CN112581026A - Joint path planning method for logistics robot on alliance chain - Google Patents

Joint path planning method for logistics robot on alliance chain Download PDF

Info

Publication number
CN112581026A
CN112581026A CN202011592153.1A CN202011592153A CN112581026A CN 112581026 A CN112581026 A CN 112581026A CN 202011592153 A CN202011592153 A CN 202011592153A CN 112581026 A CN112581026 A CN 112581026A
Authority
CN
China
Prior art keywords
training
neural network
logistics
logistics robot
action
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
CN202011592153.1A
Other languages
Chinese (zh)
Other versions
CN112581026B (en
Inventor
吴琛
谢杨洁
张帅
胡麦芳
张珂杰
詹士潇
汪小益
黄方蕾
蔡亮
李伟
邱炜伟
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.)
Hangzhou Qulian Technology Co Ltd
Original Assignee
Hangzhou Qulian Technology 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 Hangzhou Qulian Technology Co Ltd filed Critical Hangzhou Qulian Technology Co Ltd
Priority to CN202011592153.1A priority Critical patent/CN112581026B/en
Publication of CN112581026A publication Critical patent/CN112581026A/en
Application granted granted Critical
Publication of CN112581026B publication Critical patent/CN112581026B/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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • 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
    • 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
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • 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
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0835—Relationships between shipper or supplier and carriers
    • G06Q10/08355—Routing methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a method for planning a joint path of logistics robots on a alliance chain, which comprises the following steps: s1, constructing logistics robot nodes participating in training, and forming a alliance chain according to the nodes; s2, constructing a strategy model pi of the logistics robot node participating in training, initializing, and acquiring an action scheme c and information f of the logistics robot node; s3, training a strategy model pi of one of the logistics robot nodes, solving an optimal action scheme c and a neural network parameter theta of the model under the information f, and uploading the optimal action scheme c and the neural network parameter theta to a alliance chain; s4, according to the optimal action scheme c and the neural network parameter theta, carrying out comparison on the neural network parameters theta of other nodesiUpdate and send to all logistics robot sectionsPoint; and analyzing the strategy model pi according to the updated neural network parameters. The invention can carry out the combined strategy training of the logistics robot on the premise of protecting the personal privacy data of each express, and can also ensure the safety in the training process through the block chain technology.

Description

Joint path planning method for logistics robot on alliance chain
Technical Field
The invention relates to the technical field of logistics distribution, in particular to a logistics robot joint path planning method on a union chain.
Background
The rapid development of artificial intelligence technology and robot technology enables more and more intelligent robots to enter our lives, rapid payment and the rise of e-commerce enable our lives and various expressors to be closely connected, and logistics robots enter our sight line under the action of the two trends. At present, the logistics robot is mainly used for centralized dispatching through the center or delivering by a man-made preset route, the logistics robot distribution under a large-scale scene is difficult to achieve reasonable planning and intelligent dispatching, and how to carry out the safety training of the logistics robot on the premise of protecting the privacy of express information becomes the next technical problem which needs to be solved urgently.
Disclosure of Invention
The invention aims to provide a logistics robot joint path planning method on a alliance chain, which can be used for carrying out joint strategy training on a logistics robot on the premise of protecting personal privacy data on express of each party and can also ensure the safety in the training process through a block chain technology.
In order to achieve the purpose, the invention provides the following scheme: the invention provides a method for planning a joint path of logistics robots on a alliance chain, which comprises the following steps:
s1, constructing logistics robot nodes participating in training, and forming a alliance chain according to the nodes;
s2, constructing a strategy model pi of the logistics robot node participating in training, initializing, and acquiring an action scheme c and information f of the logistics robot node;
s3, training a strategy model pi of one of the logistics robot nodes, solving an optimal action scheme c and a neural network parameter theta of the model under the information f, and uploading the optimal action scheme c and the neural network parameter theta to a alliance chain;
s4, according to the optimal action scheme c and the neural network parameter theta, carrying out comparison on the neural network parameter theta of other logistics robot nodesiUpdating and sending the updated information to other logistics robot nodes; and analyzing the strategy model pi according to the updated neural network parameters to complete the path planning of the logistics robot.
Preferably, the specific process of step S3 is: taking the information of the current moment and the training parameters of the neural network as input parameters of the strategy model pi to obtain an optimal action scheme c; and updating the training parameter theta of the neural network to obtain the updated neural network parameter theta.
Preferably, the specific process of step S4 is: using the neural network parameter theta obtained by the robot node in the step S3 as an input parameter of other node models to obtain an action scheme c of other robot nodesi(ii) a According to other node action schemes ciCalculating the time for specifically completing the current logistics transportation task
Figure BDA0002869519300000021
Transporting task time by logistics
Figure BDA0002869519300000022
According to other robot node neural network parameters thetaiCalculated task time
Figure BDA0002869519300000023
Obtaining a score s; judging the score s, and if the mean value of the score s reaches a set threshold value, determining the current parameter theta of the nodeiAnd the update is performed efficiently.
Preferably, the specific process of analyzing the policy model pi according to the updated neural network parameter in step S4 is as follows: according to the updated neural network parameter thetaiJudging whether the model meets the training requirement, if so, ending the training; if not, the updating is continued.
Preferably, the training requirement is obtained by using a patience coefficient, and the specific process is as follows: obtaining training times according to the heart endurance coefficient value, and training the logistics robot according to the training times; if the action time obtained by a certain training is longer than the action time obtained last time, subtracting 1 from the training times, and if the action time obtained by a certain training is shorter than the action time obtained last time, resetting the training times; when the training times is 0, ending the training; and after the training is finished, selecting the parameter of a certain training with the shortest action time as the training requirement of the model.
Preferably, the action plan c comprises a delivery sequence of the packages and a preset action route;
the delivery sequence is represented by a vector; the preset action route is expressed by a coordinate set.
Preferably, the information f comprises the number of packages to be delivered and delivery information of the packages;
the delivery information of the package comprises address coordinates and delivery receiver information.
The invention discloses the following technical effects:
the invention can effectively carry out the combined training of the logistics robot on the premise of not directly exchanging express information, the process can effectively protect personal information of a user from being leaked to the maximum extent, the combined training of multiple robots can also obviously improve the training efficiency of a logistics robot model, and the problem that the model training of individual logistics robots is poor due to special or lack of data can be effectively avoided. Meanwhile, the safety in the training process can be well guaranteed by utilizing the block chain technology, and the training process is prevented from being maliciously attacked or the data in the training process is prevented from being stolen.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a method for planning a joint path of a logistics robot in a alliance chain according to the present invention;
FIG. 2 is a flow chart of the neural network parameter θ updating process according to 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 order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the present invention provides a method for planning a joint path of a logistics robot in a alliance chain, which includes the following steps:
and S1, constructing logistics robot nodes participating in training, and forming a alliance chain according to the nodes.
The federation chain is a special mechanism in the block chain, and is different from the traditional block chain in that the traditional block chain can access data related to the block chain by a private key, other people can access the data uploaded by the block chain by sharing a public key with the other people, and participants of the federation chain can directly access any information uploaded in the federation chain by other participants, but non-participants can only see the storage block with the chain, and cannot obtain the specific content in the storage block.
Since the embodiment involves mutual evaluation of multiple robots, parameters and scores of evaluation need to be exchanged in the evaluation process. By performing the two steps on the alliance chain, a non-alliance chain participant can only see that a certain record exists in the alliance chain, but cannot see the specific content of the record, and can directly acquire uploaded or shared data (parameters and scores) of other participants without secondary identity authentication for the participants of the alliance chain; secondly, the internal protection is also safe in a alliance chain (which can be said to be beneficial on a block chain), which can prevent the occurrence of the phenomenon that a certain internal participating logistics robot is tampered with the existing data or adds the data of malicious attack due to failure or malicious attack, because the data on the alliance chain (block chain) is backed up by all nodes, the data on 51% of devices needs to be modified to modify one data, and because the attack of the relation between the distribution of the storage nodes and the number of the devices needs to spend great computing power, the attack is often unrendered, so that the safety of the data on the block chain can be greatly improved.
S2, constructing a strategy model pi of the logistics robot participating in training, initializing, and acquiring an action scheme c and information f of the logistics robot.
The action scheme c records the delivery sequence of the packages, the preset action route and the like, a specific existence form provides a possibility, the delivery sequence can be a vector, each number in the vector is the code of the package, the sequence of elements in the vector is the delivery sequence, the action route can be a set of a series of coordinates, and the robot can move along the preset track according to the sequence of the coordinates.
The information f records the number of packages to be delivered, delivery information (address coordinates, delivery receiver information, etc.) of the packages, etc.
The policy model of this embodiment is established by using a convolutional neural network, and the policy model pi represents the probability of making the action scheme c under the information f in the step t, specifically:
π(c|f；θ)＝P(Ct＝c|Ft＝f；θt) (1)
wherein: theta is a training parameter of the neural network; ct、Ft、θtAnd (4) an action scheme, information and neural network training parameters adopted in the step t.
And S3, training the strategy model pi of one node to obtain the optimal action scheme c of the model under the information f and the neural network parameter theta, and uploading the optimal action scheme c and the neural network parameter theta to a alliance chain.
Inputting the information f of the current moment and the training parameter theta of the neural network into the formula (1) to obtain an action scheme c, and updating the training parameter theta of the neural network according to the formula (2), specifically:
Figure BDA0002869519300000061
wherein: alpha is the learning rate; n is the number of logistics robots passing evaluation; thetatParameters of the neural network at time t (current time); thetat+1Parameters of the neural network at time t +1 (next time); dθ,tThe derivative of the parameter theta of the neural network at the time t (the current time) is calculated as follows:
Figure BDA0002869519300000062
Qi(f,C)＝E[Ut|Ft＝f,Ct＝c] (4)
wherein: e2]To find the desired function; u shapetExciting a historical step, specifically:
Ut＝Rt+γRt+1+γ2Rt+2+...+γxRt+x (5)
wherein: u shapetIs the importance function of step t; gamma is a coefficient between 0 and 1; x is the total round number; rtThe reward value of the step t is specifically as follows:
Figure BDA0002869519300000071
wherein: t istThe time required for delivering the logistics robot according to the t-time scheme; t ist-1The time required for the logistics robot to deliver is determined according to the t-1 time scheme;
Figure BDA0002869519300000072
the time excitation factor is generally 0-0.1.
S4, according to the optimal action scheme c and the neural network parameter theta, carrying out comparison on the neural network parameters theta of other nodesiAnd updating and sending the updated parameters to all the logistics robot nodes.
In this embodiment, the score s of the self model scheme is calculated by using the neural network parameter θ through other nodes, which specifically includes:
obtaining the action scheme c by using the neural network parameter theta obtained by the node in the step S3 as an input parameter of other node modelsiThen by action scheme ciCalculating the time for specifically completing the current logistics transportation task
Figure BDA0002869519300000073
Time to carry logistics
Figure BDA0002869519300000074
And node optimum parameter thetaiCalculating to obtain task time
Figure BDA0002869519300000075
And comparing to obtain a score s, which is specifically as follows:
Figure BDA0002869519300000076
wherein: and sigma is a score adjusting parameter, and the value of sigma is generally 1-10.
If the average value of the scores reaches a set threshold value, the current parameter theta of the node isiIf the score average value does not reach the set threshold value, the parameter θ is not updated, the threshold value set in the present embodiment is related to the σ parameter during calculation, and when σ is 10, the threshold value is generally set to 7-9, which can be actually defined according to the request of the logistics robot for the distribution scheme.
After all the logistics robot nodes are updated, the updated parameters are sent to all the logistics robot nodes, and the parameter updating process of the embodiment is shown in fig. 2.
S5, according to the updated neural network parameter thetaiJudging whether the model meets the path planning requirement, if so, finishing the training; if not, the process returns to step S4.
In the embodiment, a patience coefficient is used as a path planning requirement of a model, a training coefficient is set to be 5, the training frequency is reduced by 1 if the action time obtained by a certain training is longer than the action time obtained last time, and the training frequency is reset if the action time obtained by a certain training is shorter than the action time obtained last time; when the training times is 0, ending the training; and after the training is finished, selecting the parameter of a certain training with the shortest action time as the training requirement of the model.
The above-described embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements of the technical solutions of the present invention can be made by those skilled in the art without departing from the spirit of the present invention, and the technical solutions of the present invention are within the scope of the present invention defined by the claims.

Claims (7)

1. A method for planning a joint path of a logistics robot on a alliance chain is characterized by comprising the following steps:
s1, constructing logistics robot nodes participating in training, and forming a alliance chain according to the nodes;
s2, constructing a strategy model pi of the logistics robot node participating in training, initializing, and acquiring an action scheme c and information f of the logistics robot node;
s3, training a strategy model pi of one of the logistics robot nodes, solving an optimal action scheme c and a neural network parameter theta of the model under the information f, and uploading the optimal action scheme c and the neural network parameter theta to a alliance chain;
s4, according to the optimal action scheme c and the neural network parameter theta, carrying out comparison on the neural network parameter theta of other logistics robot nodesiUpdating and sending the updated information to other logistics robot nodes; and analyzing the strategy model pi according to the updated neural network parameters to complete the path planning of the logistics robot.
2. The method for planning the joint path of the logistics robots in the alliance chain according to claim 1, wherein the specific process of the step S3 is as follows: taking the information of the current moment and the training parameters of the neural network as input parameters of the strategy model pi to obtain an optimal action scheme c; and updating the training parameter theta of the neural network to obtain the updated neural network parameter theta.
3. The method for planning the joint path of the logistics robots in the alliance chain according to claim 1, wherein the specific process of the step S4 is as follows: using the neural network parameter theta obtained by the robot node in the step S3 as an input parameter of other node models to obtain an action scheme c of other robot nodesi(ii) a According to other node action schemes ciCalculating the time for specifically completing the current logistics transportation task
Figure FDA0002869519290000011
Transporting task time by logistics
Figure FDA0002869519290000012
Participating in neural network according to other robot nodesNumber thetaiCalculated task time
Figure FDA0002869519290000021
Obtaining a score s; judging the score s, and if the mean value of the score s reaches a set threshold value, determining the current parameter theta of the nodeiAnd the update is performed efficiently.
4. The method for planning the joint path of the logistics robots in the alliance chain according to claim 1, wherein the specific process of analyzing the policy model pi according to the updated neural network parameters in the step S4 is as follows: according to the updated neural network parameter thetaiJudging whether the model meets the training requirement, if so, ending the training; if not, the updating is continued.
5. The jointed path planning method for logistics robots in alliances-chains according to claim 4, wherein the training requirement is obtained by using a tolerance coefficient, and the specific process is as follows: obtaining training times according to the heart endurance coefficient value, and training the logistics robot according to the training times; if the action time obtained by a certain training is longer than the action time obtained last time, subtracting 1 from the training times, and if the action time obtained by a certain training is shorter than the action time obtained last time, resetting the training times; when the training times is 0, ending the training; and after the training is finished, selecting the parameter of a certain training with the shortest action time as the training requirement of the model.
6. The integrated federation chain logistics robot path planning method of claim 1, wherein the action plan c comprises a delivery sequence of packages and a preset action route;
the delivery sequence is represented by a vector; the preset action route is expressed by a coordinate set.
7. The integrated federation chain logistics robot path planning method of claim 1, wherein the information f includes the number of parcels to be delivered, delivery information of the parcels;
the delivery information of the package comprises address coordinates and delivery receiver information.
CN202011592153.1A 2020-12-29 2020-12-29 Joint path planning method for logistics robot on alliance chain Active CN112581026B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011592153.1A CN112581026B (en) 2020-12-29 2020-12-29 Joint path planning method for logistics robot on alliance chain

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011592153.1A CN112581026B (en) 2020-12-29 2020-12-29 Joint path planning method for logistics robot on alliance chain

Publications (2)

Publication Number Publication Date
CN112581026A true CN112581026A (en) 2021-03-30
CN112581026B CN112581026B (en) 2022-08-12

Family

ID=75144077

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011592153.1A Active CN112581026B (en) 2020-12-29 2020-12-29 Joint path planning method for logistics robot on alliance chain

Country Status (1)

Country Link
CN (1) CN112581026B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113537625A (en) * 2021-07-30 2021-10-22 重庆移通学院 Data sharing method considering energy consumption efficiency in power internet of things based on block chain

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107272705A (en) * 2017-07-31 2017-10-20 中南大学 A kind of multiple neural network controlling planning method of robot path under intelligent environment
CN108803313A (en) * 2018-06-08 2018-11-13 哈尔滨工程大学 A kind of paths planning method based on ocean current prediction model
CN109711557A (en) * 2018-12-28 2019-05-03 西安航空学院 A kind of wheelpath prediction technique, system, computer equipment and storage medium
US20190377362A1 (en) * 2019-07-16 2019-12-12 Lg Electronics Inc. Artificial intelligence device installed in vehicle and method therefor
CN111061277A (en) * 2019-12-31 2020-04-24 歌尔股份有限公司 Unmanned vehicle global path planning method and device
CN111158401A (en) * 2020-01-20 2020-05-15 北京理工大学 Distributed unmanned aerial vehicle path planning system and method for encouraging space-time data exploration
CN111415048A (en) * 2020-04-10 2020-07-14 大连海事大学 Vehicle path planning method based on reinforcement learning
CN111462489A (en) * 2020-04-01 2020-07-28 腾讯云计算（北京）有限责任公司 Traffic congestion area prediction method and device
CN111860828A (en) * 2020-06-15 2020-10-30 北京仿真中心 Neural network training method, storage medium and equipment
CN112101939A (en) * 2020-09-14 2020-12-18 邢文超 Node management method and system based on block chain

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107272705A (en) * 2017-07-31 2017-10-20 中南大学 A kind of multiple neural network controlling planning method of robot path under intelligent environment
CN108803313A (en) * 2018-06-08 2018-11-13 哈尔滨工程大学 A kind of paths planning method based on ocean current prediction model
CN109711557A (en) * 2018-12-28 2019-05-03 西安航空学院 A kind of wheelpath prediction technique, system, computer equipment and storage medium
US20190377362A1 (en) * 2019-07-16 2019-12-12 Lg Electronics Inc. Artificial intelligence device installed in vehicle and method therefor
CN111061277A (en) * 2019-12-31 2020-04-24 歌尔股份有限公司 Unmanned vehicle global path planning method and device
CN111158401A (en) * 2020-01-20 2020-05-15 北京理工大学 Distributed unmanned aerial vehicle path planning system and method for encouraging space-time data exploration
CN111462489A (en) * 2020-04-01 2020-07-28 腾讯云计算（北京）有限责任公司 Traffic congestion area prediction method and device
CN111415048A (en) * 2020-04-10 2020-07-14 大连海事大学 Vehicle path planning method based on reinforcement learning
CN111860828A (en) * 2020-06-15 2020-10-30 北京仿真中心 Neural network training method, storage medium and equipment
CN112101939A (en) * 2020-09-14 2020-12-18 邢文超 Node management method and system based on block chain

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113537625A (en) * 2021-07-30 2021-10-22 重庆移通学院 Data sharing method considering energy consumption efficiency in power internet of things based on block chain

Also Published As

Publication number Publication date
CN112581026B (en) 2022-08-12

Similar Documents

Publication Publication Date Title
CN112348204B (en) Safe sharing method for marine Internet of things data under edge computing framework based on federal learning and block chain technology
WO2021114929A1 (en) Blockchain-based model combination training method and device
WO2020029590A1 (en) Sample prediction method and device based on federated training, and storage medium
Huang et al. Consensus mechanism for software-defined blockchain in internet of things
CN109035014B (en) Data transaction system
CN113408743B (en) Method and device for generating federal model, electronic equipment and storage medium
CN112257873A (en) Training method, device, system, equipment and storage medium of machine learning model
CN107196974A (en) A kind of space mass-rent worker's location privacy protection method based on difference privacy
CN112581026B (en) Joint path planning method for logistics robot on alliance chain
CN109741005A (en) A kind of processing method and processing device of recruitment information
Wang et al. Fed-ltd: Towards cross-platform ride hailing via federated learning to dispatch
CN111598417B (en) Natural language processing-based electric power non-emergency repair order distribution method
Wang et al. A trusted consensus scheme for collaborative learning in the edge ai computing domain
Xu et al. Secure and reliable transfer learning framework for 6G-enabled Internet of Vehicles
Zheng et al. Dynamic case retrieval method with subjective preferences and objective information for emergency decision making
CN113902037A (en) Abnormal bank account identification method, system, electronic device and storage medium
CN115396442A (en) Calculation force sharing system and method for urban rail transit
CN110874638B (en) Behavior analysis-oriented meta-knowledge federation method, device, electronic equipment and system
CN114330750B (en) Method for detecting federated learning poisoning attack
CN109919764A (en) Domain name transaction monitoring and managing method and device based on block chain
CN116127519A (en) Dynamic differential privacy federal learning system based on blockchain
Xu et al. A blockchain-based federated learning scheme for data sharing in industrial internet of things
CN112101577A (en) XGboost-based cross-sample federal learning and testing method, system, device and medium
CN114781940B (en) Carbon transaction management method and device
CN116821952A (en) Privacy data calculation traceability system and method based on block chain consensus mechanism

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