Path planning method for multi-cycle inspection of on-road parking spaces

Abstract

The invention discloses an on-road parking space multi-cycle inspection path planning method, which comprises the following steps: 1, establishing a secondary road network, setting an inspection time period, obtaining the length of a road section where each parking point in a node set is located, and calculating the driving time of the road section where each parking point is located; 2, obtaining the distance between nodes in the node set, and generating a driving time matrix; 3, establishing a parking point multi-cycle inspection path planning model; 4, defining a destruction operator set and a repair operator set; and 5, solving the parking point multi-cycle inspection path planning model to obtain an optimal path planning scheme. According to the method, the actual condition of current on-road parking space inspection can be combined, multi-period cooperative inspection arrangement is considered to optimize a path planning scheme, and therefore energy consumption can be reduced.

Description

Technical Field

[0001] The present invention belongs to the field of on-street parking inspection, and specifically relates to a path planning method for multi-cycle inspection of on-street parking spaces. Background Art

[0002] The Vehicle Routing Problem (VRP) is a common combinatorial optimization problem in the fields of transportation and scheduling. The aim is to find an optimal path planning scheme under the constraints such as meeting the demands of customer points. In the scenario of on-street parking space inspection, the service objects change from discrete customer points to continuous road sections. Since the VRP cannot directly describe the coverage characteristics of road sections, it is usually modeled as an Arc Routing Problem (ARP). However, when using the ARP model to describe the parking inspection problem currently, only a single inspection cycle is considered. When there are multiple inspection cycles for parking points in the road network, the existing models alone cannot handle it reasonably. Summary of the Invention

[0003] The present invention is to solve the above-mentioned deficiencies of the existing technologies, and proposes a path planning method for multi-cycle inspection of on-street parking spaces, aiming to combine the actual situation of current on-street parking space inspection, consider the coordinated arrangement of multi-cycle inspections to optimize the path planning scheme, so as to reduce energy consumption.

[0004] The present invention adopts the following technical solutions to achieve the above-mentioned invention purpose:

[0005] The path planning method for multi-cycle inspection of on-street parking spaces according to the present invention is characterized in that the road sections of the road are used as nodes in the secondary road network, and the shortest path between the end point of any road section and the start point of the next road section is used as an arc in the secondary road network, thereby constructing a secondary network , where N represents the nodes in the secondary road network, and , represents the set of start points, represents the set of end points, represents the set of parking points. Let the number of start points and end points be both m, and they correspond one by one, and the number of parking points is n; represents the arcs in the secondary road network, , represents the arc between any node i and node j. According to different inspection cycles, the parking points are divided into r types; let the inspection vehicle set K contain r×m inspection vehicles, and a line is formed by a number of parking points passed by each inspection vehicle from the start point to the end point, and a path planning scheme is formed by the lines of all inspection vehicles. The path planning method is carried out according to the following steps:

[0006] Step 1: Set the inspection period E, and obtain the average vehicle speed and the maximum vehicle speed of the vehicles traveling on the parking points within the inspection period E;

[0007] Step 2: Obtain the lengths of the road sections where each parking point in the node set N is located, and calculate the travel time of each road section where the parking point is located according to the average vehicle speed of each inspection vehicle;

[0008] Obtain the distances between each node in the node set N, and calculate the travel time between each node according to the maximum vehicle speed of each inspection vehicle, so as to generate a travel time matrix;

[0009] Step 3: Establish a multi-cycle inspection path planning model for parking points;

[0010] Step 4: Define the execution operations of the destruction operator set and the repair operator set ;

[0011] Step 5: Based on the operator set, solve the inspection path planning model for parking points to obtain an optimal path planning scheme;

[0012] Step 5.1: Set and initialize the current iteration number , define the maximum iteration number as , adopt a greedy strategy to generate the I-th generation path planning scheme , let the optimal path planning scheme ; Calculate the temperature of the I-th generation , where is the temperature control parameter;

[0013] Step 5.2: Use the destruction operator and the repair operator to for updating to obtain the updated I-th generation path planning scheme ;

[0014] Step 5.3: Let be assigned to after that, if , then output the optimal path planning scheme , otherwise, calculate the temperature of the (I + 1)-th iteration , and take as the (I + 1)-th generation path planning scheme , and return to step 5.2 to execute sequentially, where c is the cooling coefficient.

[0015] The characteristics of a path planning method for multi-cycle inspection of on-street parking spaces according to the present invention also lie in that step 3 includes:

[0016] Step 3.1: Use formula (1) to construct the objective function Z of the inspection path planning model for parking points;

[0017] (1)

[0018] In formula (1), represents the starting point assigned to inspection vehicle k, , ; represents whether inspection vehicle k immediately inspects parking point j after departing from the starting point . If so, let , otherwise, let ; represents the out - node of node i, represents whether the route of inspection vehicle k in the route planning scheme includes arc , that is, whether inspection vehicle k immediately inspects parking point j after parking at parking point i. If so, let , otherwise, let ; represents the single - longest inspection time of inspection vehicle k, represents the fixed energy consumption of the inspection vehicle during the inspection period E, represents the unit - distance energy consumption of the inspection vehicle, is a coefficient; represents arc 's length, represents the length of the road section where parking point i is located;

[0019] Step 3.3: Use formulas (2) to (11) to construct the constraint conditions of the parking point inspection route planning model:

[0020] (2)

[0021] (3)

[0022] (4)

[0023] (5)

[0024] (6)

[0025] (7)

[0026] (8)

[0027] (9)

[0028] (10)

[0029] (11)

[0030] In formulas (2) to (11), represents the end point of inspection vehicle k, represents the inspection cycle of parking point i; represents arc 's passing time, represents the inspection time of parking point i; represents the time when inspection vehicle k arrives at parking point j and starts inspection, represents the time when inspection vehicle k arrives at parking point i and starts inspection, where M is a positive number; represents whether the route of inspection vehicle k in the path planning scheme includes arc , that is, whether inspection vehicle k immediately inspects parking point i after inspecting parking point j. If so, let , otherwise, let ; represents the incoming node of node i, represents whether the route of inspection vehicle k in the path planning scheme includes arc , that is, whether inspection vehicle k inspects parking point i before reaching the end point . If so, let , otherwise, let .

[0031] Furthermore, step 4 includes:

[0032] Step 4.1: Define the first destruction operator 's execution operations include:

[0033] Step 4.1.1: Randomly select a parking point i in the current path planning scheme and add it to the set D of deleted parking points; add all parking points not included in D to the new set L of parking points;

[0034] Step 4.1.2: Randomly select a parking point i from D, calculate the similarity between parking point i and each parking point in L using formula (12), and then sort all the parking points in L in ascending order according to the similarity to obtain the sorted set of parking points ;

[0035] (12)

[0036] In formula (12), is the travel time coefficient, is the demand coefficient; represents the inspection time of parking point j in L;

[0037] Step 4.1.3: Generate a random number and calculate the index , where represents the number of parking points in

[0038] Step 4.1.4: Obtain the parking point in with index , and after adding the parking point to D, remove the parking point from ;

[0039] Step 4.1.5: If the number of parking points in D reaches q, then delete all the parking points in D from , and update the inspection cycle of all lines in and the inspection cycle of all lines in

[0040] Step 4.2: Define the second destruction operator The execution operation of is to randomly select a parking point i in the current path planning scheme , and after deleting it from , add it to the set D of deleted parking points, and update and the inspection cycle of all lines in

[0041] Step 4.3: Define the first repair operator The execution operation of

[0042] Step 4.3.1: Assume that any parking point is inserted into each position of each line in

[0043] Step 4.3.2: After finding the optimal position that makes the increment of the objective function value the smallest, insert the parking point into the optimal position of the corresponding line in the path planning scheme , and delete from D until ; ;

[0044] Step 4.4: Define the second repair operator The execution operations include:

[0045] Step 4.4.1: Assume that any parking point is inserted into the optimal position of each route, where the optimal position makes the increment of the objective function value of the smallest, and add the increment of the objective function value of each route after insertion to the increment set

[0046] Step 4.4.2: After sorting the increments in in ascending order, the sorted increment set is obtained, and the regret value of parking point i is calculated; where, , respectively represent the first and second increments in

[0047] Step 4.4.3: Select the parking point corresponding to the largest regret value in D and insert it into the optimal positions of all routes in where the optimal position is such that the increment of the objective function value of is the smallest, delete from D until

[0048] is reached, and the candidate path planning scheme

[0049] Step 5.2 further includes: and is obtained; and the repair operator and form the operator pair in the I-th iteration;

[0050] Step 5.2.1: Use the roulette wheel selection method to select the destruction operator in the I-th iteration to perform destruction on and the set of nodes removed in the I-th iteration ;

[0051] Step 5.2.2: According to the repair operator in the I-th iteration to perform repair on ;

[0052] Step 5.2.4: Calculate respectively according to formula (1) the objective function value of and the objective function value of , and thus calculate the acceptance probability of under the I-th iteration according to formula (14) :

[0053] (14)

[0054] Step 5.2.5: Generate a pseudo-random number under the I-th iteration ; if or , then take as the updated path planning scheme of the I-th generation , and let the objective function value of ; otherwise, take as the updated path planning scheme of the I-th generation , and let ;

[0055] Step 5.2.6: If , then let ; otherwise remain unchanged.

[0056] An electronic device according to the present invention includes a memory and a processor, characterized in that the memory is used to store a program for supporting the processor to execute the path planning method, and the processor is configured to execute the program stored in the memory.

[0057] A computer-readable storage medium according to the present invention, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, it executes the steps of the path planning method.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0059] 1. By establishing a two-level road network, considering the parking points with multiple inspection cycles in the road network, using the total energy consumption during the inspection period as the objective function, and constructing a mixed-integer linear programming model that minimizes the total energy consumption by using the VRP model, the present invention makes up for the deficiency that the existing ARP model cannot describe the multi-cycle inspection path planning problem of on-road parking spaces, thereby significantly reducing the inspection energy consumption under complex road networks and providing a scientific decision-making scheme that takes into account both efficiency and sustainability for urban-level intelligent parking management.

[0060] 2. The present invention uses a large neighborhood search algorithm with a hybrid simulated annealing mechanism to solve the mixed-integer linear programming model, and utilizes the advantage that the simulated annealing mechanism can jump out of the local optimum to make up for the deficiency of premature convergence of the general large neighborhood search algorithm, so as to obtain a patrol path planning scheme with the lowest total energy consumption. Description of the Drawings

[0061] Figure 1 It is the flowchart of the large neighborhood search algorithm with the hybrid simulated annealing mechanism of the present invention;

[0062] Figure 2 It is the berth diagram on one side of the one-way road of the present invention;

[0063] Figure 3 It is the berth diagram on both sides of the one-way road of the present invention;

[0064] Figure 4 It is the berth diagram on both sides and one side of the two-way lane of the present invention;

[0065] Figure 5 It is the example diagram of the patrol path planning scheme of the present invention. Detailed Embodiment

[0066] In this embodiment, a path planning method for multi-cycle patrol of on-street parking spaces takes the road sections as nodes in the secondary road network, and takes the shortest path between the end point of any road section and the start point of the next road section as the arc in the secondary road network, so as to construct a secondary network , where N represents the nodes in the secondary road network, and , represents the set of start points, represents the set of end points, represents the set of parking points. Let the number of start points and end points be both m, and they correspond one by one, and the number of parking points is n; represents the arc in the secondary road network, , represents the arc between any node i and node j; Figure 2 , Figure 3 , Figure 4 shows the relationship between the berths, the traffic flow and the driving direction of the patrol vehicle. In the figure, the solid arrows indicate the traffic flow direction, and the dashed arrows indicate the driving direction of the patrol vehicle. Among them, Figure 2 the (a 1 ) part and the (b 1 ) part in respectively represent setting berths on the right side of the traffic flow on the one-way road, Figure 2 the (a 2 ) part and the (b 2 ) part in respectively represent setting berths on the left side of the traffic flow on the one-way road, Figure 3 the (a) part and the (b) part in respectively represent setting berths on both sides of the one-way road,Figure 4 In (a 1 ), (b 1 ), the parts respectively indicate that berths are set on both sides of a two-way lane. Figure 4 In (a 2 ), (b 2 ), the parts respectively indicate that berths are set on one side of a two-way lane. According to different inspection cycles, the parking points are divided into r types; let the inspection vehicle set K contain r × m inspection vehicles, and a route is composed of a number of parking points passed by each inspection vehicle from the starting point to the ending point, and a path planning scheme is formed by the routes of all inspection vehicles. A path planning scheme contains multiple routes composed of the starting point, ending point, and inspected parking points of the vehicle, as Figure 5 shown. Figure 5 The numbers on the road section in are the road section and parking point numbers, and the numbers in the circles are the intersection numbers. Two routes are respectively marked in blue and red. The solid arrows indicate that the inspection vehicle inspects the parking points on the road section, and the dashed arrows indicate directly passing through the road section. The path planning method is carried out according to the following steps:

[0067] Step 1: Set the inspection time period E, which is usually several consecutive hours in a day, and obtain the average speed and maximum speed of vehicle travel at the parking points within the inspection time period E; the vehicle passes through the road section where the parking point is located at the average speed and reaches the next parking point at the maximum speed.

[0068] Step 2: Obtain the lengths of the road sections where each parking point in the node set N is located, and calculate the travel time of each road section where the parking point is located according to the average speed of each inspection vehicle;

[0069] Obtain the distances between each node in the node set N, and calculate the travel time between each node according to the maximum speed of each inspection vehicle, so as to generate a travel time matrix.

[0070] Step 3: Establish a multi-cycle inspection path planning model for parking points;

[0071] Step 3.1: Use Equation (1) to construct the objective function Z of the parking point inspection path planning model;

[0072] (1)

[0073] In Equation (1), represents the starting point assigned to inspection vehicle k, , ; represents whether inspection vehicle k immediately inspects parking point j after departing from the starting point . If so, let , otherwise, let ; Denote the set of out - nodes of node \(i\). For any , the set of out - nodes contains the starting point and the remaining parking points except \(i\); for any , , the set of out - nodes contains the ending point and all parking points; Denote whether the route of inspection vehicle \(k\) in the route planning scheme contains arc , that is, whether inspection vehicle \(k\) immediately inspects parking point \(j\) after inspecting parking point \(i\). If so, let , otherwise, let ; Denote the single - longest inspection time of inspection vehicle \(k\), Denote the fixed energy consumption of the inspection vehicle during the inspection period \(E\), Denote the energy consumption per unit distance of the inspection vehicle, is a coefficient; Denote the length of arc , Denote the length of the road section where parking point \(i\) is located.

[0074] Step 3.3: Use equations (2) - (11) to construct the constraint conditions of the parking - point inspection route planning model:

[0075] (2)

[0076] (3)

[0077] (4)

[0078] (5)

[0079] (6)

[0080] (7)

[0081] (8)

[0082] (9)

[0083] (10)

[0084] (11)

[0085] Equation (2) indicates that any parking spot is assigned a vehicle for inspection; Equation (3) ensures that any vehicle must start from its origin; Equation (4) ensures that the route of any vehicle from its origin to its destination is continuous, that is, for each vehicle, it must leave parking spot i after arriving at it; Equation (5) indicates that any vehicle must finally reach the destination; Equation (6) indicates that the number of vehicles used for inspection does not exceed a given quantity; Equation (7) indicates that the travel time of any vehicle cannot exceed the minimum cycle among the visited parking spots to meet the inspection time interval requirements of each parking spot; Equation (8) indicates that the driving time of any vehicle does not exceed the maximum duration; Equation (9) indicates the relationship between the start service times of two sequentially served nodes; Equation (10) indicates that the visit time meets the time window, that is, it ensures that the visit time is within the inspection cycle of the parking spot; Equation (11) indicates the value range of the decision variables.

[0086] In Equations (2) to (11), represents the destination of inspection vehicle k, represents the inspection cycle of parking spot i; represents arc the passing time of, represents the inspection time used at parking spot i; represents the time when inspection vehicle k arrives at parking spot j and starts inspection, represents the time when inspection vehicle k arrives at parking spot i and starts inspection, M is a positive number; represents whether the route of inspection vehicle k in the path planning scheme contains arc , that is, whether inspection vehicle k immediately inspects parking spot i after inspecting parking spot j. If so, let , otherwise, let ; represents the set of in-nodes of node i. For any , the set of in-nodes contains the origin and the remaining parking spots except i; represents whether the route of inspection vehicle k in the path planning scheme contains arc , that is, whether inspection vehicle k inspects parking spot i before reaching the destination . If so, let , otherwise, let .

[0087] Step 4: Define the execution operations of the destruction operator set and the repair operator set ;

[0088] Step 4.1: Define the execution operations of the first destruction operator including:

[0089] Step 4.1.1: In the current path planning scheme Randomly select a parking point \(i\) and add it to the set \(D\) of deleted parking points; add all parking points not included in \(D\) to the new set \(L\) of parking points.

[0090] Step 4.1.2: Randomly select a parking point \(i\) from \(D\), calculate the similarity between parking point \(i\) and each parking point in \(L\) using Equation (12), and then sort all the parking points in \(L\) in ascending order according to the similarity to obtain the sorted set of parking points ;

[0091] (12)

[0092] In Equation (12), is the travel time coefficient, is the demand coefficient; represents the inspection time of parking point \(j\) in \(L\);

[0093] Step 4.1.3: Generate a random number and calculate the index , where represents the number of parking points in, and \(h\) represents the coefficient;

[0094] Step 4.1.4: Obtain the parking point in with the index , add the parking point to \(D\), and then remove the parking point from ; Step 4.1.5: If the number of parking points in \(D\) reaches \(q\), delete all the parking points in \(D\) from

[0095] and update the inspection cycle of all lines in to obtain the damaged path planning scheme and the inspection cycle of all lines in as well as ; Otherwise, return to Step 4.1.2 and execute sequentially.

[0096] Step 4.2: Define the second destruction operator The execution operation of is to randomly select a parking point \(i\) in the current path planning scheme , delete it from , add it to the set \(D\) of deleted parking points, and update the inspection cycle of all lines in until the number of deleted parking points reaches \(q\), so as to obtain the damaged path planning scheme as well as the inspection cycle of all lines in

[0097] Step 4.3: Define the first repair operator The execution operation is to insert the parking points in D into the optimal positions of the damaged path planning scheme until all the deleted parking points in D are re-inserted into the damaged path planning scheme to obtain a candidate path planning scheme , including:

[0098] Step 4.3.1: Assume that any parking point is inserted into each position of each line in

[0099] and the increments of the objective function values at different positions are obtained; Step 4.3.2: After finding the optimal position that minimizes the increment of the objective function value of , insert the parking point into the optimal position of the corresponding line in the path planning scheme and delete from D until .

[0100] Step 4.4: Define the second repair operator The execution operation is to select the parking point with the largest regret value each time and insert it into the optimal position in the damaged path planning scheme until all the deleted parking points are re-inserted into the damaged path planning scheme to obtain a candidate path planning scheme , including:

[0101] Step 4.4.1: Assume that any parking point is inserted into the optimal position of each line in , where the optimal position minimizes the increment of the objective function value of and add the increment of the objective function value of each line after insertion to the increment set ;

[0102] Step 4.4.2: After sorting the increments in in ascending order, obtain the sorted increment set , and calculate the regret value of parking point i; where , respectively represent the first and second increments in

[0103] Step 4.4.3: Select the parking point corresponding to the largest regret value in D and insert it into Among the optimal positions of all the lines, the optimal position is such that the increment of the objective function value is minimized, and is deleted from D until is reached, obtaining the candidate path planning scheme .

[0104] Step 5: Based on the operator set, solve the parking point inspection path planning model to obtain the optimal path planning scheme. The algorithm flow is as Figure 1 shown;

[0105] Step 5.1: Set and initialize the current iteration number , define the maximum iteration number as , generate the I-th generation path planning scheme using the greedy strategy, and let the optimal path planning scheme ; calculate the I-th generation temperature , where is the temperature control parameter;

[0106] Step 5.2: Use the destruction operator and the repair operator to update, obtaining the updated I-th generation path planning scheme ;

[0107] Step 5.2.1: Use the roulette wheel selection method to select the destruction operator and the repair operator at the I-th iteration from and respectively, and form the operator pair at the I-th iteration. Among them, the number q of parking points to be deleted by the destruction operator satisfies , is the coefficient.

[0108] Step 5.2.2: After using the destruction operator at the I-th iteration to destroy , obtain the path scheme after destruction at the I-th iteration and the set of nodes removed at the I-th iteration;

[0109] Step 5.2.3: According to , use the repair operator at the I-th iteration to repair , obtaining the candidate path planning scheme at the I-th iteration;

[0110] Step 5.2.4: Calculate Objective function value and Objective function value , so as to calculate the acceptance probability of at the I-th iteration according to Equation (14): :

[0111] (14)

[0112] Step 5.2.5: Generate a pseudo-random number at the I-th iteration ; if or , then use as the updated path planning solution for the I-th generation , and let the objective function value of be ; otherwise, use as the updated path planning solution for the I-th generation , and let .

[0113] Step 5.2.6: If , then let ; otherwise, remains unchanged.

[0114] Step 5.3: After assigning to , if , then output the optimal path planning solution , otherwise, calculate the temperature at the (I + 1)-th iteration , use as the path planning solution for the (I + 1)-th generation , and return to Step 5.2 to execute sequentially, where c is the cooling coefficient.

[0115] In this embodiment, an electronic device includes a memory and a processor. The memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0116] In this embodiment, a computer-readable storage medium stores a computer program, and when the computer program is run by a processor, it executes the steps of the above method.

Claims

1. A path planning method for multi-cycle inspection of on-street parking spaces, characterized in that: The road sections are taken as nodes in the secondary road network, and the shortest path between the end point of any section and the starting point of the next section is taken as the arc in the secondary road network, so as to construct a secondary network. , where N represents the nodes in the secondary network, and , represents the starting point set, represents the endpoint set, Represents the set of parking points, let the number of starting points and end points be m, and they correspond one to one, and the number of parking points be n; represents the arc in the secondary road network, , Represents the arc between any node i and node j. According to different inspection cycles, the parking points are divided into r types. Let the inspection vehicle set K contain r×m inspection vehicles. A route is composed of several parking points passed by each inspection vehicle from the starting point to the end point. The routes of all inspection vehicles constitute a path planning scheme. The path planning method is performed in the following steps: Step 1: Set the inspection period E, and obtain the average speed and maximum speed of vehicles traveling at the parking point within the inspection period E; Step 2: Obtain the length of the road section where each parking point in the node set N is located, and calculate the travel time of the road section where each parking point is located based on the average speed of each patrol vehicle; Obtain the distance between each node in the node set N, and calculate the travel time between each node according to the maximum speed of each inspection vehicle, so as to generate a travel time matrix; Step 3: Establish a multi-cycle inspection path planning model for parking points; Step 4: Define the set of destruction operators and the set of repair operators Execution operation; Step 5: Based on the operator set, solve the parking inspection path planning model to obtain the optimal path planning solution; Step 5.1: Set and initialize the current number of iterations , define the maximum number of iterations as , using the greedy strategy to generate the first generation path planning solution , let the optimal path planning solution ; Calculate the temperature of the first generation ,in, is the temperature control parameter; Step 5.2: Apply the destruction operator and the repair operator to Update to obtain the updated I-generation path planning solution ; Step 5.3: Order Assign to After that, if , then output the optimal path planning solution Otherwise, calculate the I+1th iteration temperature ,Will As the I+1 generation path planning solution , return to step 5.2 and execute sequentially, where c is the cooling coefficient.

2. A path planning method for multi-cycle inspection of on-street parking spaces according to claim 1, characterized in that: Step 3 includes: Step 3.1: Use equation (1) to construct the objective function Z of the parking inspection path planning model; (1) In formula (1), Indicates the starting point where the patrol vehicle k is assigned, , ; Indicates that patrol car k starts from the starting point After departure, should we inspect parking point j immediately? If so, let , otherwise, let ; represents the outgoing node of node i, Indicates whether the route of patrol vehicle k in the path planning solution contains an arc , that is, whether the inspection vehicle k inspects the parking point j immediately after inspecting the parking point i. If so, let , otherwise, let ; It represents the longest single inspection time of inspection vehicle k. represents the fixed energy consumption of the inspection vehicle during the inspection period E, It represents the energy consumption per unit distance of the inspection vehicle. is the coefficient; Representing Arc Length, Indicates the length of the road section where parking point i is located; Step 3.3: Use equations (2) to (11) to construct the constraints of the parking inspection path planning model: (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) In formula (2) to formula (11), represents the end point of the inspection vehicle k, represents the inspection cycle of parking point i; Representing Arc The passing time, represents the inspection time of parking point i; represents the time when the inspection vehicle k arrives at the parking point j and starts the inspection, It represents the time when the inspection vehicle k arrives at the parking point i and starts the inspection. M is a positive number; Indicates whether the route of patrol vehicle k in the path planning solution contains an arc , that is, whether the inspection vehicle k immediately inspects the parking point i after inspecting the parking point j. If so, let , otherwise, let ; represents the connected node of node i, Indicates whether the route of patrol vehicle k in the path planning solution contains an arc , that is, the inspection vehicle k reaches the end point Whether parking point i has been inspected before, if so, let , otherwise, let .

3. A path planning method for multi-cycle inspection of on-street parking spaces according to claim 2, characterized in that: Step 4 includes: Step 4.1: Define the first destruction operator The execution operations include: Step 4.1.1: In the current path planning scheme Randomly select a parking point i from the set and add it to the deleted parking point set D; add all parking points not included in D to the new parking point set L; Step 4.1.2: Randomly select a parking point i from D, and use formula (12) to calculate the similarity between parking point i and each parking point in L, then sort all parking points in L in ascending order according to the similarity to obtain the sorted parking point set ; (12) In formula (12), is the travel time coefficient, is the demand coefficient; represents the inspection time of parking point j in L; Step 4.1.3: Generate random numbers , and calculate the index ,in, express The number of parking spots in the middle, h represents the coefficient; Step 4.1.4: Get The index is Parking spot and the parking spot After adding to D, the parking point from Eliminate; Step 4.1.5: If the number of parking spots in D reaches q, Delete all parking points in D and update The inspection cycle of all lines in the system is used to obtain the path planning scheme after the damage. as well as Otherwise, return to step 4.1.2 and execute sequentially; Step 4.2: Define the second destruction operator The execution operation is in the current path planning scheme Randomly select a parking point i from After being deleted, it is added to the deleted parking point set D and updated The inspection cycle of all routes in the process is repeated until the number of parking spots deleted reaches q, thus obtaining the path planning solution after the destruction. as well as Inspection cycle of all lines in the system; Step 4.3: Define the first repair operator The execution operations include: Step 4.3.1: Assume that any parking point insert At each position of each line in the network, and obtain the increment of the objective function value at different positions; Step 4.3.2: Find the After the increment of the objective function value is the smallest, the parking point Insert into the path planning solution The optimal position of the corresponding line in Delete from D until So far, the candidate path planning scheme is obtained ; Step 4.4: Define the second repair operator The execution operations include: Step 4.4.1: Assume that any parking point Insert into The optimal position of each line in The increment of the objective function value of the inserted line is the smallest, and the increment of the objective function value of each line after insertion is added to the increment set. middle; Step 4.4.2: After sorting the increments in ascending order, we get the sorted increment set , calculate the regret value of parking point i ;in, , Respectively The first and second increments; Step 4.4.3: Select the parking point corresponding to the maximum regret value in D and insert into Among the optimal positions of all lines in , the optimal position is such that The increment of the objective function value is the smallest. Delete from D until So far, the candidate path planning solution is obtained .

4. A path planning method for multi-cycle inspection of on-street parking spaces according to claim 3, characterized in that: Step 5.2 includes: Step 5.2.1: Use the roulette wheel selection method to select and Select the destruction operator under the Ith iteration and repair operator And form the operator pair under the Ith iteration ; Step 5.2.2: Using the destruction operator at iteration I right After the destruction, the path plan after the destruction in the Ith iteration is obtained And the set of nodes removed in the Ith iteration ; Step 5.2.3: According to , using the repair operator under the Ith iteration right After repair, the candidate path planning solution for the Ith iteration is obtained ; Step 5.2.4: Calculate formula (1) The objective function value of and The objective function value of , and then calculate the I-th iteration according to formula (14) The acceptance probability : (14) Step 5.2.5: Generate pseudo-random numbers for iteration I ;like or , then As an updated I-generation path planning solution , and order The objective function value of Otherwise, As an updated I-generation path planning solution , and order ; Step 5.2.6: If , then let ;otherwise, Remain unchanged.

5. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the path planning method described in any one of claims 1 to 4, and the processor is configured to execute the program stored in the memory.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the path planning method according to any one of claims 1 to 4 are executed.

Images