congestion carbon emissions caused by dynamic traffic routing. Influence. More importantly, the application of the Thiessen polygon nearest neighbor principle to the optimization method of parking location partition, combined with the function can achieve the common goal of the shortest distance. For complex path optimization research, real-time path planning and application of navigation systems, it is very important to improve the efficiency of calculation and feedback for real-time navigation and path planning systems.

(Ⅲ) The present invention uses genetic algorithm (GA) to find the optimal public parking lot set. GA has great application value in speeding up the global optimization of random search. It is especially suitable for simulating complex and high-volume problems involving realistic solutions. The most important point is that this method provides a practical and applicable method for the regulation of static traffic and dynamic traffic, and proves that static traffic has a strong ability to influence dynamic traffic. Especially in terms of congestion control, this dynamic partitioning approach can be explored for a larger range of real-time. With real-time traffic information provided by ITS's roadside sensors, dynamic zoning-based parking guidance will have the potential to improve traffic system performance with the support of smart city infrastructure. By applying dynamic partitioning and modeling methods to intelligent transportation systems, the efficiency of parking induction and the dynamic optimization of traffic allocation for future smart mobility can be ensured.

Description of drawings

Figure 1 is a schematic diagram of the location of the example base;

Figure 2 shows the status quo of road land use in the example base of the built-up area;

Figure 3 is a schematic diagram of the current situation of the base buildings;

Figure 4 is a schematic diagram of the base parking lot;

Fig. 5 is the public parking lot planning zoning diagram based on Thiessen polygon;

Fig. 6 is that genetic algorithm carries out dynamic solution to formula (3) and obtains the optimization process of total amount of carbon emissions;

Figure 6.1 is the Matlab model calculation interface;

Figure 6.2 is the iteratively generated 5 groups of road segment flow matrices;

Figure 6.3 allocates traffic for 48 road segments in each matrix;

Figure 7 shows the most suitable scheme for the public parking lot under the condition that the overall carbon emission of the road network is the smallest;

Fig. 8 is a flow chart of genetic algorithm;

Figure 9 is the real-time road condition information of the road obtained from open source map data mining.

Figure 10 Schematic diagram of the path to navigate to the equipped parking lot where the destination is located

Figure 11 Schematic diagram of the route to navigate to the public parking lot where the destination is located

Figure 12 Obtaining the parking route planning process under the optimal condition of road network carbon emissions

Figure 13 Parking site selection and route planning method based on dynamic open source map information

The specific content of the present invention will be further explained in detail below in conjunction with the examples.

Detailed ways

Specific embodiments of the present invention are provided below, and it should be noted that the present invention is not limited to the following specific embodiments, and all equivalent transformations done on the basis of the technical solutions of the present application all fall within the protection scope of the present invention.

General idea of the present invention is as follows:

Navigate to the built-in parking lot:

The real-time road condition information at time T provided by the open source map platform, the estimated arrival time of parking facility usage information, and whether there is a parking lot and available parking spaces at the terminal d.

If there is a parking lot and available parking spaces, proceed from the starting point o to the end point d along the section between the forks and intersections of the road network. When passing through each fork, judge the feasible section of the forward direction. For a feasible path k, according to formula 1, calculate the driving path k' when the carbon emissions of all feasible paths are the smallest, and navigate to the destination parking lot along the optimal path k'.

Navigate to public parking:

If there is no parking lot, determine whether there is a public parking lot within 300 meters of the end point d.

If there are public parking lots and available parking spaces, obtain all N public parking lots with vacant parking spaces in the city at time T, and select N'(2≤N'≤N) public parking lots from N to form Combination alternative schemes of public parking lots, construct a Thiessen polygon partition with the centroid of each public parking lot in the alternative scheme as the feature point, calculate the traffic trip volume and traffic attraction generated by the covered buildings in the partition, construct the centroid connecting rod, and connect The Thiessen polygon centroid to the nearest road intersection distributes the traffic volume generated by the buildings in the partition to the road network through the centroid link. According to formula 2, use the genetic algorithm to dynamically optimize and solve the formula, calculate the public parking lot combination scheme when the overall carbon emission of the road network is the smallest in the alternative scheme, and obtain the Thiessen polygon i' where the starting point o belongs to and the Thiessen polygon i' where the end point d belongs to in the combined scheme Polygon j', and the distribution results of vehicles in each road section under the optimal parking lot combination scheme, the driving path from i' to j' is obtained according to the distribution results of vehicles in each road section, and the driving path is used to navigate to the public parking lot j' where d is located Then walk to the destination d.

Generally, there are several specific methods for genetic algorithms to solve multi-objective optimization problems, such as weight coefficient change methods, parallel selection methods, layout selection methods, shared function methods, and hybrid methods. To alleviate the programming complexity and simultaneously recombine to meet traffic and parking demands, a parallel selection approach with single-factor inheritance was used in this study. The specific programming process is as follows:

1) Encoding operation

In this study, the gene (alternative public parking lot) is initialized and coded using the real number coding method, and the coding sequence is 1, 2, 3, 4, m.

2) Generate initial population

Each chromosome (possible combination scheme for a public parking lot) is made up of genes. The set of randomly selected alternative public parking lots constitutes the initial gene. For example, if there are 10 alternative public parking lots within 300 meters of the end point D, numbered 1, 2, 3...10, the chromosome can be defined as {1, 3, 7} or {1, 2, 9, 10}, etc. .

3) Define the fitness function

The fitness function (formula 3) is used to calculate the solution result of the genetic algorithm.

4) Population selection

The purpose of population selection (a set of possible combinations of public parking lots) is to select a better individual in the initial parking lot group as the parent of the next generation of genetic crosses. The criterion for judging whether an individual is excellent is the result of the fitness function of the individual solution set. The better the fitness value of an individual, the more likely it is to be selected by the next generation.

5) Recombination

Solutions in the mating pool will continually produce combinations that reduce the overall carbon footprint of the road network due to intergenerational replication effects. The fitness of the best individual in the group does not decrease due to replication, genetic processes that do not generate new candidate parking lots. The population used in the genetic recombination process is randomly selected from the mating pool. By selecting the parking lot groups with better personalities, the final generation will contain the best genetics from the paternal line.

6) Variation

For example, a gene with a current value of 0 means that the parking lot will not be selected in a genetic iteration. If a mutation occurs, the value will change to 1, which means that the chance can be selected for the next iteration. Mutations will be performed with equal probability for each public parking lot in the chromosome.

To sum up, as shown in Figure 8, the genetic algorithm is used to optimize the _CO2 emissions of the city's overall road traffic network, and the specific steps are as follows:

Step 1. Initialize. Set the population size to 8, the chromosome length to 7, the number of iterations to 5, the gene recombination mode (transposition, full mating, transposition, inversion) and mutation probability to 1/7.

Step 2. Apply binary integers to the encoding of the coordinates of each public parking lot and generate an initial population randomly.

Step 3. Compute the fitness function for each generation using an iterative method. The method is separated by three parallel parts, namely Thiessen polygon-based zoning, traffic allocation and emission calculation. Classify all individuals in the parent generation, select better individuals, and eliminate inferior individuals to generate a new combination of public parking lot location schemes.

Step 4. Perform crossover among individuals with random connections according to crossover probabilities. According to the probability of mutation, the combination of single parking lot is mutated.

Step 5. Confirm that the maximum number of iterations of 5 is reached and the calculation results no longer produce results with lower carbon emissions. If yes, then output the best solution for the location of the public parking lot. Otherwise, please return to step 3 for the next round of iterative calculation.

Example 1:

A parking lot navigation method with a low-carbon target is carried out according to the following steps:

Taking the Xincheng Square area of Xi'an as an example, calculate the parking navigation planning scheme under the condition that the destination is the optimal carbon emission of the road network in the Xincheng Square area. Optimizing the efficiency of dynamic traffic systems.

Enter East Street (o) as the starting point, and Xuji Seafood (d) as the ending point at Xincheng Plaza. Figure 2 is drawn according to the current land use map of Xi'an City and the built environment information of this area. Based on the open source map platform, obtain real-time traffic information and parking facility usage information at 8:00 a.m., and estimate the arrival time at the destination according to the real-time traffic conditions.

During the morning rush hour, most cars drive into public parking spaces, and relatively few vehicles drive out, which can be ignored. Therefore, without loss of generality, we directly use the oncoming traffic flow to derive the link net flow. According to the usage characteristics of parking spaces in the area where the example parking lot is located in Table 1, calculate the remaining number of parking spaces in the destination parking lot, the average parking duration, and the turnover rate of the parking spaces. It can be obtained that the utilization rate of parking spaces is 0.5, and the turnover rate is 3. The rate is 1 and the turnover rate is 1.

Table 1 Usage of parking spaces

At 10:00 a.m., according to the real-time traffic information (Table 3) provided by the Baidu map platform and the usage information of parking facilities, the estimated arrival time is 10:07 a.m. It is judged that there is a free parking lot when it arrives. Carry out path planning according to formula (1), and obtain 2 feasible paths between od in the whole city (k ₁ =11, 12, 6, 7; k ₂ =11, 13, 14, 15, 7), including 7 Sections (section numbers: 11, 13, 14, 12, 6, 16, 7), section information see Table 2. Among them, it can be obtained from the information in Table 1 Similarly available /> According to the optimal principle of road network carbon emissions, the carbon emissions of paths k1 and k2 in the set K are calculated according to formula 1, and the carbon emissions of each path in the set K are obtained, and MinZ(X) is solved to obtain Z( k ₁ )＝3.0388*10 ⁷ , Z(k ₂ )＝7.2437*10 ⁷ , therefore, MinZ(X)＝Z(k ₂ ), the optimal path for the minimum carbon emissions is path k ₁ The optimal path k ₁ (11, 12, 6, 7) scheme navigates to the parking lot d, as shown in Figure 10.

Table 2 section information

Table 3 Example of road speed data obtained by parsing open source maps

At 18:00 p.m., there is no available parking space in the parking lot of terminal d. Parking needs to be solved in public parking lots within 300 meters of the destination. According to the formula (2), the combination scheme of public parking lots under the optimal condition of road network carbon emission is calculated to realize the partition of the research area and the optimal flow distribution of each road section, and obtain the optimal path planning between od under the optimal condition.

Figure 3 shows the building information within this range, and the endpoint at the lower left corner of the road network is defined as the coordinate origin. The building material center point is regarded as a feature point to obtain building information. Based on the coordinate origin, mark the coordinates of each building centroid, and the parameters are shown in Table 4.

Table 4 Information of buildings and their attached parking spaces

Table 5 road information

Note: In the "Road Type" column of Table 3, A, L, and C represent arterial roads, secondary arterial roads, and urban branch roads, respectively; here, all roads listed here are two-way roads. )

Figure 4 shows the construction and public parking lots in this area, and Figure 5 shows the seven candidate public parking lots in this area. By selecting the best combination of 7 parking lots, it will affect the road network traffic flow distribution, so as to achieve the optimal traffic performance in the road network, that is, the optimization goal of minimizing carbon emissions. Obtain the optimal path between od under the optimal combination mode. The specific path planning calculation process is as follows:

In this case study, 7 separate Thiessen polygons were created as parking analysis areas based on the centroid locations of available public parking lots. The information of each polygon is represented by its centroid. Table 6 lists the road information shown in Figure 5.

Table 6 road endpoint information

Table 7 provides the basic information of these 7 public parking lots and their Thiessen polygon analysis area.

Table 7 Public parking lot and road parameters

As shown in Figure 4, links are created from the centroid of the Thiessen polygon analysis area to the nearest road node, and all traffic flows generated by buildings in the Thiessen polygon analysis area are distributed to the road network through the links. According to the generated connecting rod, measure its length, the coordinates of the center of mass of the Thiessen polygon and the coordinates of the road nodes, the parameters are shown in Table 7.

Table 8 Parameters of the center of mass connecting rod

In this example, 7 polygonal centroids are considered as feature points for traffic distribution and connected with the road network through different links.

Program the scheme based on MATLAB platform, realize the dynamic division and traffic distribution of the available public parking lot in the research area to establish Thiessen polygon analysis area, and realize the optimal location of public parking lot through parthenogenetic algorithm (PGA).

Matlab simulation results

Figure 6.1 Matlab model calculation work interface, source: programming screenshot

Matlab calculates the scheme and the results show that after 8 populations and 5 generations of inheritance, the optimal solution (minimum) of the overall carbon emission of the road network is obtained, and the optimal chromosome public parking lot combination numbers finally determined by GA are: 4, 7, 1 , 5, 6. During the calculation process, the solution to the road flow is as follows:

Figure 6.2 is the matrix result of Matlab's modeling and solving of the model. Among them, since the optimal solution appears after 5 generations of uniparental inheritance in this calculation example, there are a total of 5 groups of calculation results of road section flow, as shown in Figure 6.2(a). Among them, the optimal traffic distribution of road sections after each generation of inheritance is shown in Figure 6.2(b):

In this calculation example, there are 24 roads in two directions, so each generation of inheritance generates a total of 48 traffic data, and the matrix structure is shown in Figure 6.3. Calculate the traffic flow of the first population based on the third generation inheritance:

Figure 6.2 5 sets of road traffic matrix generated by iteration, image source: screenshot of Matlab programming

Figure 6.3 Flow distribution of 48 road sections in each matrix, image source: Matlab programming screenshot to calculate the overall road network carbon emissions based on the flow of each road section, the carbon emission calculation results after five generations of inheritance output the historical data of emissions as shown in Table 9;

Table 9 The flow carbon emission value of each road section under the optimal solution set, unit: 1.0e+0.4*(g)

Figure 7 shows the optimization effect of the overall carbon emissions of the road network during the iterative scheme of PGA. In this simulation, carbon emissions decreased slightly after the first few rounds of optimization, reached an optimum after the fourth iteration, and showed no further improvement thereafter. As shown in Figure 6. In total 1-7 available parking spaces 1, 4, 5, 6, 7 are the best combination of parking spaces. The Thiessen polygonal parking partition divided by the five parking lots divides the traffic flow of the road network, and calculates the driving path of the vehicle according to formula (2), as shown in Figure 11.

Reasonable explanation of carbon emission accounting

1. Test the rationality of the minimum carbon emissions of the road network

The total length of the road network in the study area is 4950 meters, two-way, and the road length is 9900 meters. According to the matlab simulation results of the road network flow, the total road network traffic flow is 414,742 vehicles. According to the calculated CO ₂ emission rates of different models, Table 10 estimates that the vehicle carbon emission rate is about 200g/km, and the road network emissions are about:

8.29×10 ⁷ (g).

Table 10 Carbon emissions of different vehicle operating conditions

The calculation result is 3.5 to 3.7×10 ⁷ (g). Although there are differences between the simulation results and the theoretical estimates, the magnitudes are consistent and the errors are within a reasonable range. After analysis, the possible reasons for the error in the calculation results are as follows: First, there are differences in the congestion conditions of the same road segment at different times, but the values are all within a reasonable range. The US Department of Transportation carbon emissions and experimental results were used to fit the relationship between rate and carbon emissions. Secondly, since carbon emissions are a cumulative value and the vehicle base is large, small differences in each vehicle will accumulate to produce a large difference in the overall value.

2. Calculation of carbon emissions in the genetic algorithm generation process scheme

The model uses genetic algorithm to generate multiple groups of parking lot combination schemes in the process of selecting the optimal solution, and calculates the carbon emissions of the road network according to the process scheme selected by GA. It is judged that the solution obtained by GA solution has the lowest carbon emission.

In summary, the calculation results verify the accuracy of the proposed carbon emission optimization model; at the same time, it also proves the feasibility of the parking location optimization model for congestion carbon emission control and the rationality of solving the model based on genetic algorithm.

As an important part of the low-carbon urban intelligent transportation system (ITS), this method will achieve the beneficial effect of reducing the carbon emission of congestion based on the route selection method based on parking decision. Combining parking navigation with road condition optimization through dynamic parking partitions to achieve dynamic parking guidance will help ensure the dynamic and static traffic efficiency of individual drivers in complex urban environments, and also provide the urban road network with the lowest carbon and the most travellers. Optimize and efficient overall path planning scheme.

All the public parking lots in the present invention are all the public parking lots that can be parked in real time. According to the driving conditions of the vehicle, first judge the moment, the availability of the parking lot, and then carry out the genetic algorithm to solve the problem, seek the combination of reasonable organization schemes, and divide the parking lot Partitioning, parking guidance is in progress. According to the genetic algorithm, the unreasonable parking lot and the parking lot with no available parking space are deleted from the candidate s parking lot and its corresponding Thiessen polygonal centroid, and the road traffic flow is carried out according to the Thiessen polygonal grid composed of all available parking lots. The division and calculation of the carbon emission scheme realize the planning and navigation of the optimal carbon emission plan.

The invention can fully plan the regulating effect of public parking lot resources on road traffic operation status according to static traffic facilities, and reduce the carbon emission effect caused by urban road traffic congestion. In addition to the real-time linkage between parking space information and road traffic status planning, dynamic guidance of emergency road conditions changes is added, and the route module is also used to recommend vehicles to enter the parking lot in a way that optimizes road network emissions, which not only enables drivers Travel is more convenient and efficient, and it has a beneficial effect on reducing urban carbon emissions. It is a revolution in the realization of static traffic to dynamic traffic regulation, response to road emergencies, and linkage technology.

Claims (3)

1. The navigation method based on the low-carbon target is carried out according to the following steps:

step 1: determining a starting point o and an end point d;

it is characterized in that the method comprises the steps of,

step 2: obtaining all feasible paths k among od in whole city ₁ ，k ₂ ，k ₃ ，…，k _m Is set K, k= { K ₁ ，k ₂ ，k ₃ ，…，k _m Defining a road sections in all feasible paths among the od to obtain a road section set A= {1,2,3 … a };

solving MinZ (X) according to the optimal principle of carbon emission of the road network, calculating the carbon emission of each path in the set K to obtain the carbon emission of each path in the set K, and obtaining an optimal path K with the minimum carbon emission ^’ ，k ^’ E, K, the vehicle follows the optimal path K ^’ Navigating to a destination;

the optimal principle of the carbon emission of the road network is calculated according to the following formula:

x is the carbon emission of the road network;

z is the overall carbon emission level of the traffic road network;

k represents the kth path in set K;

a is a road section number;

L _a the unit is m, which is the length of the road section a;

the unit is s for the free flow time of the road section a;

c _a the traffic capacity of the road section a is pcu/h;

alpha and beta are blocking coefficients, and 0.15 and 4 are taken respectively;

the unit is pcu for the flow on the kth path between od points with the starting point of o and the end point of d;

a road section-path variable, which is 0 or 1 variable; if the road section a belongs to the kth path between od points with the starting point of o and the end point of d, the road section a is +.>Otherwise->

f is a sixth order function subject to a speed-to-carbon emission rate conversion;

2. the parking lot navigation method based on the low-carbon target is carried out according to the following steps:

step (1): determining a starting point o and an end point d;

step (2): based on an open source map platform, acquiring real-time road condition information and parking facility use information at the current moment T, estimating the arrival time S according to the real-time road condition information, judging whether a matched parking lot has a free parking space when reaching a destination d, executing the step (3) if the matched parking lot has the free parking space when reaching the destination d, and executing the step (4) if the matched parking lot has no free parking space or the destination d has no matched parking space;

it is characterized in that the method comprises the steps of,

step (3): obtaining all feasible paths b among od in whole city ₁ ，b ₂ ，b ₃ ，…，b _m B= { B ₁ ，b ₂ ，b ₃ ，…，b _m Defining P road sections in all feasible paths among the od to obtain a road section set P= {1,2,3 … P };

solving MinZ (X) according to the optimal principle of carbon emission of the road network, and calculating the carbon emission of each path in the set B to obtain the carbon emission of each path in the set B, and obtaining an optimal path B with the minimum carbon emission ^’ ，b ^’ E P, the vehicle follows the optimal path b ^’ Navigating to a destination;

the optimal principle of the carbon emission of the road network is calculated according to the following formula:

x is the carbon emission of the road network;

z is the overall carbon emission level of the traffic road network;

b represents the B-th path in set B;

p is the road section number;

L _p the unit is m, which is the length of the road section p;

the free flow time of the road section p is s;

c _p the traffic capacity of the road section p is pcu/h；

Alpha and beta are blocking coefficients, and 0.15 and 4 are taken respectively;

the unit is pcu for the flow on the b-th path between od points with the starting point of o and the end point of d; />A road section-path variable, which is 0 or 1 variable; if the road section p belongs to the b-th path between od points with the starting point of o and the end point of d, the road section p is +.>Otherwise

f is a sixth order function subject to a speed-to-carbon emission rate conversion;

step (4): acquiring real-time road condition information and public parking lot use information at the current moment T, and pre-judging whether available public parking lots exist in 300 meters around the destination d;

if not, recommending the user to change the destination;

if so, N public parking lots with vacant parking spaces in the whole city range at the current moment are obtained, and the mass centers (N _x ，N _y ) Dividing Thiessen polygons for feature points to obtain a set O= {1,2, … i … j … N } of parking partitions, wherein i is more than or equal to 1 and less than or equal to N, j is more than or equal to 1 and less than or equal to N, and i is more than or equal to j, where i is (i) _x ，i _y ) Thiessen polygon partition starting at point, j is represented by (j _x ，j _y ) A Thiessen polygon partition for the endpoint;

n ', 2N' is less than or equal to N in N public parking lots, and the parking lots are formed Dividing an urban road network by the boundary of the alternative public parking lot combination alternative scheme to obtain a road section set R= {1,2,3 … R } and all feasible paths s among od ₁ ，s ₂ ，s ₃ ，…，s _m S = { S ₁ ，s ₂ ，s ₃ ，…，s _m }；

Solving MinZ (X) according to the principle of optimal road network carbon emission, and calculating the carbon emission of the urban road network divided by the combination alternative scheme of all public parking lots to obtain an optimal running path S' of the vehicle;

the vehicle navigates to the destination according to the optimal path S';

the specific formula is as follows:

x is the carbon emission of the road network;

z is the overall carbon emission level of the traffic road network;

r is the road section number;

ζ is the coefficient of the gravity model;

L _r the unit is m, which is the length of the road section r;

f is the proportion of travel in a self-driving mode in the Thiessen polygon partition;

P _i representing traffic volume generated by Thiessen polygon partition with i as a starting point;

B _j representing the traffic attraction generated by the Thiessen polygonal partition with j as the endpoint;

t _r (D _r ) Traffic impedance for a Thiessen polygon partition starting at i to a Thiessen polygon partition starting at j;

D _r the traffic flow on the road section r is pcu/h;

the unit is s, which is the free flow time of the road section r;

c _r the traffic capacity of the road section r is pcu/h;

g is the number of ij pairs, G is the number of ij pairs;

s is the number of possible paths; h is the number of paths through r segments in the feasible paths;

μ is the desired value;

e is a natural logarithmic base;

θ is a traffic conversion parameter, θ=3 to 3.5;

is any one of the possible paths +.>Is set, the total travel time of (a);

α and β are blocking coefficients, and α=0.15 and β=4 are taken respectively;

D _pr the number of the public parking lot positions of the r-section road is the number of the public parking lot positions of the r-section road;

D _er the number of parking lots is set for the buildings with the exits on the r sections;

E _pr the berth turnover rate is equal to the peak hour berth turnover rate of the public parking lot;

E _er the method comprises the steps of establishing a parking lot peak hour berth turnover rate for a building;

U _pr the utilization rate of berths is the peak hour of the public parking lot;

U _er the method comprises the steps of building a parking lot peak hour berth utilization rate for a building;

f is a sixth order function subject to a speed-to-carbon emission rate conversion;

3. the low-carbon target-based parking lot navigation method according to claim 2, wherein an emergency situation is encountered during navigation:

if the traffic accident occurs along the optimal path b', returning to the step (2) to recalculate and judge;

if the destination d is provided with a spare parking space in the parking lot, planning a path according to a formula (2), and updating to obtain an optimal path until the navigation to the destination d is finished;

if no available vehicle position exists in the matched parking lot, executing the step (4) according to the road condition at the time of T+nt, wherein n=1, 2,3,4 and … to obtain the vehicle running path s 'under the condition of minimum carbon emission of the whole road network' _T+nt Navigating to d; t is the current moment, and T is the time interval for acquiring the update of the open source map platform information;

if the public parking lot where the destination d is located is saturated, judging whether the current running position of the vehicle is within the range of the destination d300 meters or not;

if not, repartitioning the Thiessen polygon meshes according to the mass center of the public parking lot with the vacant parking spaces at the time of T+nt as the characteristic points, and recalculating the public parking lot positions and the optimal path s 'according to the formula (3)' _T+nt ；

If yes, navigating to a public parking lot where a Thiessen polygon partition where a road section where the vehicle is located when T+nt;

and (5) until the navigation to the destination is finished.