CN-122024515-A - Vehicle scheduling method, device, medium and light quantum computer

CN122024515ACN 122024515 ACN122024515 ACN 122024515ACN-122024515-A

Abstract

The invention provides a vehicle dispatching method, a device, a medium and a light quantum computer, wherein dispatching reference data are acquired, and then a plurality of candidate dispatching schemes are determined according to the dispatching reference data, and in each candidate dispatching scheme, a first time required by each vehicle to be lighter to reach a corresponding parking space and a second time required by each vehicle to be lighter to be parked into the corresponding parking space are determined. And determining the total cost of each candidate scheduling scheme according to the first time and the second time corresponding to the plurality of vehicles to be lighter in each candidate scheduling scheme. And determining a target scheduling scheme from the plurality of candidate scheduling schemes according to the total cost of each candidate scheduling scheme, and guiding a plurality of vehicles to be lightered to park in corresponding parking spaces based on the target scheduling scheme. By introducing the factor of the parking time of the vehicle into the selection of the path scheme, traffic jam caused by the parking operation is reduced, the smoothness of traffic flow in the parking lot is ensured, and the space utilization rate and turnover rate of the parking lot are improved.

Inventors

Kang Fengshou
YANG LIN
ZHANG LIBO

Assignees

上海图灵智算量子科技有限公司
图灵智算量子科技(北京)有限公司

Dates

Publication Date: 20260512
Application Date: 20251203

Claims (10)

1. A vehicle scheduling method, characterized by comprising: Acquiring scheduling reference data, wherein the scheduling reference data comprises initial coordinates of a plurality of vehicles to be lightered and a network diagram of a parking lot, and the vehicles to be lightered represent all vehicles needing to be scheduled to a corresponding parking space from the current position in a current scheduling task; Determining a plurality of candidate scheduling schemes according to the scheduling reference data, and in each candidate scheduling scheme, a first time required by each vehicle to be lightered to reach a corresponding parking space and a second time required by each vehicle to be lightered to be parked to the corresponding parking space, wherein each candidate scheduling scheme comprises a parking space, a traveling path and a traveling sequence corresponding to each vehicle to be lightered; Determining the total cost of each candidate scheduling scheme according to the first time and the second time corresponding to the plurality of vehicles to be rejected in each candidate scheduling scheme; And determining a target scheduling scheme from the plurality of candidate scheduling schemes according to the total cost of each candidate scheduling scheme, and guiding the plurality of vehicles to be lighter to park in the corresponding parking spaces based on the target scheduling scheme.
2. The vehicle scheduling method according to claim 1, characterized in that, before the step of acquiring scheduling reference data, the method further comprises: responding to the received dispatching task, determining a first vehicle to be dispatched and a second vehicle which does not park in a corresponding parking space in the parking lot at the current moment; and determining the plurality of vehicles to be lighter according to the first vehicle and the second vehicle.
3. The vehicle dispatching method according to claim 1, wherein the step of determining the total cost of each candidate dispatching scheme according to the first time and the second time corresponding to the plurality of vehicles to be rejected in the candidate dispatching scheme specifically comprises: Determining operation sets and blocking matrixes respectively corresponding to the plurality of vehicles to be lighter when the vehicles to be lighter are parked in the corresponding parking spaces according to the scheduling reference data, wherein the blocking matrixes are used for representing the blocking degree of each parking operation of the vehicles to be lighter on a road; Determining collision probability corresponding to each candidate scheduling scheme according to the operation set and the blocking matrix; determining the blocking time corresponding to each candidate scheduling scheme according to the blocking matrix and the operation set, and taking the blocking time as the parking cost of the candidate scheduling scheme; determining the total path length of the plurality of vehicles to be rejected corresponding to each candidate scheduling scheme according to the scheduling reference data, and taking the total path length as the management cost of the candidate scheduling scheme; and determining the total cost of the candidate scheduling scheme according to at least one of the collision probability, the parking cost and the management cost corresponding to each candidate scheduling scheme and the first time and the second time corresponding to the plurality of vehicles to be rejected in the candidate scheduling scheme.
4. A vehicle dispatching method according to claim 3, wherein the step of determining, as the management cost of each candidate dispatching scheme, the total path length of the plurality of vehicles to be rejected corresponding to the candidate dispatching scheme according to the dispatching reference data specifically comprises: determining the offline sequence of the plurality of vehicles to be lighter; determining a first cost of each parking space for the vehicle to be lighter according to the offline sequence and the distance between the parking space in the parking lot and the vehicle to be lighter; determining a second cost corresponding to each candidate scheduling scheme according to the comprehensive cost of each parking space for each vehicle to be lightered; determining the path length corresponding to each candidate scheduling scheme according to the scheduling reference data; And determining the management cost of each candidate scheduling scheme according to the second cost and the path length corresponding to the candidate scheduling scheme, wherein the management cost is used for redefining the parking space corresponding to each vehicle to be rejected.
5. The vehicle scheduling method according to any one of claims 1 to 4, wherein the step of determining a target scheduling scheme from the plurality of candidate scheduling schemes according to the total cost of each of the candidate scheduling schemes specifically includes: determining a target scheduling scheme from the plurality of candidate scheduling schemes based on at least one constraint condition according to a total cost of each of the candidate scheduling schemes; Each vehicle to be lighter has one and only one driving path; each parking space is used for parking only one vehicle to be lighter at the same time; When each road in the network diagram is at any moment, the number of vehicles to be refuted running on the road does not exceed the maximum capacity of the road; the departure time of each vehicle waiting for the lighters is not earlier than the offline time of the vehicle waiting for the lighters; When an overlapping area exists in a parking area occupied by any two vehicles to be lighter when parking in a parking space, the parking time of the two vehicles to be lighter is not overlapped; at any moment the vehicle to be refuted does not travel into a road that is completely blocked by the parking maneuver.
6. The vehicle scheduling method of claim 5, wherein the determining a target scheduling scheme from the plurality of candidate scheduling schemes based on at least one constraint condition of total cost of each of the candidate scheduling schemes, specifically comprises: Converting each parameter in the candidate scheduling schemes into a decision variable, converting the total cost into an objective function, converting the constraint condition into a punishment item, and determining a vehicle scheduling model; and inputting the vehicle dispatching model and the dispatching reference data into a quantum computer, and determining a target dispatching scheme according to the output result of the quantum computer.
7. The vehicle scheduling method of claim 1, wherein the method further comprises: Executing the target scheduling scheme in real time through a simulation system; And when the simulation result is different from the estimated result of the target scheduling scheme, re-determining the target scheduling scheme, and guiding the plurality of vehicles to be refused to park in the corresponding parking spaces through the re-determined target scheduling scheme.
8. A vehicle scheduling apparatus, characterized by comprising: The system comprises an acquisition module, a scheduling reference module and a scheduling control module, wherein the scheduling reference data comprises initial coordinates of a plurality of vehicles to be lighter and a network diagram of a parking lot, and the vehicles to be lighter represent all vehicles needing to be scheduled to a corresponding parking lot from the current position in the current scheduling task; The generation module is used for determining a plurality of candidate scheduling schemes according to the scheduling reference data, and a first time required by each vehicle to be lightered to reach a corresponding parking space and a second time required by each vehicle to be lightered to be parked to the corresponding parking space in each candidate scheduling scheme, wherein each candidate scheduling scheme comprises a parking space, a driving path and a driving sequence corresponding to each vehicle to be lightered; The cost module is used for determining the total cost of each candidate scheduling scheme according to the first time and the second time corresponding to the plurality of vehicles to be refused in each candidate scheduling scheme; And the determining module is used for determining a target scheduling scheme from the plurality of candidate scheduling schemes according to the total cost of each candidate scheduling scheme, and guiding the plurality of vehicles to be lighter to park in the corresponding parking spaces based on the target scheduling scheme.
9. A light quantum computer comprising a quantum portion and a classical computing portion, wherein: The classical computing part is used for converting a vehicle dispatching model constructed based on the vehicle dispatching method of any one of the claims 1-7 into a quadratic unconstrained binary optimization QUBO model; The quantum part is used for outputting a binary bit string according to the QUBO model; the classical calculation part is used for decoding the binary bit string and determining a target scheduling scheme.
10. A computer-readable storage medium storing a computer program for executing the vehicle scheduling method according to any one of claims 1 to 7.

Description

Vehicle scheduling method, device, medium and light quantum computer Description of the division The application relates to a method, a device, a medium and a light quantum computer for dispatching vehicles, which are divided into patent applications of China, wherein the application date is 2025, 12, 3, and the application number is CN 202511804478.4. Technical Field The invention relates to the field of logistics system optimization, in particular to a vehicle scheduling method, a vehicle scheduling device, a vehicle scheduling medium and a light quantum computer. Background With the development of the automobile manufacturing industry, the number of new vehicles which are assembled and driven off from a production line is increased, the types and the configurations of the vehicles which are driven off from the production line are also diversified, and the management complexity of a parking lot is increased sharply. In the prior art, a parking path is planned for a vehicle which goes offline usually depending on manual experience or a preset scheduling rule, but the existing scheduling method only pays attention to the running process of the vehicle on the path, and ignores the blockage of the parking operation to the road after the vehicle reaches a target parking place, so that the space utilization rate and turnover rate of the parking lot are low. Disclosure of Invention In view of the above, the embodiments of the present invention are directed to providing a vehicle dispatching method, device, medium and light quantum computer, so as to solve the problem that the existing dispatching method only focuses on the running process of the vehicle on the path, but ignores the problem that the space utilization rate and turnover rate of the parking lot are low due to the blockage of the parking operation to the road after the vehicle reaches the target parking space. In one aspect, the present invention provides a vehicle scheduling method, including: Acquiring scheduling reference data, wherein the scheduling reference data comprises initial coordinates of a plurality of vehicles to be lightered and a network diagram of the parking lot, and the vehicles to be lightered represent all vehicles needing to be scheduled to a corresponding parking space from the current position in the current scheduling task; Determining a plurality of candidate scheduling schemes according to the scheduling reference data, and in each candidate scheduling scheme, a first time required by each vehicle to be lighter to reach the corresponding parking space and a second time required by each vehicle to be lighter to be parked into the corresponding parking space, wherein each candidate scheduling scheme comprises a parking space, a driving path and a driving sequence corresponding to each vehicle to be lighter; Determining the total cost of each candidate scheduling scheme according to the first time and the second time corresponding to the plurality of vehicles to be rejected in each candidate scheduling scheme; And determining a target scheduling scheme from the plurality of candidate scheduling schemes according to the total cost of each candidate scheduling scheme, and guiding the plurality of vehicles to be lighter to park in the corresponding parking spaces based on the target scheduling scheme. In some embodiments of the application, a first vehicle to be dispatched and a second vehicle which does not park in a corresponding parking space in the parking lot at the current moment are determined in response to a received dispatching task, and the plurality of vehicles to be rejected are determined according to the first vehicle and the second vehicle. In some embodiments of the present application, the step of determining the total cost of the candidate scheduling scheme according to the first time and the second time corresponding to the plurality of vehicles to be refuted in each candidate scheduling scheme specifically includes determining, according to the scheduling reference data, an operation set and a blocking matrix respectively corresponding to the plurality of vehicles to be refuted when the plurality of vehicles to be refuted are parked in the corresponding parking spaces, wherein the blocking matrix is used for representing the blocking degree of each parking operation of the vehicles to be refuted on a road, determining, according to the operation set and the blocking matrix, a collision probability corresponding to each candidate scheduling scheme, determining, according to the blocking matrix and the operation set, a blocking time corresponding to each candidate scheduling scheme as a parking cost of the candidate scheduling scheme, determining, according to the scheduling reference data, a total path length of the plurality of vehicles to be refuted corresponding to each candidate scheduling scheme as a management cost of the candidate scheduling scheme, determining, according to the collision probability co