CN110298512B - Driving scheme optimization method, load balance analyzer and intelligent driving optimization system - Google Patents

Driving scheme optimization method, load balance analyzer and intelligent driving optimization system Download PDF

Info

Publication number
CN110298512B
CN110298512B CN201910589318.0A CN201910589318A CN110298512B CN 110298512 B CN110298512 B CN 110298512B CN 201910589318 A CN201910589318 A CN 201910589318A CN 110298512 B CN110298512 B CN 110298512B
Authority
CN
China
Prior art keywords
driving
scheme
optimization
terminal
driving scheme
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910589318.0A
Other languages
Chinese (zh)
Other versions
CN110298512A (en
Inventor
王智明
徐雷
陶冶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China United Network Communications Group Co Ltd
Original Assignee
China United Network Communications Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China United Network Communications Group Co Ltd filed Critical China United Network Communications Group Co Ltd
Priority to CN201910589318.0A priority Critical patent/CN110298512B/en
Publication of CN110298512A publication Critical patent/CN110298512A/en
Application granted granted Critical
Publication of CN110298512B publication Critical patent/CN110298512B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q50/40—
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count

Abstract

The application discloses a driving scheme optimization method, a load balance analyzer and an intelligent driving optimization system, wherein the method comprises the following steps: receiving driving requirements sent by each driving terminal; determining each driving scheme corresponding to each driving terminal according to each driving requirement and a depth analysis method; judging whether the information data of each driving scheme meets preset evaluation conditions or not; and if the information data of at least one driving scheme does not meet the evaluation conditions, respectively carrying out optimization analysis on the driving schemes which do not meet the evaluation conditions in an iterative mode according to a depth analysis method until each optimized driving scheme meets the evaluation conditions. The finally obtained driving scheme has the effects of high load balance degree of a traffic hub and contribution to realizing navigation intelligentization and three-dimensional multi-layer type traffic tool route switching of the traffic tool.

Description

Driving scheme optimization method, load balance analyzer and intelligent driving optimization system
Technical Field
The application belongs to the technical field of intelligent driving, and particularly relates to a driving scheme optimization method, a load balance analyzer and an intelligent driving optimization system.
Background
With the rapid growth of 5G and edge cloud load-bearing services, the problems that traffic hub load imbalance, vehicle navigation mostly depends on manual assistance operation, and the switching of routes of planar single-layer vehicles is limited are increasingly prominent. The existing plane single-layer type traffic system does not fully consider the problems that traffic hub loads are unbalanced, vehicle navigation depends on manual auxiliary operation, and the route switching of the plane single-layer type vehicle is limited.
Disclosure of Invention
The present application aims to solve at least one of the technical problems in the prior art, and provides a driving scheme optimization method and device.
The application provides a driving scheme optimization method, which comprises the following steps:
receiving driving requirements sent by each driving terminal;
determining each driving scheme corresponding to each driving terminal according to each driving requirement and a depth analysis method;
judging whether the information data of each driving scheme meets preset evaluation conditions or not;
and if the information data of at least one driving scheme does not meet the evaluation conditions, respectively carrying out optimization analysis on the driving schemes which do not meet the evaluation conditions in an iterative mode according to a depth analysis method until each optimized driving scheme meets the evaluation conditions.
Preferably, the information data of the driving scenario includes at least: the ratio of the load balance degree of the traffic hub, the energy consumption of the driving terminal and the total mileage is calculated;
each of the driving scenarios is stored in the form of a two-dimensional vector as:
Figure GDA0003058961050000011
wherein k is the current iteration number, (i, j, t) is a three-dimensional coordinate, and i belongs to [1, m ]]，j∈[1,n]，t∈[1,q]The driving scheme corresponds to the driving demand based on the three-dimensional coordinates;
Figure GDA0003058961050000021
and
Figure GDA0003058961050000022
respectively the load balance degree of the traffic hub with three-dimensional coordinates of (i, j, t), the energy consumption of the driving terminal and the energy consumption of the driving terminal in the kth iteration processRatio of total mileage.
Preferably, the step of performing optimization analysis on the driving schemes which do not satisfy the evaluation condition according to a depth analysis method includes:
optimizing the current driving scheme according to a preset unsupervised learning method according to the current driving scheme and the historical driving scheme of the driving terminal to obtain a plurality of optimized driving schemes of the driving terminal;
and selecting one optimized driving scheme meeting preset optimization conditions from the optimized driving schemes.
Preferably, the step of optimizing the current driving scheme according to a preset unsupervised learning method according to the current driving scheme and the historical driving scheme of the driving terminal to obtain a plurality of optimized driving schemes of the driving terminal includes:
calculating the (k + 1) th depth unsupervised learning enhancement factor according to the formula (1)
Figure GDA0003058961050000023
Figure GDA0003058961050000024
Calculating the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) obtained by the (k + 1) th iteration according to the formula (2)
Figure GDA0003058961050000025
Figure GDA0003058961050000026
Wherein the content of the first and second substances,
Figure GDA0003058961050000027
load balance degree of the traffic hub in the current k-th iteration process, CmaxGFor the historical maximum traffic hub load balancing,
Figure GDA0003058961050000028
is the ratio of the energy consumption of the driving terminal to the total mileage in the current k iteration process, WminGIs the ratio of the historical minimum driving terminal energy consumption to the total mileage,
Figure GDA0003058961050000029
and (5) obtaining the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) for the kth iteration.
Preferably, the travel demand includes: a source location, a pathway location, and a target location; selecting an optimized driving scheme meeting preset optimization conditions according to the following formula:
Figure GDA00030589610500000210
Figure GDA0003058961050000031
Figure GDA0003058961050000032
wherein the content of the first and second substances,
Figure GDA0003058961050000033
is a longitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration,
Figure GDA0003058961050000034
For the longitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration,
Figure GDA0003058961050000035
Is the latitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration process,
Figure GDA0003058961050000036
Is the latitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration process,
Figure GDA0003058961050000037
for the longitude value of the source location in the driving demand corresponding to the three-dimensional coordinate (i, j, t) during the kth iteration,
Figure GDA0003058961050000038
is the latitude value of the source position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k iteration process.
Preferably, the evaluation condition includes:
Figure GDA0003058961050000039
wherein i belongs to [1, m ], j belongs to [1, n ], t belongs to [1, q ].
The present application further provides a load balancing analyzer, comprising:
the receiving module is used for receiving the driving requirements sent by each driving terminal;
the determining module is used for determining each driving scheme corresponding to each driving terminal according to each driving requirement and a depth analysis method;
the judging module is used for judging whether the information data of each driving scheme meets preset evaluation conditions or not;
and the optimization module is used for respectively carrying out optimization analysis on the running schemes which do not meet the evaluation conditions according to a depth analysis method in an iterative mode until each optimized running scheme meets the evaluation conditions if the information data of at least one running scheme does not meet the evaluation conditions.
The present application further provides an intelligent driving optimization system, including:
the load balancing analyzer described above;
the driving terminals are used for sending driving demands to the load balance analyzer; and also for receiving a driving scheme.
Preferably, the system further comprises:
and the edge server is used for controlling the driving terminal to drive according to the driving scheme.
Preferably, the system further comprises: a network transmission unit for transmitting the driving demand; and/or for transmitting the driving profile.
In the driving scheme optimization method provided in this embodiment, a plurality of driving schemes are determined based on driving requirements sent by each driving terminal, whether each current driving scheme can be output is determined according to information data in each driving scheme, and when a driving scheme which does not satisfy an output condition (that is, an evaluation condition) exists, a final driving scheme corresponding to each driving terminal which does not satisfy the evaluation condition is obtained by performing optimization analysis on the driving scheme which does not satisfy the evaluation condition. Therefore, the finally obtained driving scheme has the effects of high load balance degree of the traffic hub and contribution to the realization of route switching of the intelligent and three-dimensional multi-layer type traffic tool for the navigation of the traffic tool.
Drawings
Fig. 1 is a flowchart of an intelligent driving optimization method according to a first embodiment of the present disclosure;
fig. 2 is a schematic structural diagram of a load balancing analyzer according to a second embodiment of the present application.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The application provides an intelligent driving optimization method, device and system. The following detailed description is made with reference to the drawings of the embodiments provided in the present application, respectively.
The present application is based on an intelligent driving optimization system implementation, and the following principles simply receive the intelligent driving optimization system implementation. The intelligent driving optimization system comprises a load balance analyzer and a plurality of driving terminals, wherein the load balance analyzer is used for receiving the driving requirements of the driving terminals, analyzing according to the driving requirements of the driving terminals to obtain a final driving scheme, and then returning the final driving scheme to the driving terminals. And the driving terminal is used for sending the driving demand to the load balance analyzer and receiving the driving scheme.
In a preferred mode, the system further includes: and the edge server is used for controlling the driving terminal to drive according to the driving scheme. The edge server can be arranged at the local of the driving terminal and can control the driving state of the driving terminal according to the driving scheme received by the driving terminal. In this application, the driving scheme may be transmitted to the edge server by the driving terminal, or the driving scheme may be directly transmitted to the edge server by the load balancing analyzer.
An intelligent driving optimization method provided by a first embodiment of the application is as follows:
in this embodiment, the driving terminal may be an automobile, a navigator, or the like, and is preferably an autonomous automobile. In the present embodiment, a driving terminal is taken as an example of an autonomous vehicle to be specifically described. The execution subject of the present embodiment is a load balancing analyzer. As shown in fig. 1, which shows a flowchart of an intelligent driving optimization method provided in an embodiment of the present application, the method includes the following steps.
Step S101, receiving the driving demands sent by each driving terminal.
The driving requirement may be a path planning request, and specifically may include a source location, a target location, a route location, and the like.
It is understood that in the present embodiment, there may be a plurality of driving terminals.
In this embodiment, the driving requirement may be actively reported by the driving terminal, or the load balancing analyzer periodically inquires the driving terminal for obtaining the driving requirement according to a preset periodic inquiry mechanism.
And S102, determining each driving scheme corresponding to each driving terminal according to each driving requirement and a depth analysis method.
In this step, corresponding driving schemes are respectively made according to the driving requirements of the driving terminals. Different driving schemes can be given for the same driving requirements, and the specific paths given by the driving schemes are different, so that the information data such as the load balance degree of the corresponding traffic hub, the ratio of the energy consumption of the driving terminal to the total mileage and the like are also different. That is, each driving scheme has information data such as load balance degree of a specific traffic junction, ratio of energy consumption of a driving terminal to total mileage and the like, so that each driving scheme has respective advantages and disadvantages. In the present embodiment, the running course is evaluated based on these pieces of information data of the running course to find a running course that is suitable for the driver's terminal to execute.
Optimally, the information data of the driving scenario comprise at least: the ratio of the load balance degree of the traffic hub, the energy consumption of the driving terminal and the total mileage is calculated; each of the driving scenarios is stored in the form of a two-dimensional vector as:
Figure GDA0003058961050000051
wherein k is the current iteration number, (i, j, t) is a three-dimensional coordinate, and i belongs to [1, m ]]，j∈[1,n]，t∈[1,q]The driving scheme corresponds to the driving demand based on the three-dimensional coordinates;
Figure GDA0003058961050000061
and
Figure GDA0003058961050000062
in the k-th iteration process, the three-dimensional coordinates are the load balance degree of the traffic hub and the ratio of the energy consumption of the driving terminal to the total mileage of (i, j, t). And the load balance degree of each transportation junction is equal to the sum of the load weights of the driving terminals at a certain moment of each transportation junction/the number of the driving terminals at a certain moment.
Step S103, judging whether the information data of each driving scheme meets a preset evaluation condition, and if the information data of at least one driving scheme does not meet the evaluation condition, executing step S104; and if the information data of at least one driving scheme does not meet the evaluation condition, ending the process.
In this embodiment, since the plurality of driving terminals are located in the same large area, and the driving plans corresponding to the driving demands of the different driving terminals may collide, there may be a problem that the loads are unbalanced when the vehicles actually drive according to the driving plans.
Therefore, in this step, the driving data of each driving scheme is evaluated comprehensively, whether the information data of each driving scheme meets the preset evaluation condition is judged, if at least one piece of information data of the driving scheme does not meet the evaluation condition, the whole driving scheme is not feasible, the driving scheme not meeting the evaluation condition needs to be optimized, and the driving scheme meeting the evaluation condition is sent to the corresponding driving terminal. And when the driving scheme does not exist, namely each driving scheme simultaneously meets the preset evaluation condition, the whole driving scheme is feasible, and at the moment, each current driving scheme can be sent to the corresponding driving terminal.
Optimally, the evaluation conditions include:
Figure GDA0003058961050000063
wherein i ∈ [1, m ]]，j∈[1,n]，t∈[1,q]，
Figure GDA0003058961050000064
And
Figure GDA0003058961050000065
in the k-th iteration process, the three-dimensional coordinates are the load balance degree of the traffic hub and the ratio of the energy consumption of the driving terminal to the total mileage of (i, j, t). In the driving scheme of each driving terminal
Figure GDA0003058961050000066
And
Figure GDA0003058961050000067
the formula is brought into the evaluation condition, whether inequality is established or not is judged, and if the inequality is established, the driving scheme is judged to meet the evaluationConditions; and if the inequality is not satisfied, judging that the running scheme does not meet the evaluation condition.
And step S104, respectively carrying out optimization analysis on the driving schemes which do not meet the evaluation conditions according to a depth analysis method in an iterative mode until each optimized driving scheme meets the evaluation conditions.
In the step, optimization analysis is respectively carried out on the driving schemes which do not meet the evaluation conditions according to a depth analysis method in an iteration mode until each optimized driving scheme meets the evaluation conditions.
It should be noted that the load balancing analyzer optimizes the driving scheme of each driving terminal that does not satisfy the evaluation condition, obtains a final driving scheme for each driving terminal that does not satisfy the evaluation condition, and sends the driving scheme to the corresponding driving terminal. The iteration mode specifically comprises a step a and a step b.
And a, respectively carrying out optimization analysis on each driving scheme according to a depth analysis method, and adding 1 to the current iteration number k.
In this step, when it is determined that there is at least one driving scheme whose information data does not satisfy the preset evaluation condition, the driving schemes need to be optimized, and 1 is added to the current iteration number k every time the driving schemes are optimized. Each optimization process is performed on a plurality of driving plans of the driving terminals which do not satisfy the evaluation condition, and the optimized driving plans are also a plurality of driving plans.
Preferably, step S104 includes step 1) and step 2):
step 1), according to a current driving scheme and a historical driving scheme of a driving terminal, optimizing the current driving scheme according to a preset unsupervised learning method to obtain a plurality of optimized driving schemes of the driving terminal.
Optimally, the step of optimizing the current driving scheme according to a preset unsupervised learning method according to the current driving scheme and the historical driving scheme of the driving terminal to obtain a plurality of optimized driving schemes of the driving terminal comprises the following steps:
calculating the (k + 1) th depth unsupervised learning enhancement factor according to the formula (1)
Figure GDA0003058961050000071
Figure GDA0003058961050000072
Calculating the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) obtained by the (k + 1) th iteration according to the formula (2)
Figure GDA0003058961050000073
Figure GDA0003058961050000074
Wherein the content of the first and second substances,
Figure GDA0003058961050000075
load balance degree of the traffic hub in the current k-th iteration process, CmaxGFor the historical maximum traffic hub load balancing,
Figure GDA0003058961050000076
is the ratio of the energy consumption of the driving terminal to the total mileage in the current k iteration process, WminGIs the ratio of the historical minimum driving terminal energy consumption to the total mileage,
Figure GDA0003058961050000077
and (5) obtaining the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) for the kth iteration.
According to the formula of the unsupervised learning method, aiming at the driving scheme of one driving terminal, a plurality of different optimized driving schemes can be obtained based on different values of delta and mu.
And 2) selecting an optimized driving scheme meeting preset optimization conditions from the optimized driving schemes.
It can be understood that each driving scheme which does not satisfy the evaluation condition obtains a plurality of optimized driving schemes, that is, the driving requirement of each driving terminal corresponds to a plurality of optimized driving schemes. In practice, only one of the driving schemes needs to be selected for each driving demand. In this step, one optimized driving scheme corresponding to each driving demand is selected from the optimized driving schemes in step 1) according to preset optimization conditions as a final driving scheme of the driving terminal.
Optimally, the driving demand comprises: a source location, a pathway location, and a target location; selecting an optimized driving scheme meeting preset optimization conditions according to the following formula:
Figure GDA0003058961050000081
Figure GDA0003058961050000082
Figure GDA0003058961050000083
wherein the content of the first and second substances,
Figure GDA0003058961050000084
is a longitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration,
Figure GDA0003058961050000085
For the longitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration,
Figure GDA0003058961050000086
Is the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration processThe latitude value of the target position,
Figure GDA0003058961050000087
Is the latitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration process,
Figure GDA0003058961050000088
for the longitude value of the source location in the driving demand corresponding to the three-dimensional coordinate (i, j, t) during the kth iteration,
Figure GDA0003058961050000089
is the latitude value of the source position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k iteration process.
As can be seen from the above formula of the optimization conditions, each optimized driving scheme is substituted into the formula, and a minimum value can be calculated, and the driving scheme corresponding to the minimum value is the final driving scheme of the driving terminal.
B, judging whether the current iteration times k are smaller than or equal to a preset threshold value, if so, judging whether the information data of each driving scheme meets preset evaluation conditions; otherwise, the iteration is ended.
In the step, before the current driving scheme of the driving terminal is evaluated again, the current iteration number k is judged, and when the current iteration number reaches a preset threshold value, the selected recommendation scheme is considered to be infinitely close to meeting the first evaluation condition, so that even if the recommendation scheme cannot meet the first evaluation condition, the recommendation scheme can be output, so that the problem that iterative optimization is carried out infinitely, and the waste of computing resources is caused is avoided. The iteration number k needs to satisfy the condition that k is 1,2, …, d, wherein d is preferably 50. When the iteration number is less than or equal to the preset threshold, returning to step S103 to evaluate each current driving scenario again, that is, determining whether the information data of each driving scenario satisfies the preset evaluation condition, so as to determine that there is at least one driving scenario whose information data does not satisfy the preset evaluation condition, and then optimizing the driving scenario again. By repeating the iteration once, the driving schemes which do not satisfy the evaluation condition are continuously optimized, and finally, the optimal driving schemes are output.
In the driving scheme optimization method provided by this embodiment, a plurality of driving schemes are determined based on driving requirements sent by each driving terminal, whether each current driving scheme can be output is determined according to information data in each driving scheme, and when a driving scheme which does not satisfy an output condition (that is, an evaluation condition) exists, a final driving scheme corresponding to each driving terminal which does not satisfy the evaluation condition is obtained by performing one or more iterative optimization analyses on the driving scheme which does not satisfy the evaluation condition. In the optimization process of the driving scheme, the method of single-source shortest path passing through multiple intermediate nodes, weighted constraint, deep unsupervised learning and the like is combined, so that the finally obtained driving scheme has high load balance degree of a traffic hub, and the effects of intelligent navigation and three-dimensional multi-layer type vehicle route switching of a vehicle are facilitated.
A load balancing analyzer provided in a second embodiment of the present application is as follows:
fig. 2 is a schematic structural diagram illustrating a load balancing analyzer according to an embodiment of the present application, and includes the following modules.
The receiving module 11 is used for receiving the driving demands sent by each driving terminal;
the first determining module 12 is configured to determine, according to each driving requirement and a depth analysis method, each driving scheme corresponding to each driving terminal;
the judging module 13 is used for judging whether the information data of each driving scheme meets preset evaluation conditions;
and the optimization module 14 is configured to, if the information data of at least one driving scheme does not satisfy the evaluation condition, perform optimization analysis on the driving schemes that do not satisfy the evaluation condition in an iterative manner according to a depth analysis method until each optimized driving scheme satisfies the evaluation condition.
Preferably, the information data of the driving scenario includes at least: the ratio of the load balance degree of the traffic hub, the energy consumption of the driving terminal and the total mileage is calculated;
each of the driving scenarios is stored in the form of a two-dimensional vector as:
Figure GDA0003058961050000101
wherein k is the current iteration number, (i, j, t) is a three-dimensional coordinate, and i belongs to [1, m ]]，j∈[1,n]，t∈[1,q]The driving scheme corresponds to the driving demand based on the three-dimensional coordinates;
Figure GDA0003058961050000102
and
Figure GDA0003058961050000103
in the k-th iteration process, the three-dimensional coordinates are the load balance degree of the traffic hub and the ratio of the energy consumption of the driving terminal to the total mileage of (i, j, t).
Preferably, the optimization module includes:
the first optimization submodule is used for optimizing the current driving scheme according to the current driving scheme and the historical driving scheme of the driving terminal and a preset unsupervised learning method to obtain a plurality of optimized driving schemes of the driving terminal;
and the second optimization submodule is used for selecting an optimized driving scheme meeting preset optimization conditions from the optimized driving schemes.
Preferably, the first optimization submodule is specifically configured to:
calculating the (k + 1) th depth unsupervised learning enhancement factor according to the formula (1)
Figure GDA0003058961050000104
Figure GDA0003058961050000105
Calculating the optimized corresponding three-dimensional coordinate (i, j, t) obtained by the (k + 1) th iteration according to the formula (2)) Of the driving scheme
Figure GDA0003058961050000106
Figure GDA0003058961050000107
Wherein the content of the first and second substances,
Figure GDA0003058961050000108
load balance degree of the traffic hub in the current k-th iteration process, CmaxGFor the historical maximum traffic hub load balancing,
Figure GDA0003058961050000109
is the ratio of the energy consumption of the driving terminal to the total mileage in the current k iteration process, WminGIs the ratio of the historical minimum driving terminal energy consumption to the total mileage,
Figure GDA0003058961050000111
and (5) obtaining the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) for the kth iteration.
Preferably, the travel demand includes: a source location, a pathway location, and a target location; selecting an optimized driving scheme meeting preset optimization conditions according to the following formula:
Figure GDA0003058961050000112
Figure GDA0003058961050000113
Figure GDA0003058961050000114
wherein the minimum ZkThe corresponding scheme is the optimized driving scheme,
Figure GDA0003058961050000115
is a longitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration,
Figure GDA0003058961050000116
For the longitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration,
Figure GDA0003058961050000117
Is the latitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration process,
Figure GDA0003058961050000118
Is the latitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration process,
Figure GDA0003058961050000119
for the longitude value of the source location in the driving demand corresponding to the three-dimensional coordinate (i, j, t) during the kth iteration,
Figure GDA00030589610500001110
is the latitude value of the source position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k iteration process.
Preferably, the evaluation condition includes:
Figure GDA00030589610500001111
wherein i belongs to [1, m ], j belongs to [1, n ], t belongs to [1, q ].
An intelligent driving optimization system provided by a third embodiment of the present application is as follows:
the intelligent driving optimization system provided by the embodiment of the application comprises: the load balancing analyzer comprises a load balancing analyzer and a plurality of driving terminals, wherein the driving terminals are used for sending driving demands to the load balancing analyzer; and also for receiving a driving scheme.
Preferably, the system further comprises:
and the edge server is used for controlling the driving terminal to drive according to the driving scheme.
The edge server can be arranged at the local driving terminal and can control the driving state of the driving terminal according to the driving scheme received by the driving terminal. In this embodiment, the driving terminal may transmit the driving scheme to the edge server, or the load balancing analyzer may directly transmit the driving scheme to the edge server.
Preferably, the system further comprises: a network transmission unit for transmitting the driving demand; and/or for transmitting the driving profile. And the system is used for transmitting the driving demand sent by the driving terminal to the load balance analyzer through the network and transmitting the optimized driving scheme to the driving terminal. The network transmission unit may specifically include: operator base stations, satellites, etc.
Further, the intelligent driving optimization system further comprises: and the gateway unit can comprise a plurality of traffic gateways and is used for ensuring the safety of network transmission in the driving scheme optimization system.
In the driving scheme optimization system provided in this embodiment, after the driving terminals send the driving demands, the driving demands are sent to the load balancing analyzer through the network transmission unit, and the load balancing analyzer determines the driving schemes corresponding to the driving demands and having high overall feasibility based on the driving demands sent by the driving terminals, and returns the driving schemes to the corresponding driving terminals through the network transmission unit. In the process of determining the optimal driving scheme, the load balance analyzer determines the optimal driving scheme by performing one or more times of iterative optimization analysis on the driving schemes, so that the finally obtained driving scheme has high load balance degree of the traffic hub, and the method is favorable for realizing the effects of intelligent navigation and three-dimensional multi-layer vehicle route switching of the vehicle.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.

Claims (7)

1. A driving scenario optimization method, comprising:
receiving driving requirements sent by each driving terminal;
determining each driving scheme corresponding to each driving terminal according to each driving requirement and a depth analysis method;
judging whether the information data of each driving scheme meets preset evaluation conditions or not;
if the information data of at least one driving scheme does not meet the evaluation conditions, respectively carrying out optimization analysis on the driving schemes which do not meet the evaluation conditions in an iterative mode according to a depth analysis method until each optimized driving scheme meets the evaluation conditions;
the information data of the driving scheme at least includes: the ratio of the load balance degree of the traffic hub, the energy consumption of the driving terminal and the total mileage is calculated;
each of the driving scenarios is stored in the form of a two-dimensional vector as:
Figure FDA0003058961040000011
wherein k is the current iteration number, (i, j, t) is a three-dimensional coordinate, and i belongs to [1, m ]]，j∈[1,n]，t∈[1,q]The driving scheme corresponds to the driving demand based on the three-dimensional coordinates;
Figure FDA0003058961040000012
and
Figure FDA0003058961040000013
respectively the load balance degree and the driving of the traffic hub with three-dimensional coordinates (i, j, t) in the kth iteration processThe ratio of the terminal energy consumption to the total mileage;
the step of respectively carrying out optimization analysis on the driving schemes which do not meet the evaluation conditions according to a depth analysis method comprises the following steps of:
optimizing the current driving scheme according to a preset unsupervised learning method according to the current driving scheme and the historical driving scheme of the driving terminal to obtain a plurality of optimized driving schemes of the driving terminal;
selecting an optimized driving scheme meeting preset optimization conditions from the optimized driving schemes;
the step of optimizing the current driving scheme according to a preset unsupervised learning method according to the current driving scheme and the historical driving scheme of the driving terminal to obtain a plurality of optimized driving schemes of the driving terminal comprises the following steps:
calculating the (k + 1) th depth unsupervised learning enhancement factor according to the formula (1)
Figure FDA0003058961040000014
Figure FDA0003058961040000021
Calculating the optimized driving scheme M with the corresponding three-dimensional coordinate (i, j, t) obtained by the (k + 1) th iteration according to the formula (2)ijt k+1：
Figure FDA0003058961040000022
Wherein, delta and mu are adjustable parameters,
Figure FDA0003058961040000023
load balance degree of the traffic hub in the current k-th iteration process, CmaxGFor the historical maximum traffic hub load balancing,
Figure FDA0003058961040000024
is the ratio of the energy consumption of the driving terminal to the total mileage in the current k iteration process, WminGIs the ratio of the historical minimum driving terminal energy consumption to the total mileage,
Figure FDA0003058961040000025
and (5) obtaining the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) for the kth iteration.
2. The travel plan optimization method according to claim 1, wherein the travel demand includes: a source location, a pathway location, and a target location; selecting an optimized driving scheme meeting preset optimization conditions according to the following formula:
Figure FDA0003058961040000026
Figure FDA0003058961040000027
Figure FDA0003058961040000028
wherein the minimum ZkThe corresponding scheme is the optimized driving scheme,
Figure FDA0003058961040000029
is a longitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration,
Figure FDA00030589610400000210
For the longitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration,
Figure FDA00030589610400000211
Is the latitude value of the target position in the driving demand corresponding to the three-dimensional coordinates (i, j, t) in the k-th iteration process,
Figure FDA00030589610400000212
Is the latitude value of the path position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k-th iteration process,
Figure FDA00030589610400000213
for the longitude value of the source location in the driving demand corresponding to the three-dimensional coordinate (i, j, t) during the kth iteration,
Figure FDA00030589610400000214
is the latitude value of the source position in the driving demand corresponding to the three-dimensional coordinate (i, j, t) in the k iteration process.
3. The running scenario optimization method according to any one of claims 1-2, wherein the evaluation condition includes:
Figure FDA0003058961040000031
wherein i belongs to [1, m ], j belongs to [1, n ], t belongs to [1, q ], and k is the current iteration number.
4. A load balancing analyzer, comprising:
the receiving module is used for receiving the driving requirements sent by each driving terminal;
the determining module is used for determining each driving scheme corresponding to each driving terminal according to each driving requirement and a depth analysis method;
the judging module is used for judging whether the information data of each driving scheme meets preset evaluation conditions or not;
the optimization module is used for respectively carrying out optimization analysis on the running schemes which do not meet the evaluation conditions according to a depth analysis method in an iterative mode until each optimized running scheme meets the evaluation conditions if the information data of at least one running scheme does not meet the evaluation conditions;
the information data of the driving scheme at least includes: the ratio of the load balance degree of the traffic hub, the energy consumption of the driving terminal and the total mileage is calculated;
each of the driving scenarios is stored in the form of a two-dimensional vector as:
Figure FDA0003058961040000032
wherein k is the current iteration number, (i, j, t) is a three-dimensional coordinate, and i belongs to [1, m ]]，j∈[1,n]，t∈[1,q]The driving scheme corresponds to the driving demand based on the three-dimensional coordinates;
Figure FDA0003058961040000033
and
Figure FDA0003058961040000034
in the kth iteration process, the three-dimensional coordinates are the load balance degree of the traffic hub and the ratio of the energy consumption of the driving terminal to the total mileage of (i, j, t);
the optimization module comprises:
the first optimization submodule is used for optimizing the current driving scheme according to the current driving scheme and the historical driving scheme of the driving terminal and a preset unsupervised learning method to obtain a plurality of optimized driving schemes of the driving terminal;
the second optimization submodule is used for selecting an optimized driving scheme meeting preset optimization conditions from the optimized driving schemes;
the first optimization submodule is specifically configured to:
calculating the (k + 1) th depth unsupervised learning enhancement factor according to the formula (1)
Figure FDA0003058961040000035
Figure FDA0003058961040000041
Calculating the optimized driving scheme M with the corresponding three-dimensional coordinate (i, j, t) obtained by the (k + 1) th iteration according to the formula (2)ijt k+1：
Figure FDA0003058961040000042
Wherein, delta and mu are adjustable parameters,
Figure FDA0003058961040000043
load balance degree of the traffic hub in the current k-th iteration process, CmaxGFor the historical maximum traffic hub load balancing,
Figure FDA0003058961040000044
is the ratio of the energy consumption of the driving terminal to the total mileage in the current k iteration process, WminGIs the ratio of the historical minimum driving terminal energy consumption to the total mileage,
Figure FDA0003058961040000045
and (5) obtaining the optimized driving scheme with the corresponding three-dimensional coordinate (i, j, t) for the kth iteration.
5. An intelligent driving optimization system, comprising:
the load balancing analyzer of claim 4;
the driving terminals are used for sending driving demands to the load balance analyzer; and also for receiving a driving scheme.
6. The intelligent driving optimization system of claim 5, further comprising:
and the edge server is used for controlling the driving terminal to drive according to the driving scheme.
7. The intelligent driving optimization system of claim 5, further comprising: a network transmission unit for transmitting the driving demand; and/or for transmitting the driving profile.
CN201910589318.0A 2019-07-02 2019-07-02 Driving scheme optimization method, load balance analyzer and intelligent driving optimization system Active CN110298512B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910589318.0A CN110298512B (en) 2019-07-02 2019-07-02 Driving scheme optimization method, load balance analyzer and intelligent driving optimization system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910589318.0A CN110298512B (en) 2019-07-02 2019-07-02 Driving scheme optimization method, load balance analyzer and intelligent driving optimization system

Publications (2)

Publication Number Publication Date
CN110298512A CN110298512A (en) 2019-10-01
CN110298512B true CN110298512B (en) 2021-06-29

Family

ID=68029867

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910589318.0A Active CN110298512B (en) 2019-07-02 2019-07-02 Driving scheme optimization method, load balance analyzer and intelligent driving optimization system

Country Status (1)

Country Link
CN (1) CN110298512B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112363840B (en) * 2020-11-23 2023-01-17 中国联合网络通信集团有限公司 Resource load balancing scheme optimization method and device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108833550A (en) * 2018-06-21 2018-11-16 中国联合网络通信集团有限公司 A kind of method and device perceiving agriculture Optimized Operation
CN108830422A (en) * 2018-06-21 2018-11-16 中国联合网络通信集团有限公司 Optimization method, the apparatus and system of intelligent driving

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9875445B2 (en) * 2014-02-25 2018-01-23 Sri International Dynamic hybrid models for multimodal analysis
US10943578B2 (en) * 2016-12-13 2021-03-09 Ultrahaptics Ip Ltd Driving techniques for phased-array systems
CN107995126B (en) * 2017-12-12 2021-08-03 中国联合网络通信集团有限公司 Load balancing route analysis method and load balancing route analyzer

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108833550A (en) * 2018-06-21 2018-11-16 中国联合网络通信集团有限公司 A kind of method and device perceiving agriculture Optimized Operation
CN108830422A (en) * 2018-06-21 2018-11-16 中国联合网络通信集团有限公司 Optimization method, the apparatus and system of intelligent driving

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Joint beamforming and antenna subarray formation for secrecy capacity maximization in MIMO wiretap channels;Jianli Huang;《2016 IEEE Conference on Communications and Network Security (CNS)》;20170223;第1-2页 *
无穷维向量最优化问题的最优性条件;谢小风;《经济数学》;20061231;第23卷(第4期);第426-431页 *

Also Published As

Publication number Publication date
CN110298512A (en) 2019-10-01

Similar Documents

Publication Publication Date Title
EP1402496B1 (en) Improved method and system for mapping traffic predictions with respect to telematics and route guidance applications
US10620635B2 (en) Moving object movement system and movement path selection method
CN106042963B (en) The cooperative optimization method and system of electrified transportation network and electric system
US20030033582A1 (en) Representations for estimating distance
CN109343946B (en) Software-defined Internet of vehicles computing task migration and scheduling method
CN110210806B (en) Cloud-based unmanned vehicle frame structure with 5G edge calculation function and control evaluation method thereof
CN109612488B (en) Big data micro-service-based mixed travel mode path planning system and method
CN107704980B (en) Multi-agent autonomous decision-making method for processing requirement of newly increased express delivery
CN109709985B (en) Unmanned aerial vehicle task optimization method, device and system
CN109788515B (en) Data collaboration method, MEC (media independent center) equipment and server
CN110298512B (en) Driving scheme optimization method, load balance analyzer and intelligent driving optimization system
CN110135641B (en) Driving scheme optimization method and device and intelligent driving optimization system
CN111367275A (en) Intelligent driving control method, device and system and storage medium
Wang et al. Radio resource allocation for bidirectional offloading in space-air-ground integrated vehicular network
Raiyn Road traffic congestion management based on a search-allocation approach
CN113382383A (en) Method for unloading calculation tasks of public transport vehicle based on strategy gradient
De Nunzio et al. Bi-objective eco-routing in large urban road networks
CN102128630B (en) Navigation method and navigator
Roy et al. DENSE: Dynamic edge node selection for safety-as-a-service
Tao et al. A traffic scheduling scheme for load balancing in SDN-based space-air-ground integrated networks
Zhao et al. Research on logistics distribution route based on multi-objective sorting genetic algorithm
CN111445715B (en) Intelligent city traffic scheduling method and scheduling equipment based on Internet of things communication
CN115511226A (en) Vehicle path optimization method based on improved differential evolution algorithm
CN115375030A (en) Vehicle route planning method, device, equipment and computer readable storage medium
KR20230082753A (en) Server, method and computer program for providing route information for logistics

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant