A spatiotemporal matching method for shared parking demand and parking spaces for driverless vehicles
Through time segmentation and simulated annealing algorithm, the displacement of unmanned vehicles is optimized, and the waste of resources and traffic accidents caused by frequent displacement of unmanned vehicles during shared parking is solved, and more efficient parking space utilization and parking requirements are achieved.
Patent Information
- Application Number
- CN202210077617.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-24
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2042-01-24
AI Technical Summary
During shared parking, the frequent displacement of driverless vehicles leads to increased costs of using vehicles, waste of resources and frequent traffic accidents. The existing technology is difficult to effectively solve the problem of matching parking spaces where driverless vehicles and humanized vehicles coexist.
Using time segmentation and simulated annealing algorithm, by dividing the parking demand time and parking space supply time into smaller time periods, a method for matching shared parking demand and parking spaces for unmanned driving is designed to optimize the displacement of unmanned vehicles. Reduce resource waste and traffic accidents.
It effectively improves the utilization rate of parking spaces, meets parking needs, and reduces the number of shifts of unmanned vehicles, reducing the cost of using vehicles and the incidence of traffic accidents.
Smart Images

Figure CN114418413B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of shared parking and unmanned driving technology, and in particular to a method for spatiotemporal matching of shared parking demands and parking spaces for unmanned driving. Background Art
[0002] As the number of cars in my country continues to rise, it not only causes road traffic congestion, but also aggravates the problem of parking difficulties in cities. The number of traditional parking spaces can no longer meet the growing parking demand. With the development of electronic technology, shared parking has become a feasible way to integrate existing resources to solve the parking problem. Shared parking has the advantages of low cost and short transformation time. Compared with new parking facilities, it not only saves time and cost, but also has better and more obvious effects. Shared parking is to use the idle time of parking facilities to establish a sharing relationship with adjacent communities, so that parking space owners can obtain income by renting out the idle time of parking spaces, and vehicle owners can park their vehicles in shared parking spaces, which not only improves the utilization rate of social resources while seeking convenience and benefits for people.
[0003] At present, driverless buses, driverless taxis and driverless work vehicles have entered people's lives, and driverless technology is becoming more and more mature. While using shared parking to solve the problem of parking difficulties, it is necessary to consider the impact of driverless vehicles on shared parking. In the process of shared parking, driverless vehicles are different from manned vehicles. On the one hand, driverless vehicles have higher requirements for parking facilities and related technical means. On the other hand, driverless vehicles have the function of "empty driving" and can move autonomously under information control or remote operation; in the process of shared parking, driverless vehicles can use the idle hours of parking spaces to park through autonomous shifting, improve the utilization rate of parking spaces, and maximize the satisfaction of parking needs.
[0004] However, unreasonable parking plans may also lead to frequent relocation of driverless vehicles, resulting in increased vehicle usage costs, waste of resources, and increased accident rates during vehicle relocation. Summary of the invention
[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a method for matching shared parking needs and parking spaces in time and space for unmanned driving, which utilizes the "autonomous shifting" feature of unmanned vehicles to improve parking space utilization and meet parking needs, while optimizing the number of unmanned vehicle shifts during shared parking to reduce resource waste and reduce the incidence of traffic accidents in parking areas. In order to achieve the above-mentioned purpose and other advantages of the present invention, a method for matching shared parking needs and parking spaces in time and space for unmanned driving is provided, comprising:
[0006] S1. Propose the relevant assumptions needed to design the supply-demand matching method;
[0007] S2, time segmentation, by dividing the parking demand time and parking space supply time into smaller time periods, characterizing the autonomous shifting of unmanned vehicles during shared parking;
[0008] S3. According to the spatiotemporal matching characteristics of parking demand and parking supply of manned and unmanned vehicles in the shared parking process, a targeted simulated annealing algorithm is designed on the basis of time segmentation to determine the supply and demand matching method that meets the parking demand with the least number of vehicle moves.
[0009] Preferably, the relevant assumptions in step S1 include:
[0010] (1) The parking demand time and parking space supply time of shared parking are known;
[0011] (2) To ensure that all parking demands are met, it is assumed that the parking demand at any moment is less than or equal to the parking space supply;
[0012] (3) A manned vehicle must park in one parking space during the parking demand period; while an unmanned vehicle can park in different parking spaces during two consecutive parking demand periods;
[0013] (4) Assume that displacement only occurs at the beginning and end of parking demand and supply, and assume that the cost of each displacement is the same.
[0014] Preferably, the time segmentation in step S2 includes the following steps:
[0015] S21, putting the start time and end time of parking demand and parking space supply into one set;
[0016] S22, remove the same elements in the set, and arrange the remaining elements in ascending order to obtain an ordered set;
[0017] S23. Taking the time nodes in the set as breakpoints, all parking demand times and parking space supply times are divided.
[0018] Preferably, step S3 includes defining a feasible solution of the simulated annealing algorithm by using E i =(e1,ej,......,el) represents the time period pairing composed of all pairs in time period i; and Indicates pairing of two different time periods in time period i, using M = {E 1 ,E 2 ,......,E i ,.....,E n} represents the pairing scheme composed of all time period pairings; M a and M b Represents two different pairing schemes, that is, two feasible solutions of the simulated annealing algorithm.
[0019] Preferably, step S3 includes determining the neighbors of the solution by defining the neighbors of the pairing scheme. For any pairing scheme There is a pairing scheme When satisfied and When M b ∈U(M a ), namely M b M a A neighbor of . a ) is M a 's neighborhood.
[0020] Preferably, based on the relevant definitions of feasible solutions and feasible solution neighbors, the simulated annealing algorithm comprises the following steps:
[0021] S31, initialization parameters, assuming that the initial temperature is T0, the cooling coefficient is α, and the end temperature is T m , the maximum number of iterations is β m , the maximum number of neighborhood searches γ m . Assume that the current temperature T = T0, the current number of iterations β = 1, and the current number of neighborhood searches γ = 0;
[0022] S32, generate an initial solution M0, first pair the parking demand of any time period in the network with the shared parking space, then reasonably combine the time period pairings of all time periods to establish an initial pairing scheme M0, and let the current solution M cur and the optimal solution M best Equal to M0;
[0023] S33, neighborhood search to obtain M cur Neighbor M cur2 , randomly generate a time period pairing on a randomly selected time period i Then, use Replace the current pairing scheme M cur Time period pairing within time period i Get M cur A neighbor of M cur2 ;
[0024] S34, if the number of car moves in the new pairing scheme is S cur2 Less than the number of car moves S of the current pairing solution cur , then accept the new solution and update the current solution to the new solution; otherwise, calculate the increment ΔS = S cur2 -S cur The value of and generates a random number δ in [0,1]; when δ≤e -Δs / T When , update the current solution to the new solution, otherwise the current solution remains unchanged;
[0025] S35, determine whether to cool down, if γ>γ m , then the temperature drops; let T = T0(1-α) n , β=β+1, γ=0, otherwise, set γ=γ+1 and go to step S33;
[0026] S36, determine whether to terminate the calculation. When the termination temperature T is reached m When the number of times that the new solution is not selected reaches a given value or the number of car moves of the current pairing solution is 0, the current pairing solution is output and the calculation ends; otherwise, go to step S33.
[0027] Compared with the prior art, the beneficial effects of the present invention are: solving the problem of matching the demand and parking spaces for shared parking where unmanned vehicles and manned vehicles coexist; refining the parking process of shared parking vehicles; and reducing the increased vehicle use costs, waste of resources and traffic accidents caused by frequent vehicle shifting by optimizing the number of shifts of unmanned vehicles during shared parking. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 It is an algorithm flow chart of the method for spatiotemporal matching of shared parking demand and parking space for unmanned driving according to the present invention;
[0029] Figure 2 It is a schematic diagram of a pairing scheme for the shared parking demand and the spatiotemporal matching method of parking spaces for unmanned driving according to the present invention. DETAILED DESCRIPTION
[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0031] Reference Figure 1-2 , a method for spatiotemporal matching of shared parking demand and parking spaces for unmanned driving, including: S1, based on the characteristics of "autonomous displacement" of unmanned vehicles, analyzing the impact of unmanned vehicles on shared parking, and proposing relevant assumptions required for designing a supply and demand matching method;
[0032] S2, time segmentation, by dividing the parking demand time and parking space supply time into smaller time periods, characterizing the autonomous shifting of unmanned vehicles during shared parking;
[0033] S3. According to the spatiotemporal matching characteristics of parking demand and parking supply of manned and unmanned vehicles in the shared parking process, a targeted simulated annealing algorithm is designed on the basis of time segmentation to determine the supply and demand matching method that meets the parking demand with the least number of vehicle moves.
[0034] Furthermore, the relevant assumptions in step S1 include:
[0035] (1) The parking demand time and parking space supply time of shared parking are known;
[0036] (2) To ensure that all parking demands are met, it is assumed that the parking demand at any moment is less than or equal to the parking space supply;
[0037] (3) A manned vehicle must park in one parking space during the parking demand period; while an unmanned vehicle can park in different parking spaces during two consecutive parking demand periods;
[0038] (4) Assuming that the shift only occurs at the beginning and end of the parking demand and supply, and assuming that the cost of each shift is the same, based on the above assumptions, the autonomous shift of unmanned vehicles can maximize the parking demand, but unreasonable parking plans may lead to frequent shifts of vehicles, resulting in waste of resources and frequent traffic accidents. Therefore, the present invention is committed to finding a method to meet the shared parking demand with the least number of shifts.
[0039] Furthermore, the time segmentation in step S2 includes the following steps:
[0040] S21, putting the start time and end time of parking demand and parking space supply into one set;
[0041] S22, remove the same elements in the set, and arrange the remaining elements in ascending order to obtain an ordered set;
[0042] S23. Taking the time nodes in the set as breakpoints, all parking demand times and parking space supply times are divided.
[0043] Furthermore, in order to determine a shared parking solution that meets parking demand with the least number of vehicle moves, a simulated annealing algorithm is designed based on the spatiotemporal matching characteristics of parking demand and parking supply, with the optimization goal of minimizing the number of vehicle moves of driverless vehicles. It is necessary to define two key concepts of the simulated annealing algorithm: feasible solution and solution neighborhood. Step S3 includes defining a feasible solution of the simulated annealing algorithm, with e(v, p, i) representing a pairing of vehicle v parked in parking space p during time period i; e(0, p, i) represents that there is no vehicle parked in parking space p during time period i. Since the demand at any moment is less than the supply, there is no situation where there is no parking space for the vehicle. By using E i=(e1,ej,......,el) represents the time period pairing composed of all pairs in time period i; and Indicates pairing of two different time periods in time period i, using M = {E 1 ,E 2 ,......,E i ,.....,E n} represents the pairing scheme composed of all time period pairings; M a and M b Represents two different pairing schemes, that is, two feasible solutions of the simulated annealing algorithm.
[0044] Further, step S3 includes determining the neighbors of the solution by defining the neighbors of the pairing scheme. For any pairing scheme There is a pairing scheme When satisfied and When M b ∈U(M a ), namely M b M a A neighbor of . a ) is M a 's neighborhood.
[0045] Further, based on the relevant definitions of feasible solutions and feasible solution neighbors, the simulated annealing algorithm includes the following steps:
[0046] S31, initialization parameters, assuming that the initial temperature is T0, the cooling coefficient is α, and the end temperature is T m , the maximum number of iterations is β m , the maximum number of neighborhood searches γ m . Assume that the current temperature T = T0, the current number of iterations β = 1, and the current number of neighborhood searches γ = 0;
[0047] S32, generate an initial solution M0, first pair the parking demand of any time period in the network with the shared parking space, then reasonably combine the time period pairings of all time periods to establish an initial pairing scheme M0, and let the current solution M cur and the optimal solution M best Equal to M0;
[0048] S33, neighborhood search to obtain M cur Neighbor M cur2 , randomly generate a time period pairing on a randomly selected time period i Then, use Replace the current pairing scheme M cur Time period pairing within time period i Get M cur A neighbor of Mcur2 ;
[0049] S34, if the number of car moves in the new pairing scheme is S cur2 Less than the number of car moves S of the current pairing solution cur , then accept the new solution and update the current solution to the new solution; otherwise, calculate the increment ΔS = S cur2 -S cur The value of and generates a random number δ in [0,1]; when δ≤e -Δs / T When , update the current solution to the new solution, otherwise the current solution remains unchanged;
[0050] S35, determine whether to cool down, if γ>γ m , then the temperature drops; let T = T0(1-α) n , β=β+1, γ=0, otherwise, set γ=γ+1 and go to step S33;
[0051] S36, determine whether to terminate the calculation. When the termination temperature Tm is reached, the number of times the new solution is not selected reaches a given value, or the number of car moves of the current pairing solution is 0, output the current pairing solution and end the calculation; otherwise, go to step S33. Specific embodiment 1
[0053] Step 1: Based on the characteristics of "autonomous shifting" of unmanned vehicles, analyze the impact of unmanned vehicles on shared parking, and propose relevant assumptions needed to design the supply and demand matching method. 1) The parking demand time and parking space supply time of shared parking are known. 2) It is assumed that the parking demand at any time is less than or equal to the parking space supply. 3) Manned vehicles can only park in one parking space during their parking demand time; unmanned vehicles can park in different parking spaces in two consecutive time periods, that is, autonomous shifting is achieved. 4) Shifting only occurs at the first and last moments of parking demand and supply, and it is assumed that the cost of each shift is the same.
[0054] Step 2: Divide the parking demand time and parking space supply time into smaller time periods to describe the displacement of driverless vehicles during shared parking. When the parking demand at each moment is met, all parking demands will be met. v,s represents the starting time of vehicle parking demand, t v,e represents the end time of vehicle parking demand, t p,s Indicates the start time of the parking space opening, t p,e Indicates the end time of the parking space opening; the method of dividing the parking demand time and the parking space supply time is as follows:
[0055] Step 2.1: Set the start and end times t of all parking demands and parking supply v,s , t v,e , t p,s , tp,e Put in a collection.
[0056] Step 2.2: Remove identical elements from the set and arrange the remaining elements in ascending order to obtain an ordered set of start and end times.
[0057] Step 2.3: Split parking demand time and parking space supply time. Compared with the above set, all parking demand time and parking space supply time are split into shorter time periods using the time nodes in the set as breakpoints.
[0058] Step 3: According to the spatiotemporal matching characteristics of parking demand and parking space supply of manned and unmanned vehicles in the shared parking process, a targeted simulated annealing algorithm is designed on the basis of time segmentation to determine the demand and parking space matching method that can meet the parking demand with the least number of vehicle moves.
[0059] First, define the feasible solution of the simulated annealing algorithm for this problem. Let e(v, p, i) represent a pairing of vehicle v parked in parking space p during period i; e(0, p, k) represents that there is no vehicle parked in parking space p during period i. Since the demand at any moment is less than the supply, there is no situation where there is no parking space for the vehicle.
[0060] E i =(e1, ..., e j , ..., e l ) represents the time period pairing composed of all pairs in time period i, l represents the number of pairs in time period i; and Indicates pairing of two different time periods in time period i.
[0061] M(V,P)={E 1 , E 2 , ..., E i , ..., E n} represents a pairing scheme (parking scheme) consisting of all time period pairs;
[0062] M a and M b Represents two different pairing schemes, that is, two feasible solutions of the simulated annealing algorithm.
[0063] For example, the opening hours of the 8 shared parking spaces are p1 (7:30-12:00, 13:00-16:30), p2 (6:00-11:30), p3 (8:30-14:15, 16:00-20:00), p4 (7:15-11:45, 13:30-18:15), p5 (9:30-16:00, 16:45-19:30), p6 (6:45-12:45, 14:45-18:15), p7 (9:00-13:45, 15:00-18:30), and p8 (8:00-15:15, 6:00-18:45). The parking demand time of 12 vehicles is v1 (8:45-11:30), v2 (13:15-18:30), v3 (7:30-14:30), v4 (7:00-12:30), v5 (8:30-13:30), v6 (16:00-19:45), v7 (6:45-11:00), v8 (15:45-18:00), v9 (15:00-19:30), v 10 (7:15-11:30),v 11 (9:30-12:30), v 12 (9:00-16:30). From the above data, we can see that the parking space opening time at each moment is greater than the parking demand time. If all 12 vehicles are manned vehicles, the parking needs of 3 vehicles will not be met; if all are driverless vehicles, using the shared parking demand and parking space time-space matching method for driverless vehicles proposed in this paper, all parking needs can be met after 4 moves. The visual matching scheme is as follows Figure 2 shown.
[0064] Next, the neighborhood of the solution is determined by defining the neighborhood of the pairing scheme. a The pairing scheme with only one time period pairing difference is called M a Neighbors, all with M a There is only one pairing scheme that pairs at different time periods, which constitutes M a That is, for any pairing scheme There is a pairing scheme Make and Then M b ∈U(M a ), namely M b M a A neighbor of . a ) is M a 's neighborhood.
[0065] The simulated annealing algorithm for the spatiotemporal matching of shared parking demand and parking spaces for driverless vehicles is as follows:
[0066] S31: Initialize parameters. Assume that the initial temperature is T0, the cooling coefficient is α, and the end temperature is T m , the maximum number of iterations is β m , the maximum number of neighborhood searches γ m Assume that the current temperature T=T0, the current number of iterations β=1, and the current number of neighborhood searches γ=0.
[0067] S32: Generate an initial solution M0. First, pair the parking demand of any time period in the network with the shared parking space, and then reasonably combine the time period pairings of all time periods to establish an initial pairing solution M0. Let the current solution M cur and the optimal solution M best Equal to M0.
[0068] S33: Neighborhood search to obtain M cur Neighbor M cur2 By changing a time period pairing in the current pairing scheme, a neighbor of the current pairing scheme is obtained. The specific method is as follows: 1) Randomly generate a time period pairing on a randomly selected time period i 2) Use Replace the current pairing scheme M cur Time period pairing for time period i Get M cur A neighbor of M cur2 .
[0069] S34: Update the current solution. If the number of car moves in the new pairing solution is S cur2 Less than the number of car moves S of the current pairing solution cur , then accept the new solution and update the current solution to the new solution; otherwise, calculate the increment ΔS = S cur2 -S cur The value of and generates a random number δ in [0, 1]; when δ≤e -ΔS / T When , the current solution is updated to a new solution, so that the algorithm converges to the global optimal solution with a certain probability; otherwise, the current solution remains unchanged.
[0070] S35: Determine whether to cool down. If γ>γ m , then the temperature drops; let T = T0(1-α) n , β=β+1, γ=0. Otherwise, let γ=γ+1 and go to S33.
[0071] S36: Determine whether to terminate the calculation. m When the number of times that the new solution is not selected reaches a given value or the number of car moves of the current pairing solution is 0, the current pairing solution is output and the calculation ends; otherwise, go to step S33.
[0072] The number of devices and processing scales described here are used to simplify the description of the present invention, and the application, modification and variation of the present invention will be obvious to those skilled in the art.
[0073] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and implementation modes. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.
Claims
1. A method for spatiotemporal matching of shared parking demand and parking space for unmanned driving, characterized in that: The following steps are involved: S1. Propose the relevant assumptions needed to design the supply-demand matching method; S2, time segmentation, by dividing the parking demand time and parking space supply time into smaller time periods, characterizing the autonomous shifting of unmanned vehicles during shared parking; the time segmentation in step S2 includes the following steps: S21, putting the start time and end time of parking demand and parking space supply into one set; S22, remove the same elements in the set, and arrange the remaining elements in ascending order to obtain an ordered set; S23, using the time nodes in the set as breakpoints, dividing all parking demand times and parking space supply times; S3. Based on the spatiotemporal matching characteristics of parking demand and parking space supply of manned and unmanned vehicles in the shared parking process, a targeted simulated annealing algorithm is designed on the basis of time segmentation to determine the supply and demand matching method that meets the parking demand with the least number of vehicle moves; Step S3 includes defining a feasible solution of the simulated annealing algorithm by using E i =(e1,ej,......,el) represents the time period pairing composed of all pairs in time period i; and Indicates pairing of two different time periods in time period i, using M = {E 1 ,E 2 ,......,E i ,.....,E n } represents the pairing scheme composed of all time period pairings; M a and M b represents two different pairing schemes, i.e., two feasible solutions of the simulated annealing algorithm; By defining the neighbors of the pairing scheme, the neighbors of the solution are determined. For any pairing scheme There is a pairing solution When satisfied and When M b ∈U(M a ), namely M b M a A neighbor of a ) is M a Neighborhood of; Based on the relevant definitions of feasible solutions and feasible solution neighbors, the simulated annealing algorithm includes the following steps: S31, initialization parameters, assuming that the initial temperature is T0, the cooling coefficient is α, and the end temperature is T m , the maximum number of iterations is β m , the maximum number of neighborhood searches γ m, Assume the current temperature T = T0, the current number of iterations β = 1, and the current number of neighborhood searches γ = 0; S32, generate an initial solution M0, first pair the parking demand of any time period in the network with the shared parking space, then reasonably combine the time period pairings of all time periods to establish an initial pairing scheme M0, and let the current solution M cur and the optimal solution M best Equal to M0; S33, neighborhood search to obtain M cur Neighbor M cur2 , randomly generate a time period pairing on a randomly selected time period i Then, use Replace the current pairing scheme M cur Time period pairing within time period i Get M cur A neighbor of M cur2 ; S34, if the number of car moves in the new pairing scheme is S cur2 Less than the number of car moves S of the current pairing solution cur , then accept the new solution and update the current solution to the new solution; otherwise, calculate the increment ΔS = S cur2 -S cur The value of and generates a random number δ in [0,1]; when δ≤e -Δs / T When , update the current solution to the new solution, otherwise the current solution remains unchanged; S35, determine whether to cool down, if γ>γ m , then the temperature drops; let T = T0(1-α) n , β=β+1, γ=0, otherwise, set γ=γ+1 and go to step S33; S36, determine whether to terminate the calculation; when the termination temperature T is reached m When the number of times that the new solution is not selected reaches a given value or the number of car moves of the current pairing solution is 0, the current pairing solution is output and the calculation ends; otherwise, go to step S33.
2. A method for spatiotemporal matching of shared parking demand and parking space for unmanned driving as claimed in claim 1, characterized in that: The relevant assumptions in step S1 include: (1) The parking demand time and parking space supply time of shared parking are known; (2) To ensure that all parking demands are met, it is assumed that the parking demand at any moment is less than or equal to the parking space supply; (3) A manned vehicle must park in one parking space during the parking demand period; while an unmanned vehicle can park in different parking spaces during two consecutive parking demand periods; (4) Assume that displacement only occurs at the beginning and end of parking demand and supply, and assume that the cost of each displacement is the same.
