CN111461403A - Vehicle path planning method and device, computer readable storage medium and terminal - Google Patents

Vehicle path planning method and device, computer readable storage medium and terminal Download PDF

Info

Publication number
CN111461403A
CN111461403A CN202010151343.3A CN202010151343A CN111461403A CN 111461403 A CN111461403 A CN 111461403A CN 202010151343 A CN202010151343 A CN 202010151343A CN 111461403 A CN111461403 A CN 111461403A
Authority
CN
China
Prior art keywords
training
distance
name sequence
path
sequence
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.)
Granted
Application number
CN202010151343.3A
Other languages
Chinese (zh)
Other versions
CN111461403B (en
Inventor
郑仁
项党
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.)
SAIC Motor Corp Ltd
Original Assignee
SAIC Motor Corp 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 SAIC Motor Corp Ltd filed Critical SAIC Motor Corp Ltd
Priority to CN202010151343.3A priority Critical patent/CN111461403B/en
Publication of CN111461403A publication Critical patent/CN111461403A/en
Application granted granted Critical
Publication of CN111461403B publication Critical patent/CN111461403B/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"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3407—Route searching; Route guidance specially adapted for specific applications
    • G01C21/343—Calculating itineraries, i.e. routes leading from a starting point to a series of categorical destinations using a global route restraint, round trips, touristic trips
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Automation & Control Theory (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides a vehicle path planning method and device, a computer readable storage medium and a terminal, wherein a path planning model is obtained by training a pointer network type deep network, a name sequence and a distance sequence of an access station to be planned, which are obtained by coding, are input into the path planning model, the name sequence is reordered by the path planning model based on the distance sequence, and a new name sequence which can enable the distance of the planned path of the access station to be the shortest is output, so that the planned path corresponding to the new name sequence is obtained. Based on the invention, a path planning strategy can be modeled by a deep network, so that the optimization result of the vehicle path planning problem can be directly calculated without searching.

Description

Vehicle path planning method and device, computer readable storage medium and terminal
Technical Field
The invention relates to the technical field of logistics, in particular to a vehicle path planning method and device, a computer readable storage medium and a terminal.
Background
The Vehicle Routing Problem (VRP) is a combinatorial optimization problem that has proven to be an NP-hard problem, the optimal solution of which cannot be determined unless all combinations are exhaustively possible. In common scenes such as actual logistics transportation, vehicle scheduling, route planning and the like, the solution requirement of the vehicle path planning problem is often met.
At present, the common optimization method for solving the problems is a heuristic algorithm, such as a genetic algorithm, an ant colony algorithm, a particle swarm algorithm, simulated annealing, tabu search and the like. When the calculation scale of the problem is not large, the optimal path can be solved by using a dynamic programming algorithm, the time complexity and the space complexity required for solving the problem by dynamic programming are high, generally more than 20 sites are visited, and the calculation speed is very slow. However, the problem usually has no specific optimization rule, and the heuristic algorithm gradually finds a solution with better performance in a random search manner, so that time is undoubtedly consumed to wait for the continuous random search. That is, the more likely the search is, the closer to the optimal solution is likely.
However, considering that the solution time of the problem in the actual scenario is limited, how to balance the solution performance and the computation loss time becomes an urgent problem to be solved.
Disclosure of Invention
In view of the above, the present invention provides a vehicle path planning method and apparatus, a computer-readable storage medium, and a terminal. The technical scheme is as follows:
a vehicle path planning method, the method comprising:
coding a name sequence and a distance sequence of an access station to be planned, wherein the name sequence comprises a name coding value of each station in the access station, and the distance sequence comprises a distance value between every two stations in the access station;
inputting the name sequence and the distance sequence into a pre-trained path planning model so that the path planning model reorders the name sequence based on the distance sequence and outputs a new name sequence, wherein the new name sequence can enable the planned path distance for accessing the access site to be shortest, and the path planning model is obtained by pre-training a deep network of a pointer network type;
and acquiring a planning path corresponding to the new name sequence.
Preferably, the process of training the depth network of the pointer network type in advance to obtain the path planning model includes:
acquiring training samples, wherein the training samples comprise name sequence samples and distance sequence samples of a plurality of groups of access sites for training, and one group of name sequence samples corresponds to one planning path for training and one distance sequence sample;
extracting a group of name sequence samples for the training from the training samples, inputting the group of name sequence samples and corresponding distance sequence samples into the deep network of the pointer network type, so that the deep network of the pointer network type outputs a target name sequence sample which can enable the corresponding planned path distance for the training to be converged in the group of name sequence samples based on the input distance sequence samples, wherein the initial value of the network parameter of the deep network of the pointer network type is preset;
adjusting the network parameters by calculating the distance of the planned path for training corresponding to the target name sequence sample, so that the deep network of the pointer network type re-executes the training when outputting one target name sequence sample capable of converging the distance of the planned path for training in the set of name sequence samples based on the input distance sequence sample to meet the end condition of the training, and sets a pointer for the final target name sequence sample output by the deep network of the pointer network type when the training is ended, wherein the pointer is the basis or foundation for re-training the deep network of the pointer network type;
and taking the depth network of the pointer network type after multiple times of training as the path planning model.
Preferably, the adjusting the network parameter by calculating the distance of the planned path for training corresponding to the target name sequence sample includes:
calculating the distance of a planning path for training corresponding to the target name sequence sample;
calculating the reference distance of the group of name sequence samples for the training;
calculating the gradient of the network parameter according to the distance of the planning path for training and the reference distance;
updating the network parameter based on the gradient of the network parameter.
Preferably, the calculating the reference distance of the group of name sequence samples in the training includes:
under the condition that the training is not the first training, acquiring the historical reference distance of the last training;
and calculating the reference distance of the group of name sequence samples of the current training by using the distance of the planning path for training and the historical reference distance.
Preferably, the calculating a reference distance of the group of name sequence samples in the current training further includes:
and under the condition that the training is the first training, determining an optimized path of the group of name sequence samples by using a heuristic algorithm, and taking the distance of the optimized path as the reference distance of the group of name sequence samples in the training.
A vehicle path planning apparatus, the apparatus comprising: the system comprises a coding module, a path planning module and a path acquisition module, wherein the path planning module comprises a model training unit;
the model training unit is used for training the depth network of the pointer network type in advance to obtain a path planning model;
the coding module is used for coding a name sequence and a distance sequence of an access station to be planned, wherein the name sequence comprises a name coding value of each station in the access station, and the distance sequence comprises a distance value between every two stations in the access station;
the path planning module is used for inputting the name sequence and the distance sequence into the path planning model, so that the path planning model reorders the name sequence based on the distance sequence and outputs a new name sequence, and the new name sequence can enable the planned path distance for accessing the access station to be shortest;
and the path acquisition module is used for acquiring the planning path corresponding to the new name sequence.
Preferably, the model training unit, configured to train a depth network of a pointer network type in advance to obtain a path planning model, is specifically configured to:
acquiring training samples, wherein the training samples comprise name sequence samples and distance sequence samples of a plurality of groups of access sites for training, and one group of name sequence samples corresponds to one planning path for training and one distance sequence sample; extracting a group of name sequence samples for the training from the training samples, inputting the group of name sequence samples and corresponding distance sequence samples into the deep network of the pointer network type, so that the deep network of the pointer network type outputs a target name sequence sample which can enable the corresponding planned path distance for the training to be converged in the group of name sequence samples based on the input distance sequence samples, wherein the initial value of the network parameter of the deep network of the pointer network type is preset; adjusting the network parameters by calculating the distance of the planned path for training corresponding to the target name sequence sample, so that the deep network of the pointer network type re-executes the training when outputting one target name sequence sample capable of converging the distance of the planned path for training in the set of name sequence samples based on the input distance sequence sample to meet the end condition of the training, and sets a pointer for the final target name sequence sample output by the deep network of the pointer network type when the training is ended, wherein the pointer is the basis or foundation for re-training the deep network of the pointer network type; and taking the depth network of the pointer network type after multiple times of training as the path planning model.
Preferably, the model training unit, configured to adjust the network parameter by calculating a distance of a planned training path corresponding to the target name sequence sample, is specifically configured to:
calculating the distance of a planning path for training corresponding to the target name sequence sample; calculating the reference distance of the group of name sequence samples for the training; calculating the gradient of the network parameter according to the distance of the planning path for training and the reference distance; updating the network parameter based on the gradient of the network parameter.
A computer readable storage medium having computer instructions stored thereon, wherein the computer instructions when executed perform the steps of any of the vehicle path planning methods.
A terminal comprising a memory and a processor, the memory having stored thereon computer instructions executable on the processor, wherein the processor, when executing the computer instructions, performs any of the steps of the vehicle path planning method.
The vehicle path planning method and device, the computer-readable storage medium and the terminal provided by the invention obtain the path planning model by training the pointer network type deep network, input the coded name sequence and distance sequence of the access station to be planned into the path planning model, reorder the name sequence by the path planning model based on the distance sequence and output a new name sequence which can enable the distance of the planned path of the access station to be shortest, thereby obtaining the planned path corresponding to the new name sequence. Based on the invention, a path planning strategy can be modeled by a deep network, so that the optimization result of the vehicle path planning problem can be directly calculated without searching.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of a method for vehicle path planning according to an embodiment of the present invention;
FIG. 2 is a network architecture diagram of a pointer network type deep network provided by an embodiment of the present invention;
FIG. 3 is a partial method flow diagram of a vehicle path planning method provided by an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a vehicle path planning device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiment of the invention provides a vehicle path planning method, the flow chart of which is shown in figure 1, and the method comprises the following steps:
and S10, coding a name sequence and a distance sequence of the visited sites to be planned, wherein the name sequence comprises a name coding value of each site in the visited sites, and the distance sequence comprises a distance value between every two sites in the visited sites.
In the embodiment of the invention, the name of each site in the sites to be visited on the path to be planned can be subjected to integer coding, a sequence containing the name coding value of each site is marked as a name sequence, and the name of one site corresponds to a unique name coding value; and simultaneously coding the distance between every two stations in the access stations, and recording a sequence containing the distance value between every two stations as a distance sequence.
For convenience of understanding, A, B, C three visiting stations on the path to be planned are taken as an example, where the distance from station a to station B is a, the distance from station a to station C is B, the distance from station B to station C is C, and B is a + C:
a, B, C, carrying out integer coding on the site names of three visiting sites, wherein the coded site A has a name coding value of 1, the coded site B has a name coding value of 2, and the coded site C has a name coding value of 3, and at the moment, a name sequence X is [ 123 ];
a, B, C the distance between two stations of three visiting stations is coded, the distance value from station A to station A is (A, A):0, the distance value from station A to station B is (A, B): a, the distance value from station A to station C is (A, C): B, the distance value from station B to station A is (B, A): a, the distance value from station B to station B is (B, B):0, the distance value from station B to station C is (B, C): C, the distance value from station C to station A is (C, A): B, the distance value from station C to station B is (C, B): C, the distance value from station C to station C is (C, C):0, A, B), the distance sequence Y { (A, A):0(A, B): a, C): B, (B, A): 0(B, C), A) b (C, B) and C (C, C) are 0.
And S20, inputting the name sequence and the distance sequence into a pre-trained path planning model, so that the path planning model reorders the name sequence based on the distance sequence and outputs a new name sequence, wherein the new name sequence can enable the planned path distance of the access station to be shortest, and the path planning model is obtained by pre-training a deep network of a pointer network type.
See fig. 2 for a network architecture diagram of a pointer network type deep network. The pointer network type deep network is an improved version of the attention mechanism network, and the network is provided with two special cyclic neural networks, wherein one cyclic neural network is used as an encoder to encode an input sequence and then convert the input sequence into a vector, and the other cyclic neural network is used as a decoder to re-decode the encoded vector into a sequence and output the sequence. Compared with the attention mechanism network, the output sequence and the input sequence of the network have a one-to-one pointer relationship, so that the length and the content of the input sequence and the output sequence of the network are the same, and only the arrangement sequence is different.
In the embodiment of the invention, the network can be trained in a reinforcement learning mode, and the path planning model obtained after a period of training can automatically approach to a real optimal path planning strategy. The path planning model may encode the input name sequence and convert the encoded name sequence into a vector, and re-decode the encoded vector into a new name sequence based on the input distance sequence and output the new name sequence. The new name sequence and the name sequence have the same length and content, only the arrangement sequence is different, and the distance of the planned path indicated by the new name sequence is shortest, namely an optimized path is planned.
In a specific implementation process, the deep network of the pointer network type can be trained and modified repeatedly in a reinforcement learning mode, so that a path planning strategy can be adjusted and improved according to actual conditions. In step S20, the process of training the depth network of the pointer network type in advance to obtain the path planning model may include the following steps, and a flowchart of the method is shown in fig. 2:
s201, training samples are obtained, wherein the training samples comprise a plurality of groups of name sequence samples and distance sequence samples of the access sites for training, and one group of name sequence samples corresponds to one planning path for training and corresponds to one distance sequence sample.
In the embodiment of the invention, when training samples are accumulated for the training of the pointer network type deep network, it is assumed that K groups of access sites for training in the training samples need to be optimized, and one group of access sites for training corresponds to one planning path for training. Aiming at each group of training access sites in the K groups of training access sites, carrying out integer coding on the name of each site in the group of training access sites, and recording a sequence containing the name coding value of each site as a name sequence sample; and simultaneously coding the distance between every two stations in the group of visiting stations for training, and recording a sequence containing the distance value between every two stations as a distance sequence sample.
S202, a group of name sequence samples for the training is extracted from the training samples, the group of name sequence samples and the corresponding distance sequence samples are input into a pointer network type deep network, so that the pointer network type deep network outputs one target name sequence sample which can enable the corresponding training planning path distance to be converged in the group of name sequence samples based on the input distance sequence samples, and the initial value of the network parameter of the pointer network type deep network is preset.
In the embodiment of the invention, a small batch of samples are randomly extracted from training samples to train the deep network of the pointer network type, and the number of the samples is set as a fixed value B. Using a group of name sequence samples in the randomly extracted samples to carry out the training, wherein the group of name sequence samples is marked as siThe target name sequence sample of the deep network solution output of the pointer network type is recorded as oiWherein i is more than or equal to 1 and less than or equal to B.
Note that, the network parameter θ of the pointer network type deep network is initialized in advance, and this network parameter determines the output characteristic of the network.
S203, adjusting network parameters by calculating the distance of the planned path for training corresponding to the target name sequence sample, so that the deep network of the pointer network type re-executes outputting one target name sequence sample capable of converging the distance of the planned path for training in the group of name sequence samples based on the input distance sequence sample until the end condition of the training is met, and sets a pointer for the final target name sequence sample output by the deep network of the pointer network type when the training is ended, wherein the pointer is the basis or foundation for re-training the deep network of the pointer network type.
In the embodiment of the invention, the input sequence and the output sequence of the deep network are determined according to the type of the representation pointer networkConditional probability p (o) of mathematical relationship between columnsi|si) To calculate the target name sequence samples oiDistance Dis (o) of the planned path for training represented byi|si)。
Further, network parameters are adjusted through a preset parameter adjustment strategy corresponding to the distance of the planned path for training, and a deep network of the pointer network type outputs a target name sequence sample which can enable the distance of the corresponding planned path for training to be converged again until a finishing condition of the training is met, such as the iteration times reach the maximum value, and the training is finished when the distance of the planned path for training meets a performance index.
In some other embodiments, the step S203 of "adjusting the network parameter by calculating the distance of the planned path for training corresponding to the target name sequence sample" may include the following steps:
calculating the distance of a planning path for training corresponding to the target name sequence sample; calculating the reference distance of the group of name sequence samples for the training; calculating the gradient of the network parameters according to the distance of the planning path for training and the reference distance; updating the network parameters based on the gradient of the network parameters.
In the embodiment of the present invention, the conditional probability p (o) can be determined as described abovei|si) To calculate the distance Dis (o) of the planned path for trainingi|si)。
In the process of calculating the reference distance, if the training is the first training, determining the group of name sequence samples s by using a heuristic algorithmiAnd taking the distance of the optimized path as the reference distance b of the training of the group of name sequence samplesi. It should be noted that the heuristic algorithm used may be one of a genetic algorithm, simulated annealing, a particle swarm algorithm, an ant colony algorithm, and a tabu search, but is not limited to the above algorithm.
If the training is not the first training, the reference distance used by the last training, namely the historical reference distance b, can be obtainedi', and calculating the set of name sequence samples of the training according to the following formula (1)Reference distance b ofi：
bi＝bi′+α(Dis(oi|si)-bi′) (1)
Where α is a preset learning rate.
Setting a reference distance b in an embodiment of the present inventioniThe purpose of (a) is to ensure that the final optimization effect of the deep reinforcement learning is at least higher than the benchmark b in the subsequent trainingiGood results are obtained.
Further, the gradient g of the network parameter θ can be calculated according to the following formula (2)θ：
Figure BDA0002402545010000081
And finally, updating the network parameter theta by using a gradient descent method, wherein the updating process can adopt a traditional gradient descent method, and can also adopt an improved gradient updating method such as a Newton method and an ADAM method.
And S204, taking the depth network of the pointer network type after multiple times of training as a path planning model.
In the embodiment of the invention, a pointer network type deep network trained by a small batch of randomly extracted samples from training samples is used as a path planning model. The path planning model is trained to achieve the optimal effect, and only the visiting stations to be planned need to be input into the path planning model in a sequence format, the path planning model outputs a sequence representing the visiting sequence of the actual stations, and the output sequence is the solution of the planned path with the shortest distance.
After the path planning model is trained, an optimized path planning result can be directly calculated according to an access station to be planned, a large number of searching and probing processes are not needed, and the quality of a solution is not lower than that of a heuristic algorithm providing a reference. In addition, the path planning model is trained based on the strength learning mode, parameters can be updated in real time, the planning performance of the model is improved, the model gradually adapts to a real scene along with the increase of time and experience, and the planning result is further improved and can be closer to the theoretical optimal solution. Finally, the path planning model can be suitable for most combination optimization problems, and the algorithm is strong in universality.
And S30, acquiring the planning path corresponding to the new name sequence.
In the embodiment of the invention, one name code value in the name sequence uniquely corresponds to the name of one access station to be planned, and the new name sequence is the same as the name sequence in length and content and only has different arrangement sequence, so that the sequence of accessing the access station to be planned can be obtained based on the sequence of the name code values in the new name sequence, the corresponding planned path is determined, and the distance of the planned path is the shortest.
According to the vehicle path planning method provided by the embodiment of the invention, a path planning model is obtained by training a pointer network type deep network, the coded name sequence and distance sequence of the access station to be planned are input into the path planning model, the name sequence is reordered by the path planning model based on the distance sequence, and a new name sequence which can enable the planned path distance of the access station to be the shortest is output, so that the planned path corresponding to the new name sequence is obtained. Based on the invention, a path planning strategy can be modeled by a deep network, so that the optimization result of the vehicle path planning problem can be directly calculated without searching.
Based on the vehicle path planning method provided by the above embodiment, an embodiment of the present invention provides an apparatus for executing the vehicle path planning method, and a schematic structural diagram of the apparatus is shown in fig. 4, where the apparatus includes: the system comprises a coding module 10, a path planning module 20 and a path obtaining module 30, wherein the path planning module 20 comprises a model training unit 201;
the model training unit 201 is used for training a depth network of a pointer network type in advance to obtain a path planning model;
the system comprises an encoding module 10, a calculation module and a calculation module, wherein the encoding module is used for encoding a name sequence and a distance sequence of an access station to be planned, the name sequence comprises a name encoding value of each station in the access station, and the distance sequence comprises a distance value between every two stations in the access station;
the path planning module 20 is configured to input the name sequence and the distance sequence into the path planning model, so that the path planning model reorders the name sequence based on the distance sequence and outputs a new name sequence, where the new name sequence can minimize a planned path distance for accessing the access station;
and a path obtaining module 30, configured to obtain a planned path corresponding to the new name sequence.
Optionally, the model training unit 201 is configured to train a deep network of a pointer network type in advance to obtain a path planning model, and is specifically configured to:
acquiring training samples, wherein the training samples comprise a plurality of groups of name sequence samples and distance sequence samples of the access sites for training, and one group of name sequence samples corresponds to one planning path for training and corresponds to one distance sequence sample; extracting a group of name sequence samples for the training from the training samples, inputting the group of name sequence samples and corresponding distance sequence samples into a pointer network type deep network, so that the pointer network type deep network outputs a target name sequence sample which can enable the corresponding training planning path distance to be converged in the group of name sequence samples based on the input distance sequence samples, and the initial value of the network parameter of the pointer network type deep network is preset; adjusting network parameters by calculating the distance of a planned path for training corresponding to a target name sequence sample, so that the deep network of the pointer network type re-executes the process of outputting one target name sequence sample capable of converging the distance of the corresponding planned path for training in the group of name sequence samples based on the input distance sequence sample until the finishing condition of the training is met, and setting a pointer for the final target name sequence sample output by the deep network of the pointer network type when the training is finished, wherein the pointer is the basis or foundation for re-training the deep network of the pointer network type; and taking the pointer network type deep network after multiple times of training as a path planning model.
Optionally, the model training unit 201 is configured to adjust the network parameter by calculating a distance of the planned training path corresponding to the target name sequence sample, and specifically configured to:
calculating the distance of a planning path for training corresponding to the target name sequence sample; calculating the reference distance of the group of name sequence samples for the training; calculating the gradient of the network parameters according to the distance of the planning path for training and the reference distance; updating the network parameters based on the gradient of the network parameters.
Optionally, the model training unit 201 configured to calculate a reference distance for training the group of name sequence samples at this time is specifically configured to:
under the condition that the training is not the first training, acquiring the historical reference distance of the last training; and calculating the reference distance of the group of name sequence samples in the training by using the distance of the planning path for training and the historical reference distance.
Optionally, the model training unit 201 configured to calculate a reference distance for training the group of name sequence samples at this time is further configured to:
and under the condition that the training is the first training, determining an optimized path of the group of name sequence samples by using a heuristic algorithm, and taking the distance of the optimized path as the reference distance of the group of name sequence samples in the training.
The vehicle path planning device provided by the embodiment of the invention obtains a path planning model by training a pointer network type deep network, inputs a name sequence and a distance sequence of an access station to be planned, which are obtained by coding, into the path planning model, reorders the name sequence based on the distance sequence by the path planning model, and outputs a new name sequence which can enable the distance of the planned path of the access station to be shortest, thereby obtaining the planned path corresponding to the new name sequence. Based on the invention, a path planning strategy can be modeled by a deep network, so that the optimization result of the vehicle path planning problem can be directly calculated without searching.
The embodiment of the invention also provides a computer-readable storage medium, on which computer instructions are stored, and when the computer instructions are executed, the steps of the vehicle path planning method shown in the embodiment of the invention are executed. The computer-readable storage medium may include, for example, a non-volatile (non-volatile) or non-transitory (non-transitory) memory, and may also include an optical disc, a mechanical hard disk, a solid state hard disk, and the like.
The embodiment of the invention also provides a terminal, which comprises a memory and a processor, wherein the memory is stored with computer instructions capable of running on the processor, and the processor executes the steps of the vehicle path planning method shown in the embodiment of the invention when running the computer instructions. The terminal includes, but is not limited to, a mobile phone, a computer, a tablet computer and other terminal devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the system or system embodiments are substantially similar to the method embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described system and system embodiments are only illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing is only a preferred embodiment of the present application and it should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims (10)

1. A vehicle path planning method, characterized in that the method comprises:
coding a name sequence and a distance sequence of an access station to be planned, wherein the name sequence comprises a name coding value of each station in the access station, and the distance sequence comprises a distance value between every two stations in the access station;
inputting the name sequence and the distance sequence into a pre-trained path planning model so that the path planning model reorders the name sequence based on the distance sequence and outputs a new name sequence, wherein the new name sequence can enable the planned path distance for accessing the access site to be shortest, and the path planning model is obtained by pre-training a deep network of a pointer network type;
and acquiring a planning path corresponding to the new name sequence.
2. The method according to claim 1, wherein the process of training a deep network of pointer network type in advance to obtain the path planning model comprises:
acquiring training samples, wherein the training samples comprise name sequence samples and distance sequence samples of a plurality of groups of access sites for training, and one group of name sequence samples corresponds to one planning path for training and one distance sequence sample;
extracting a group of name sequence samples for the training from the training samples, inputting the group of name sequence samples and corresponding distance sequence samples into the deep network of the pointer network type, so that the deep network of the pointer network type outputs a target name sequence sample which can enable the corresponding planned path distance for the training to be converged in the group of name sequence samples based on the input distance sequence samples, wherein the initial value of the network parameter of the deep network of the pointer network type is preset;
adjusting the network parameters by calculating the distance of the planned path for training corresponding to the target name sequence sample, so that the deep network of the pointer network type re-executes the training when outputting one target name sequence sample capable of converging the distance of the planned path for training in the set of name sequence samples based on the input distance sequence sample to meet the end condition of the training, and sets a pointer for the final target name sequence sample output by the deep network of the pointer network type when the training is ended, wherein the pointer is the basis or foundation for re-training the deep network of the pointer network type;
and taking the depth network of the pointer network type after multiple times of training as the path planning model.
3. The method of claim 2, wherein the adjusting the network parameters by calculating the distance of the planned path for training corresponding to the target name sequence sample comprises:
calculating the distance of a planning path for training corresponding to the target name sequence sample;
calculating the reference distance of the group of name sequence samples for the training;
calculating the gradient of the network parameter according to the distance of the planning path for training and the reference distance;
updating the network parameter based on the gradient of the network parameter.
4. The method of claim 3, wherein calculating the reference distance for the set of samples of the name sequence of the current training comprises:
under the condition that the training is not the first training, acquiring the historical reference distance of the last training;
and calculating the reference distance of the group of name sequence samples of the current training by using the distance of the planning path for training and the historical reference distance.
5. The method of claim 4, wherein calculating the reference distance for the set of name sequence samples in the current training further comprises:
and under the condition that the training is the first training, determining an optimized path of the group of name sequence samples by using a heuristic algorithm, and taking the distance of the optimized path as the reference distance of the group of name sequence samples in the training.
6. A vehicle path planning apparatus, the apparatus comprising: the system comprises a coding module, a path planning module and a path acquisition module, wherein the path planning module comprises a model training unit;
the model training unit is used for training the depth network of the pointer network type in advance to obtain a path planning model;
the coding module is used for coding a name sequence and a distance sequence of an access station to be planned, wherein the name sequence comprises a name coding value of each station in the access station, and the distance sequence comprises a distance value between every two stations in the access station;
the path planning module is used for inputting the name sequence and the distance sequence into the path planning model, so that the path planning model reorders the name sequence based on the distance sequence and outputs a new name sequence, and the new name sequence can enable the planned path distance for accessing the access station to be shortest;
and the path acquisition module is used for acquiring the planning path corresponding to the new name sequence.
7. The apparatus according to claim 6, wherein the model training unit, configured to train a deep network of a pointer network type in advance to obtain a path planning model, is specifically configured to:
acquiring training samples, wherein the training samples comprise name sequence samples and distance sequence samples of a plurality of groups of access sites for training, and one group of name sequence samples corresponds to one planning path for training and one distance sequence sample; extracting a group of name sequence samples for the training from the training samples, inputting the group of name sequence samples and corresponding distance sequence samples into the deep network of the pointer network type, so that the deep network of the pointer network type outputs a target name sequence sample which can enable the corresponding planned path distance for the training to be converged in the group of name sequence samples based on the input distance sequence samples, wherein the initial value of the network parameter of the deep network of the pointer network type is preset; adjusting the network parameters by calculating the distance of the planned path for training corresponding to the target name sequence sample, so that the deep network of the pointer network type re-executes the training when outputting one target name sequence sample capable of converging the distance of the planned path for training in the set of name sequence samples based on the input distance sequence sample to meet the end condition of the training, and sets a pointer for the final target name sequence sample output by the deep network of the pointer network type when the training is ended, wherein the pointer is the basis or foundation for re-training the deep network of the pointer network type; and taking the depth network of the pointer network type after multiple times of training as the path planning model.
8. The apparatus according to claim 7, wherein the model training unit configured to adjust the network parameter by calculating a distance of a planned training path corresponding to the target name sequence sample is specifically configured to:
calculating the distance of a planning path for training corresponding to the target name sequence sample; calculating the reference distance of the group of name sequence samples for the training; calculating the gradient of the network parameter according to the distance of the planning path for training and the reference distance; updating the network parameter based on the gradient of the network parameter.
9. A computer readable storage medium having computer instructions stored thereon, wherein the computer instructions when executed perform the steps of the vehicle path planning method of any one of claims 1 to 5.
10. A terminal comprising a memory and a processor, the memory having stored thereon computer instructions executable on the processor, wherein the processor, when executing the computer instructions, performs the steps of the vehicle path planning method according to any one of claims 1 to 5.
CN202010151343.3A 2020-03-06 2020-03-06 Vehicle path planning method and device, computer readable storage medium and terminal Active CN111461403B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010151343.3A CN111461403B (en) 2020-03-06 2020-03-06 Vehicle path planning method and device, computer readable storage medium and terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010151343.3A CN111461403B (en) 2020-03-06 2020-03-06 Vehicle path planning method and device, computer readable storage medium and terminal

Publications (2)

Publication Number Publication Date
CN111461403A true CN111461403A (en) 2020-07-28
CN111461403B CN111461403B (en) 2023-09-29

Family

ID=71682699

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010151343.3A Active CN111461403B (en) 2020-03-06 2020-03-06 Vehicle path planning method and device, computer readable storage medium and terminal

Country Status (1)

Country Link
CN (1) CN111461403B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116090688A (en) * 2023-04-10 2023-05-09 中国人民解放军国防科技大学 Moving target traversal access sequence planning method based on improved pointer network

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106225797A (en) * 2016-06-30 2016-12-14 银江股份有限公司 A kind of paths planning method
CN107038496A (en) * 2017-03-29 2017-08-11 中南大学 Paths planning method and system are delivered in a kind of unmanned plane express delivery automatically
CN109489667A (en) * 2018-11-16 2019-03-19 楚天智能机器人（长沙）有限公司 A kind of improvement ant colony paths planning method based on weight matrix
CN109724611A (en) * 2019-01-08 2019-05-07 北京三快在线科技有限公司 Paths planning method, device, electronic equipment and storage medium
CN109764884A (en) * 2019-01-02 2019-05-17 北京科技大学 A kind of school bus paths planning method and device for planning
CN110348636A (en) * 2019-07-12 2019-10-18 南方科技大学 Path planning prediction technique, device, equipment and computer readable storage medium
CN209525760U (en) * 2019-04-11 2019-10-22 上海汽车集团股份有限公司 A kind of vehicle location assistance searching system
WO2019232869A1 (en) * 2018-06-04 2019-12-12 平安科技（深圳）有限公司 Handwriting model training method, text recognition method and apparatus, device, and medium
US20200005448A1 (en) * 2018-06-29 2020-01-02 Photogauge, Inc. SYSTEM AND METHOD FOR USING IMAGES FROM A COMMODITY CAMERA FOR OBJECT SCANNING, REVERSE ENGINEERING, METROLOGY, ASSEMBLY, and ANALYSIS

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106225797A (en) * 2016-06-30 2016-12-14 银江股份有限公司 A kind of paths planning method
CN107038496A (en) * 2017-03-29 2017-08-11 中南大学 Paths planning method and system are delivered in a kind of unmanned plane express delivery automatically
WO2019232869A1 (en) * 2018-06-04 2019-12-12 平安科技（深圳）有限公司 Handwriting model training method, text recognition method and apparatus, device, and medium
US20200005448A1 (en) * 2018-06-29 2020-01-02 Photogauge, Inc. SYSTEM AND METHOD FOR USING IMAGES FROM A COMMODITY CAMERA FOR OBJECT SCANNING, REVERSE ENGINEERING, METROLOGY, ASSEMBLY, and ANALYSIS
CN109489667A (en) * 2018-11-16 2019-03-19 楚天智能机器人（长沙）有限公司 A kind of improvement ant colony paths planning method based on weight matrix
CN109764884A (en) * 2019-01-02 2019-05-17 北京科技大学 A kind of school bus paths planning method and device for planning
CN109724611A (en) * 2019-01-08 2019-05-07 北京三快在线科技有限公司 Paths planning method, device, electronic equipment and storage medium
CN209525760U (en) * 2019-04-11 2019-10-22 上海汽车集团股份有限公司 A kind of vehicle location assistance searching system
CN110348636A (en) * 2019-07-12 2019-10-18 南方科技大学 Path planning prediction technique, device, equipment and computer readable storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘晓;戴敏;郑刚;黄庆军;: "运钞车车辆路径规划策略", 计算机应用 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116090688A (en) * 2023-04-10 2023-05-09 中国人民解放军国防科技大学 Moving target traversal access sequence planning method based on improved pointer network

Also Published As

Publication number Publication date
CN111461403B (en) 2023-09-29

Similar Documents

Publication Publication Date Title
CN109313586B (en) System for iterative training of artificial intelligence using cloud-based metrics
EP3485432B1 (en) Training machine learning models on multiple machine learning tasks
CN110366734B (en) Optimizing neural network architecture
EP3822880A1 (en) Load prediction method and apparatus based on neural network
CN109195135B (en) Base station selection method based on deep reinforcement learning in LTE-V
US20230237375A1 (en) Dynamic placement of computation sub-graphs
US20210014872A1 (en) Method and apparatus for facilitating resource pairing using a deep q-network
EP3218854A1 (en) Generating natural language descriptions of images
CN110728376B (en) Federal learning method and device based on tree topology structure
CN108106624B (en) Multi-user reservation scheduling path planning method and related device
CN109947940B (en) Text classification method, device, terminal and storage medium
CN105407158A (en) Method and device for building model and pushing message
CN115829024B (en) Model training method, device, equipment and storage medium
WO2018172513A1 (en) Training neural networks using posterior sharpening
CN115271130B (en) Dynamic scheduling method and system for maintenance order of ship main power equipment
CN112926789A (en) Satellite cloud picture prediction method, prediction device and readable storage medium
CN111461403B (en) Vehicle path planning method and device, computer readable storage medium and terminal
CN111210017B (en) Method, device, equipment and storage medium for determining layout sequence and data processing
CN114158105B (en) Routing method and device of wireless sensor network applied to comprehensive energy system
CN116756536B (en) Data identification method, model training method, device, equipment and storage medium
CN112104863B (en) Method and related device for training video quality evaluation model and evaluating video quality
CN115630566B (en) Data assimilation method and system based on deep learning and dynamic constraint
CN116017528A (en) Traffic prediction method, traffic prediction device and server
CN115496121A (en) Model training method and device based on federal learning
CN115268259A (en) PID control loop setting method and device

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