WO2022141865A1 - 汽车道路监测数据处理方法、装置、设备及存储介质 - Google Patents
汽车道路监测数据处理方法、装置、设备及存储介质 Download PDFInfo
- Publication number
- WO2022141865A1 WO2022141865A1 PCT/CN2021/083954 CN2021083954W WO2022141865A1 WO 2022141865 A1 WO2022141865 A1 WO 2022141865A1 CN 2021083954 W CN2021083954 W CN 2021083954W WO 2022141865 A1 WO2022141865 A1 WO 2022141865A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- monitoring
- node
- smart car
- credit score
- encrypted information
- Prior art date
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 418
- 238000003672 processing method Methods 0.000 title abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 75
- 230000008859 change Effects 0.000 claims abstract description 74
- 238000000034 method Methods 0.000 claims abstract description 40
- 230000004044 response Effects 0.000 claims description 28
- 230000002776 aggregation Effects 0.000 claims description 9
- 238000004220 aggregation Methods 0.000 claims description 9
- 238000012790 confirmation Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000007774 longterm Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Definitions
- the present application relates to the technical field of cloud monitoring, and in particular, to a method, device, device and storage medium for processing vehicle road monitoring data.
- automated intelligent vehicles have become the main trend of future automotive technology development.
- automated cars generate a variety of images and numbers when driving, such as driving records, road conditions, data from the car itself, and even weather conditions.
- the inventor realizes that, in the prior art, to monitor the road or monitor the smart car, the smart car driving on the road is generally constrained by fixed traffic rules, so that after the smart car violates the traffic rules, Punish the driver; however, the above method in the prior art is only a time-interrupted processing method for the driver or the smart car, for the driver or the smart car, there is no long-term, Comprehensive and effective and accurate monitoring and evaluation data, so the true intelligence of smart cars cannot be realized.
- the embodiments of the present application provide a method, device, equipment and storage medium for processing road monitoring data of automobiles, so as to monitor and evaluate data effectively and accurately in a long-term, comprehensive manner, thereby realizing the true intelligence of a smart car.
- Monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change in the credit score of the smart car, record the monitoring node that detects the change in the credit score of the smart car as Iteratively propose nodes;
- Make the iterative proposal node send data iteration instructions to all the monitoring nodes except the iterative proposal node, so as to obtain the real-time road conditions of the driving sections corresponding to all the monitoring nodes, and the driving records of the smart car ;
- a method for processing vehicle road monitoring data, applied to a cloud service end node comprising:
- each of the monitoring nodes After each of the monitoring nodes receives the update parameter, and more than a preset number of monitoring nodes determine the validity of the update parameter, it is determined that the change in the credit score of the smart car is valid.
- a vehicle road monitoring data processing device which is applied to a monitoring node of a terminal, wherein the monitoring nodes of the terminal are multiple, including:
- the credit score monitoring module is used to monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change of the credit score of the smart car, the credit score of the smart car will be monitored to change.
- the monitoring nodes of are recorded as iteratively proposed nodes;
- a data acquisition module configured to make the iterative proposal node send data iteration instructions to all the monitoring nodes except the iterative proposal node, so as to acquire the real-time road conditions of the driving sections corresponding to all the monitoring nodes, and the driving record of the smart car;
- a model gradient loss determination module configured to determine the model gradient and model loss of each of the monitoring nodes according to the real-time road conditions and the driving record of the smart car;
- An encrypted information sending module configured to encrypt the model gradient and the model loss to obtain encrypted information, and send the encrypted information to the cloud server node, so that the cloud server node can perform encryption processing according to each Monitor the encrypted information of the node to determine the update parameters;
- the update parameter receiving module is used to receive the update parameter sent by the cloud server node, and when more than a preset number of monitoring nodes determine that the update parameter is valid, generate a parameter block that stores the update parameter, and The parameter block is added to the tail end of the blockchain associated with each of the monitoring nodes to determine that the change in the credit score of the smart car is valid.
- a vehicle road monitoring data processing device applied to a cloud service end node, includes:
- the encrypted information receiving module is used to receive the encrypted information sent by each monitoring node in response to the change of the credit score of the smart car; the encrypted information is after each monitoring node receives the data iteration instruction, according to the The model gradient and model loss corresponding to the monitoring nodes are determined; the model gradient and the model loss are determined according to the real-time road conditions of the driving section corresponding to each monitoring node and the driving record of the smart car;
- an update parameter sending module configured to determine an update parameter according to the encrypted information of each of the monitoring nodes, and send the update parameter to each of the monitoring nodes;
- a credit score change confirmation module configured to determine that the credit score change of the smart car is valid after each monitoring node receives the update parameter and more than a preset number of monitoring nodes determine the validity of the update parameter.
- a computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, applied to a monitoring node of a terminal, wherein the monitoring nodes of the terminal are multiple, wherein , the processor implements the following steps when executing the computer-readable instructions:
- Monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change in the credit score of the smart car, record the monitoring node that detects the change in the credit score of the smart car as Iteratively propose nodes;
- the iteratively proposed node to send data iteration instructions to the other monitoring nodes except the iteratively proposed node, so as to obtain the real-time road conditions of the driving sections corresponding to all the monitoring nodes and the driving records of the smart car;
- the tail end of the blockchain associated with each of the monitoring nodes is used to determine that the change in the credit score of the smart car is valid.
- a computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, applied in a cloud server node, wherein the processor executes the computer
- the following steps are implemented when readable instructions:
- each of the monitoring nodes After each of the monitoring nodes receives the update parameter, and more than a preset number of monitoring nodes determine that the update parameter is valid, it is determined that the change in the credit score of the smart car is valid.
- One or more readable storage media storing computer-readable instructions, applied to the monitoring node of the terminal, the monitoring node of the terminal is multiple, wherein, when the computer-readable instructions are executed by one or more processors , causing the one or more processors to perform the following steps:
- Monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change in the credit score of the smart car, record the monitoring node that detects the change in the credit score of the smart car as Iteratively propose nodes;
- the iteratively proposed node to send data iteration instructions to the other monitoring nodes except the iteratively proposed node, so as to obtain the real-time road conditions of the driving sections corresponding to all the monitoring nodes and the driving records of the smart car;
- the tail end of the blockchain associated with each of the monitoring nodes is used to determine that the change in the credit score of the smart car is valid.
- One or more readable storage media storing computer-readable instructions, which are applied in a cloud service end node, wherein, when the computer-readable instructions are executed by one or more processors, the one or more processing The device performs the following steps:
- each of the monitoring nodes After each of the monitoring nodes receives the update parameter, and more than a preset number of monitoring nodes determine that the update parameter is valid, it is determined that the change in the credit score of the smart car is valid.
- the present application can improve the effectiveness and accuracy of the processing of vehicle road monitoring data. It can also effectively protect the safety of the driving record of the smart car.
- FIG. 1 is a schematic diagram of an application environment of a method for processing vehicle road monitoring data in an embodiment of the present application
- FIG. 2 is a flowchart of a method for processing vehicle road monitoring data in an embodiment of the present application
- FIG. 3 is another flowchart of a method for processing vehicle road monitoring data in an embodiment of the present application.
- FIG. 5 is a flowchart of another method for processing vehicle road monitoring data in an embodiment of the present application.
- FIG. 6 is a schematic block diagram of a vehicle road monitoring data processing device in an embodiment of the present application.
- FIG. 7 is a schematic block diagram of another vehicle road monitoring data processing device in an embodiment of the present application.
- FIG. 8 is a schematic diagram of a computer device in an embodiment of the present application.
- the vehicle road monitoring data processing method provided in the embodiment of the present application can be applied to the application environment shown in FIG. 1 , that is, the vehicle road monitoring data processing method can be applied to the monitoring nodes of the terminal and In the cloud server node of the cloud server, a communication connection is made between the monitoring node and the cloud server node for data transmission.
- the terminal can be installed on but not limited to personal computers and notebook computers.
- the cloud server can be implemented by an independent server or a server cluster composed of multiple servers.
- a method for processing vehicle road monitoring data is provided, and the method is applied to the monitoring node in FIG. 1 as an example for description, including the following steps:
- S10 Monitor in real time whether the credit scores of the smart cars within the scope of the monitoring nodes are changed, and in response to the changes in the credit scores of the smart cars, record the monitoring nodes that monitor the changes in the credit scores of the smart cars as Iteratively proposes nodes.
- the monitoring node refers to a node deployed in the blockchain network of the terminal, and the monitoring node is used to record the driving behavior of the smart car and detect whether the credit score of the smart car changes.
- a monitoring node is associated with one or more roadside monitoring instruments, and each roadside monitoring instrument is responsible for monitoring different driving sections, so as to obtain the driving records and current road conditions of the smart car while driving on the corresponding driving section.
- a smart car refers to a car that has added sensors such as radar sensors and camera sensors, controllers, and actuators.
- the smart car can automatically record various images or data such as driving records during driving.
- the credit score refers to the score for the driving record of each smart car, that is, a smart car is associated with a credit score, and a smart car is associated with a vehicle credit account.
- the vehicle credit accounts of all smart cars are stored in the On the credit ledger of the automobile road monitoring data processing system based on blockchain; exemplarily, if the driver of the smart car violates the traffic rules, the credit score of the smart car will be reduced; or the driving of the smart car If there is no violation of traffic rules for a period of time (such as 30 days), the credit score of the smart car can be improved; therefore, whether it is to improve the credit score of the smart car or reduce the credit score of the smart car, it depends on The credit score for smart cars has changed.
- S20 Make the iterative proposal node send data iteration instructions to the monitoring nodes other than the iterative proposal node, so as to obtain the real-time road conditions of the driving sections corresponding to all the monitoring nodes, and the driving conditions of the smart car Record;
- a monitoring node is associated with a driving section, that is, the driving sections that each monitoring node is responsible for monitoring are different, and then when one of the monitoring nodes detects that the credit score of the smart car changes, the proposed node sends data through this iteration. Iterative instructions are sent to other monitoring nodes to obtain real-time road conditions of driving sections corresponding to all monitoring nodes, as well as driving records of smart cars whose credit scores have changed.
- a monitoring node is associated with a roadside monitoring instrument, and then the roadside monitoring instrument can obtain the real-time road conditions of the corresponding driving section and the driving records of the smart car;
- the real-time road conditions can be road conditions (such as road congestion) , the road is smooth, or whether there is a traffic accident, etc.), or the weather conditions of the road corresponding to the driving section (such as whether the road surface is slippery, current temperature, weather, etc.).
- Driving records include, but are not limited to, the driving speed, origin or destination of the smart car.
- S30 Determine the model gradient and model loss of each of the monitoring nodes according to the real-time road conditions and the driving record of the smart car;
- each monitoring node there is an independently modeled monitoring model in each monitoring node, that is, after the credit score of the monitored smart car changes, each monitoring node performs local modeling to obtain monitoring results. model; after obtaining the corresponding real-time road conditions and the driving records of the smart car, input the real-time road conditions and the driving records of the smart car into the monitoring model for learning, and obtain the model gradient and model loss output by the monitoring model; further, each The real-time road conditions obtained by the monitoring nodes and the driving records of the smart car may be different. Therefore, the model gradient and model loss output by the monitoring model of each monitoring node may also be different.
- the monitoring model may be a model constructed based on a recurrent neural network, and the monitoring model is used for iterative training according to the real-time road conditions obtained by each monitoring node and the driving records of the smart car, thereby obtaining the model gradient and model loss.
- the model loss can be determined using a mean square error loss function. As the number of iterative updates of the monitoring model increases, the model gradient decreases gradually.
- S40 Encrypt the model gradient and the model loss to obtain encrypted information, and send the encrypted information to the cloud server node, so that the cloud server node can perform encryption processing according to the encrypted information of each monitoring node Determine update parameters;
- a homomorphic encryption algorithm such as a uniform hash algorithm, etc. may be used to encrypt the model gradient and the model loss.
- a uniform encryption algorithm such as a uniform hash algorithm
- each monitoring node encrypts the model gradient and model loss, and after obtaining the encrypted information, sends the encrypted information to the cloud server node, and then after the cloud server node receives the encrypted information, performs encryption on each encrypted information.
- the federated aggregation process is performed, and the cloud service model of the cloud server node is constructed according to the encrypted information after the federated aggregation processing; parameters, and send the updated parameters to each monitoring node through the cloud server node, so that each monitoring node can determine the validity of the updated parameters.
- the update parameters include the gradient update parameters corresponding to the model gradient and the loss update parameters corresponding to the model loss.
- S50 Receive the update parameter sent by the cloud server node, and when more than a preset number of monitoring nodes determine that the update parameter is valid, generate a parameter block for storing the update parameter, and store the parameter area in the parameter block. A block is added to the tail of the blockchain associated with each of the monitoring nodes to determine that the change in the credit score of the smart car is valid.
- the preset number is 2/3 of all monitoring nodes, and it is exemplarily assumed that there are 9 monitoring nodes in total, and after more than 6 monitoring nodes (such as 7, 8, etc.) are required to determine the validity of the update parameters, Generates a parameter block that stores update parameters.
- a monitoring node is associated with a blockchain, that is, each monitoring node is associated with a blockchain containing multiple blocks, and the last block of the blockchain represents the number of federated learning iterations, that is, Each time an update parameter sent by the cloud server node is received, and when more than a preset number of monitoring nodes determine that the update parameter is valid, a parameter block for storing the update parameter is generated, and the parameter block is added to the To the end of the blockchain associated with each monitoring node, and the last block represents the latest updated parameters after iteration, so as to determine that the change of the credit score of the smart car is valid.
- the update parameter cannot be approved by most monitoring nodes, and the update parameter is not stored in the new parameter block. If the number of iterations is not updated, the verification of the change of the credit score of the smart car is unsuccessful, and then the credit score of the smart car is restored to the score before the change. In this way, the credit score of the smart car can be changed through a monitoring node and The cloud server node performs verification to improve the validity and accuracy of vehicle road monitoring data processing.
- the current credit score is determined. Judging when changes occur can improve the effectiveness and accuracy of vehicle road monitoring data processing. Further, through the combination of blockchain and federated learning, the security of the driving records of smart cars can also be effectively protected.
- step S50 that is, after adding the parameter block to the end of the blockchain associated with each monitoring node, it includes:
- the number of node iterations is the total number of blocks in the blockchain associated with each monitoring node.
- step S50 it is pointed out that after receiving the update parameters sent by the cloud server node, and when more than a preset number of monitoring nodes determine the validity of the update parameters, a parameter block for storing the update parameters is generated, That is to say, each time a parameter block for storing update parameters is generated, it can be regarded as an iterative process, and then by obtaining the number of blocks corresponding to the blocks associated with each monitoring node, the corresponding iteration number of each node can be obtained.
- all monitoring nodes need to ensure that the number of node iterations of the blockchain associated with them is the latest, that is, the number of blocks of each monitoring node is the same, and when the number of iterations of each node is different, the largest value will be set.
- the blockchain corresponding to the number of node iterations is recorded as the replacement blockchain, and other blockchains except the replacement blockchain are recorded as the blockchain to be replaced; and all the blockchains to be replaced are replaced Replace the blockchain for the described.
- step S50 that is, after determining that the change of the credit score of the smart car is valid, the method further includes:
- S01 Acquire a credit score corresponding to the driver's smart car in response to a pre-travel instruction including a travel route sent by the driver;
- the pre-travel instruction refers to the instruction sent by the driver to the vehicle road monitoring data processing system before the trip, and after the vehicle road monitoring data processing system receives the pre-travel instruction, the monitoring node will be sent from the vehicle road monitoring data processing system.
- obtain the credit score corresponding to the driver's smart car may be obtained from the credit account book by matching the driver's name or the license plate number.
- the travel route may include a departure point, a destination point, and a driving section that needs to be passed between the departure point and the destination point.
- the preset score threshold may be 90 points, 95 points, or the like.
- the credit score is compared with the preset score threshold, and if the credit score is higher than or equal to the preset score threshold, obtain the current road conditions (such as congestion, slippage, weather conditions, etc.) corresponding to the driving section associated with the travel route in the pre-travel instruction, and obtain the credits of all smart cars on the driving section score.
- the current road conditions such as congestion, slippage, weather conditions, etc.
- the monitoring node can capture the current road conditions of the associated driving section and capture images of all smart cars on the driving section to obtain all the information in the image.
- the smart car performs license plate image recognition to obtain the license plate numbers corresponding to all smart cars on the driving section, and then obtains the credit score corresponding to each license plate number from the credit book in the automobile road monitoring data processing system.
- S04 Determine the recommended value of the travel route according to the current road conditions and the credit scores of all the smart cars, and send the recommended value to the driver, so that the driver can follow the recommendation The value determines whether to continue traveling on the travel route.
- the monitoring node After obtaining the current road conditions corresponding to the driving section associated with the travel route, and obtaining the credit scores of all smart cars on the driving section, the monitoring node, according to the current road conditions and the credit scores of all smart cars, Determine recommended values for travel routes.
- the monitoring node may, according to the current road conditions, for example, the recommended value corresponding to sunny weather is higher, the recommended value for rainy weather is lower, the recommended value is lower when the road is congested, and the recommended value is lower when the road is clear; And after obtaining the credit scores of all smart cars, compare the credit scores of each smart car with the preset score threshold. If the credit score is lower than the preset score threshold, it means that the driver on this travel route has more violations, and the probability of accidents may be high. Then, according to the current road conditions and the credit scores of all the smart cars, determine The recommended value of the travel route is sent, and the recommended value is sent to the driver, so that the driver can determine whether to continue to travel on the travel route according to the recommended value.
- the monitoring node can also recommend a suitable travel route for the driver according to the departure and destination in the driver's travel route. , for the driver to determine whether to travel with the travel route.
- a method for processing vehicle road monitoring data is proposed, and the method is applied to the cloud server node in FIG. 1 as an example for description, including the following steps:
- S90 Receive encrypted information sent by each of the monitoring nodes in response to a change in the credit score of the smart car; the encrypted information is based on a model corresponding to each of the monitoring nodes after each of the monitoring nodes receives the data iteration instruction The gradient and the model loss are determined; the model gradient and the model loss are determined according to the real-time road conditions of the driving section corresponding to each monitoring node and the driving record of the smart car;
- the credit score refers to the score for the driving record of each smart car, that is, a smart car is associated with a credit score, and a smart car is associated with a vehicle credit account.
- the vehicle credit accounts of all smart cars are stored in the On the credit ledger of the automobile road monitoring data processing system based on blockchain; exemplarily, if the driver of the smart car violates the traffic rules, the credit score of the smart car will be reduced; or the driving of the smart car If there is no violation of traffic rules for a period of time (such as 30 days), the credit score of the smart car can be improved; therefore, whether it is to improve the credit score of the smart car or reduce the credit score of the smart car, it depends on The credit score for smart cars has changed.
- the monitoring node detects a change in the credit score of the smart car
- the monitoring node that detects the change in the credit score of the smart car sends a data iteration instruction to all other monitoring nodes, so that all monitoring nodes obtain the corresponding driving Real-time road conditions of the road section, and driving records corresponding to the smart car.
- there is an independently modeled monitoring model in each monitoring node that is, after the credit score of the monitored smart car changes, each monitoring node performs local modeling to obtain monitoring results.
- each The real-time road conditions obtained by the monitoring nodes and the driving records of the smart car may be different. Therefore, the model gradient and model loss output by the monitoring model of each monitoring node may also be different.
- each monitoring node determines the corresponding model gradient and model loss
- each monitoring node encrypts the model gradient and model loss through an encryption algorithm such as a homomorphic encryption algorithm, a uniform encryption algorithm (such as a uniform hash algorithm), and then obtains Encrypt the information, and send the encrypted information to the cloud server node.
- an encryption algorithm such as a homomorphic encryption algorithm, a uniform encryption algorithm (such as a uniform hash algorithm), and then obtains Encrypt the information, and send the encrypted information to the cloud server node.
- S100 Determine update parameters according to encrypted information of each monitoring node, and send the update parameters to each monitoring node;
- the monitoring nodes after receiving the encrypted information sent by each of the monitoring nodes, perform federated aggregation processing on each encrypted information, and construct a cloud service model of the cloud server node according to the encrypted information after the federated aggregation processing; For example, after the model gradient and model loss of each monitoring node are weighted and averaged, the update parameters are determined, and the update parameters are sent to each monitoring node, so that each monitoring node can determine the validity of the update parameters.
- the encrypted information for each monitoring node is determined according to the current road conditions and driving records of the corresponding driving section obtained by each monitoring node, that is, the encrypted information for each monitoring node is different, but the The data features are the same (for example, each monitoring node collects current road conditions and driving records), so in this embodiment, a horizontal federated learning method is used to determine the update parameters.
- the preset number is 2/3 of all monitoring nodes, and it is exemplarily assumed that there are 9 monitoring nodes in total, and after more than 6 monitoring nodes (such as 7, 8, etc.) are required to determine the validity of the update parameters, Generates a parameter block that stores update parameters.
- a monitoring node is associated with a blockchain, that is, each monitoring node is associated with a blockchain containing multiple blocks, and the last block of the blockchain represents the number of federated learning iterations, that is, Each time an update parameter sent by the cloud server node is received, and when more than a preset number of monitoring nodes determine the validity of the update parameter, a parameter block for storing the update parameter is generated, and the parameter block is stored in the parameter block. It is added to the tail end of the blockchain associated with each monitoring node, and the last block represents the latest updated parameters after iteration, so as to determine that the change of the credit score of the smart car is valid.
- the method before receiving the encrypted information sent by each monitoring node, the method includes:
- the vehicle road monitoring data processing system includes a credit book for storing vehicle credit accounts; one vehicle credit account is associated with a smart car; one smart car is associated with one credit score;
- the credit book is a distributed storage database based on blockchain.
- a vehicle credit account associated with the smart car is established in the credit book, and a vehicle credit account is associated with a credit score.
- the credit score will be based on the driver's credit score. Adjusting driving habits, such as illegal driving habits will reduce credit scores, good driving habits can improve credit scores.
- s i is the credit score associated with the i-th vehicle credit account in the credit book;
- ⁇ i refers to the standardized error cost corresponding to the i-th vehicle credit account , the number of violations of the labeling rules is determined);
- ⁇ i refers to the damage index corresponding to the ith vehicle credit account (the damage index can be determined according to the loss caused by the violation of traffic rules by the ith vehicle credit account in history) , such as the loss of the current smart car, as well as the loss of other smart cars, etc.);
- ⁇ refers to the current traffic condition;
- ⁇ T refers to the number of times the credit score of the i-th vehicle credit account drops during the T time period.
- sp is the credit score associated with the p -th vehicle credit account in the credit book
- t is the number of days in which the credit score associated with the p-th vehicle credit account has not experienced a credit score reduction.
- the monitoring nodes are deployed in the vehicle road monitoring data processing system; one of the monitoring nodes is associated with one of the driving sections.
- one monitoring node is associated with one driving section, that is, the driving sections that each monitoring node is responsible for monitoring are different.
- a monitoring node is associated with a roadside monitoring instrument so that the roadside monitoring instrument can obtain the real-time road conditions of the corresponding driving section and the driving record of the smart car.
- a monitoring node is also associated with a block chain containing multiple blocks, receives the update parameters sent by the cloud server node at each monitoring node, and determines the update parameters at the monitoring nodes exceeding a preset number. When valid, generate a parameter block storing the updated parameters, and add the parameter block to the end of the blockchain associated with each monitoring node to determine that the change in the credit score of the smart car is valid .
- a vehicle road monitoring data processing apparatus which is applied to a monitoring node of a terminal, and the vehicle road monitoring data processing apparatus corresponds one-to-one with the vehicle road monitoring data processing method in the above embodiment.
- the vehicle road monitoring data processing device includes a credit score monitoring module 10 , a data acquisition module 20 , a model gradient loss determination module 30 , an encrypted information sending module 40 and an update parameter receiving module 50 .
- the detailed description of each functional module is as follows:
- the credit score monitoring module 10 is used to monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change of the credit score of the smart car, it will monitor the changes in the credit score of the smart car.
- the monitoring node is recorded as iteratively proposed node;
- the first data acquisition module 20 is configured to make the iterative proposal node send data iteration instructions to the monitoring nodes other than the iterative proposal node, so as to acquire the real-time road conditions of the driving sections corresponding to all the monitoring nodes, and the driving record of the smart car;
- a model gradient loss determination module 30 configured to determine the model gradient and model loss of each of the monitoring nodes according to the real-time road conditions and the driving record of the smart car;
- the encrypted information sending module 40 is configured to perform encryption processing on the model gradient and the model loss to obtain encrypted information, and send the encrypted information to the cloud server node, so that the cloud server node can perform encryption processing according to each
- the encrypted information of the monitoring node is used to determine the update parameters;
- the update parameter receiving module 50 is configured to receive the update parameter sent by the cloud server node, and when more than a preset number of monitoring nodes determine that the update parameter is valid, generate a parameter block that stores the update parameter, The parameter block is added to the tail end of the block chain associated with each monitoring node to determine that the change in the credit score of the smart car is valid.
- the vehicle road monitoring data processing device further includes:
- a node iteration number acquisition module used to acquire the node iteration numbers corresponding to the blockchains associated with each monitoring node
- the block chain recording module is used to record the block chain corresponding to the node iteration number with the largest value as the replacement block chain when the number of iterations of each node is different, and record the block chain other than the replacement block chain
- the chain record is the blockchain to be replaced;
- a blockchain replacement module configured to replace all the blockchains to be replaced with the replacement blockchains.
- the vehicle road monitoring data processing device further includes:
- a credit score obtaining module configured to obtain a credit score corresponding to the driver's smart car when receiving a pre-travel instruction including a travel route sent by the driver;
- a credit score comparison module for comparing the credit score with a preset score threshold
- the second data acquisition module is configured to acquire the current road conditions corresponding to the driving section associated with the travel route when the credit score is higher than or equal to the preset score threshold, and acquire all the information related to the driving section on the driving section. Credit scores for smart cars;
- a recommended value determination module configured to determine the recommended value of the travel route according to the current road conditions and the credit scores of all the smart cars, and send the recommended value to the driver, so that the driver can The operator determines whether to continue to travel on the travel route according to the recommended value.
- FIG. 7 another vehicle road monitoring data processing device is proposed, which is applied to a cloud server node, including:
- the encrypted information receiving module 90 is configured to receive the encrypted information sent by each of the monitoring nodes in response to the change of the credit score of the smart car; the encrypted information is after each of the monitoring nodes receives the data iteration instruction, according to the The model gradient and the model loss corresponding to the monitoring node are determined; the model gradient and the model loss are determined according to the real-time road conditions of the driving section corresponding to each of the monitoring nodes and the driving record of the smart car;
- an update parameter sending module 100 configured to determine update parameters according to encrypted information of each monitoring node, and send the update parameters to each monitoring node;
- the credit score change confirmation module 110 is configured to determine that the credit score change of the smart car is valid after each monitoring node receives the update parameter and more than a preset number of monitoring nodes determine the validity of the update parameter .
- the vehicle road monitoring data processing device further includes:
- the automobile road monitoring data processing system includes a credit book for storing vehicle credit accounts; one of the vehicle credit accounts is associated with a smart car; A smart car is associated with a credit score;
- a monitoring node deployment module is used to deploy the monitoring nodes in the vehicle road monitoring data processing system; one of the monitoring nodes is associated with one of the driving sections.
- the encrypted information receiving module includes:
- a federated aggregation unit for performing federated aggregation processing on each of the encrypted information to construct a cloud service model
- the federated splitting unit is used for federated splitting of the cloud service model and determining update parameters of the cloud service model.
- Each module in the above-mentioned vehicle road monitoring data processing device can be implemented in whole or in part by software, hardware and combinations thereof.
- the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
- a computer device is provided, and the computer device may be a server, and its internal structure diagram may be as shown in FIG. 8 .
- the computer device includes a processor, memory, a network interface, and a database connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities.
- the memory of the computer device includes a readable storage medium, an internal memory.
- the readable storage medium stores an operating system, computer readable instructions and a database.
- the internal memory provides an environment for the execution of the operating system and computer-readable instructions in the readable storage medium.
- the database of the computer device is used to store the data used in the processing of the vehicle road monitoring data in the above embodiment.
- the network interface of the computer device is used to communicate with an external terminal through a network connection.
- the computer-readable instructions when executed by the processor, implement a method for processing vehicle road monitoring data.
- the readable storage medium provided by this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium
- a computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, applied to a monitoring node of a terminal, the terminal There are multiple monitoring nodes, wherein the processor implements the following steps when executing the computer-readable instructions:
- Monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change in the credit score of the smart car, record the monitoring node that detects the change in the credit score of the smart car as Iteratively propose nodes;
- the iteratively proposed node to send data iteration instructions to the other monitoring nodes except the iteratively proposed node, so as to obtain the real-time road conditions of the driving sections corresponding to all the monitoring nodes and the driving records of the smart car;
- the tail end of the blockchain associated with each of the monitoring nodes is used to determine that the change in the credit score of the smart car is valid.
- another computer device including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, and applied in a cloud server node, wherein , the processor implements the following steps when executing the computer-readable instructions:
- each of the monitoring nodes After each of the monitoring nodes receives the update parameter, and more than a preset number of monitoring nodes determine that the update parameter is valid, it is determined that the change in the credit score of the smart car is valid.
- one or more readable storage media storing computer-readable instructions are provided and applied to a monitoring node of a terminal, where there are multiple monitoring nodes of the terminal, wherein the computer-readable instructions are When executed by one or more processors, the one or more processors are caused to perform the following steps:
- Monitor in real time whether the credit score of the smart car within the range of each monitoring node changes, and in response to the change in the credit score of the smart car, record the monitoring node that detects the change in the credit score of the smart car as Iteratively propose nodes;
- the iteratively proposed node to send data iteration instructions to the other monitoring nodes except the iteratively proposed node, so as to obtain the real-time road conditions of the driving sections corresponding to all the monitoring nodes and the driving records of the smart car;
- the tail end of the blockchain associated with each of the monitoring nodes is used to determine that the change in the credit score of the smart car is valid.
- another or more readable storage media storing computer-readable instructions are provided for application in a cloud service end node, wherein when the computer-readable instructions are executed by one or more processors , causing the one or more processors to perform the following steps:
- each of the monitoring nodes After each of the monitoring nodes receives the update parameter, and more than a preset number of monitoring nodes determine that the update parameter is valid, it is determined that the change in the credit score of the smart car is valid.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Geometry (AREA)
- Game Theory and Decision Science (AREA)
- Multimedia (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Computer Hardware Design (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
一种汽车道路监测数据处理方法、装置、设备及介质,属于云监控技术领域。该方法令监测到智能汽车的信用评分发生改变的迭代提出节点发送数据迭代指令至除迭代提出节点外的监测节点,获取所有监测节点对应的驾驶路段的实时道路状况及智能汽车的驾驶记录(S20);根据实时道路状况以及驾驶记录,确定各监测节点的模型梯度以及模型损失(S30);对模型梯度以及模型损失进行加密处理得到加密信息,将加密信息发送至云服务端节点,令云服务端节点根据加密信息确定更新参数(S40);在超过预设数量的监测节点确定更新参数有效时,将生成的参数区块加入至各监测节点关联的区块链的尾端,确定智能汽车的信用评分改变有效(S50)。提高了汽车道路监测数据处理的准确性。
Description
本申请要求于2020年12月31日提交中国专利局、申请号为202011640125.2,发明名称为“汽车道路监测数据处理方法、装置、设备及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及云监控技术领域,尤其涉及一种汽车道路监测数据处理方法、装置、设备及存储介质。
随着汽车与人工智能、5G通讯、互联网、传感器技术等领域的快速融合,自动化智能汽车已成为未来汽车技术发展的主要趋势。相对于传统汽车,自动化汽车行驶时会产生各种各样图像和数字等数据,例如:行驶记录,道路状况,汽车本身的数据,甚至是天气状况。
发明人意识到,现有技术中,对道路进行监测亦或者对智能汽车进行监控,一般是通过固定的交通规则,对行驶在道路上的智能汽车进行约束,以在智能汽车存在违反交通规则之后对驾驶员进行惩罚;但是,现有技术中的上述方式仅为一种对于驾驶员或智能汽车的一种即时间断的处理方式,对于该驾驶员或者智能汽车来说,并不存在一个长期、综合且有效准确的监控评价数据,如此,无法实现智能汽车的真正智能。
申请内容
本申请实施例提供一种汽车道路监测数据处理方法、装置、设备及存储介质,以长期、综合且有效准确的监控评价数据,进而实现智能汽车的真正智能。
一种汽车道路监测数据处理方法,应用于终端的监测节点中,所述终端的监测节点为多个,包括:
实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;
令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的所有所述监测节点中,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
一种汽车道路监测数据处理方法,应用于云服务端节点中,包括:
响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路 段的实时道路状况以及所述智能汽车的驾驶记录确定;
根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效性之后,确定所述智能汽车的信用评分改变有效。
一种汽车道路监测数据处理装置,应用于终端的监测节点中,所述终端的监测节点为多个,包括:
信用评分监测模块,用于实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的监测节点记录为迭代提出节点;
数据获取模块,用于令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的所有所述监测节点中,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
模型梯度损失确定模块,用于根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
加密信息发送模块,用于对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
更新参数接收模块,用于接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
一种汽车道路监测数据处理装置,应用于云服务端节点中,包括:
加密信息接收模块,用于响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
更新参数发送模块,用于根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
信用评分变更确认模块,用于在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效性之后,确定所述智能汽车的信用评分改变有效。
一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述处理器执行所述计算机可读指令时实现如下步骤:
实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;
令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,应用于云服务端节点中,其中,所述处理器执行所述计算机可读指令时实现如下步骤:
响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
一个或多个存储有计算机可读指令的可读存储介质,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行如下步骤:
实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;
令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
一个或多个存储有计算机可读指令的可读存储介质,应用于云服务端节点中,其中,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行如下步骤:
响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
本申请,可以提高汽车道路监测数据处理的有效性以及准确性。还可以有效保护智能汽车的驾驶记录的安全性。
图1是本申请一实施例中汽车道路监测数据处理方法的一应用环境示意图;
图2是本申请一实施例中汽车道路监测数据处理方法的一流程图;
图3是本申请一实施例中汽车道路监测数据处理方法的另一流程图;
图4是本申请一实施例中汽车道路监测数据处理方法的又一流程图;
图5是本申请一实施例中另一汽车道路监测数据处理方法的一流程图;
图6是本申请一实施例中汽车道路监测数据处理装置的一原理框图;
图7是本申请一实施例中另一汽车道路监测数据处理装置的一原理框图;
图8是本申请一实施例中计算机设备的一示意图。
本申请实施例提供的汽车道路监测数据处理方法,该汽车道路监测数据处理方法可以应用于如图1所示的应用环境中,也即该汽车道路监测数据处理方法可以应用于终端的监测节点以及云服务端的云服务端节点中,监测节点与云服务端节点之间通信连接,以进行数据传输。其中,终端可安装在但不限于个人计算机、笔记本电脑上。云服务端可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
在一实施例中,如图2所示,提供一种汽车道路监测数据处理方法,以该方法应用在图1中的监测节点为例进行说明,包括如下步骤:
S10:实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的监测节点记录为迭代提出节点。
可以理解地,监测节点指的是被部署在终端的区块链网络中的节点,该监测节点用于对智能汽车的驾驶行为进行记录,以及检测智能汽车的信用评分是否发生改变。一个监测节点关联一个或多个路边监测仪器,每一个路边监测仪器负责监控不同的驾驶路段,进而可以获取智能汽车在对应的驾驶路段行驶时的驾驶记录以及当前道路状况。
可以理解地,智能汽车指的是增加了如雷达传感器、摄像传感器等传感器、控制器以及执行器等装置的汽车,该智能汽车在行驶过程中能够自动记录各种图像或者如驾驶记录等数据。
可以理解地,信用评分指的是针对每一个智能汽车的驾驶记录的评分,也即一个智能汽车关联一个信用评分,且一个智能汽车关联一个车辆信用账号,所有智能汽车的车辆信用账号均存储在基于区块链构建的汽车道路监测数据处理***的信用账本上;示例性地,假设智能汽车的驾驶员存在违反交通规则的行为,则会降低该智能汽车的信用评分;亦或者智能汽车的驾驶员在一段时间内(如30天)连续没有存在违反交通规则的行为,则可以提高该智能汽车的信用评分;因此,无论是提高智能汽车的信用评分或者是降低智能汽车的信用评分,均视为智能汽车的信用评分发生改变。
S20:令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的所述监测节点中,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
可以理解地,一个监测节点关联一个驾驶路段,也即各监测节点负责监控的驾驶路段是不同的,进而在其中一个监测节点监测到智能汽车的信用评分发生改变时,通过该迭代提出节点发送数据迭代指令至其它监测节点中,以获取所有监测节点对应的驾驶路段的实时道路状况,以及发生信用评分改变的智能汽车的驾驶记录。其中,一个监测节点关联一个路边监测仪器,进而该路边监测仪器可以获取与其对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;实时道路状况可以为道路情况(如道路拥堵、道路通畅,或者是否存在交通事故等),也可以为与驾驶路段对应的道路的天气情况(如道路地面是否 湿滑、当前温度、天气等)。驾驶记录包括但不限于智能汽车的行驶速度、出发地或者目的地等。
S30:根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
可以理解地,针对所有监测节点,在每一个监测节点中均存在一个独立建模的监测模型,也即在监测智能汽车的信用评分发生改变之后,每一监测节点均进行本地建模,得到监测模型;获取对应的实时道路状况以及智能汽车的驾驶记录之后,将实时道路状况以及智能汽车的驾驶记录输入至监测模型中进行学习,得到监测模型输出的模型梯度以及模型损失;进一步地,每一个监测节点获取到的实时道路状况以及智能汽车的驾驶记录可能是不同的,因此,每一个监测节点的监测模型输出的模型梯度以及模型损失也可能是不同的。其中,监测模型可以为基于递归神经网络构建的模型,该监测模型用于根据各监测节点获取的实时道路状况以及智能汽车驾驶记录进行迭代训练,进而得到模型梯度以及模型损失。示例性地,模型损失可以采用均方差损失函数确定。随着监测模型的迭代更新次数的增加,模型梯度在逐渐下降。
S40:对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
示例性地,对模型梯度以及模型损失进行加密处理可以采用同态加密算法、均匀加密算法(如均匀哈希算法)等。
进一步地,在各监测节点对模型梯度以及模型损失进行加密处理,得到加密信息后,将加密信息发送至云服务端节点中,进而在云服务端节点接收到加密信息之后,对各加密信息进行联邦聚合处理,并根据联邦聚合处理后的加密信息构建云服务端节点的云服务模型;对云服务模型进行联邦拆分,如对各监测节点的模型梯度以及模型损失进行加权平均后,确定更新参数,并通过云服务端节点将更新参数发送至各监测节点中,以令各监测节点确定更新参数的有效性。其中,更新参数中包括与模型梯度对应的梯度更新参数,以及与模型损失对应的损失更新参数,该更新参数用于令各监测节点在下一次监测到智能汽车的信用评分发生改变时,根据此次发生改变的智能汽车的实时道路状况、智能汽车的驾驶记录,更新参数,对各监测节点的监测模型进行训练,进而在不断的训练过程中,提高各监测节点的监测模型根据实时道路状况以及智能汽车的驾驶记录,对智能汽车的信用评分的改变进行判断的准确率。
S50:接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
优选地,预设数量为所有监测节点的2/3,示例性地的假设监测节点共有9个,则需要超过6个监测节点(如7个、8个等)确定更新参数的有效性之后,生成一个存储更新参数的参数区块。
可以理解地,一个监测节点关联一个区块链,也即每一个监测节点中均关联一个包含多个区块的区块链,该区块链的最后一个区块表征联邦学习迭代数,也即每接收一次云服务端节点发送的更新参数,且在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,进而最后一个区块表征最新的迭代后的更新参数,以确定智能汽车的信用评分改变有效。
进一步地,若少于预设数量的监测节点不确定更新参数的有效性,则表征该更新参数无法得到大部分监测节点的同意,进而不将该更新参数存储至新的参数区块中,此次迭代数不更新,智能汽车的信用评分改变校验不成功,进而将智能汽车的信用评分恢复至改变 之前的评分,如此,可以使得智能汽车的信用评分的改动,可以通过过个监测节点以及云服务端节点进行校验,提高汽车道路监测数据处理的有效性以及准确性。
在本实施例中,通过区块链以及联邦学习结合的方式,并根据智能汽车发生信用评分改变时的当前道路状况以及驾驶记录,进而根据长期且具有综合性的监控数据,对本次信用评分发生改变进行评判,可以提高汽车道路监测数据处理的有效性以及准确性。进一步地,通过区块链以及联邦学习结合的方式,还可以有效保护智能汽车的驾驶记录的安全性。
在一实施例中,如图3所示,步骤S50之后,也即将所述参数区块加入至各所述监测节点关联的区块链的尾端之后,包括:
S60:获取与各所述监测节点关联的区块链分别对应的各节点迭代数;
可以理解地,节点迭代数也即与各监测节点关联的区块链的区块总数。在步骤S50中指出,在接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效性时,生成一个存储所述更新参数的参数区块,也即每一次生成一个存储更新参数的参数区块即可视为一次迭代过程,进而通过获取与各监测节点关联的区块分别对应的区块数,即可得到对应的各节点迭代数。
S70:在各节点迭代数存在不同时,将数值最大的节点迭代数对应的区块链记录为替换区块链,将除所述替换区块链之外的其他区块链记录为待替换区块链;
S80:将所有所述待替换区块链均替换为所述替换区块链。
可以理解地,所有监测节点均需要保证与其关联的区块链的节点迭代数是最新的,也即每一个监测节点的区块数相同,进而在各节点迭代数存在不同时,将数值最大的节点迭代数对应的区块链记录为替换区块链,将除所述替换区块链之外的其他区块链记录为待替换区块链;并将所有所述待替换区块链均替换为所述替换区块链。
在一实施例中,如图4所示,步骤S50之后,也即确定所述智能汽车的信用评分改变有效之后,还包括:
S01:响应于驾驶员发送的包含出行路线的预出行指令,获取与所述驾驶员的智能汽车对应的信用评分;
其中,预出行指令是指驾驶员出行前发送至汽车道路监测数据处理***中的指令,进而在汽车道路监测数据处理***接收到该预出行指令之后,监测节点会自汽车道路监测数据处理***中的信用账本中,获取与该驾驶员的智能汽车对应的信用评分。示例性地,可以通过驾驶员姓名匹配或者车牌号匹配方式,自信用账本中获取与该驾驶员的智能汽车对应的信用评分。可以理解地,出行路线可以包含出发地、目的地,以及出发地至目的地之间需要通过的驾驶路段。
S02:将所述信用评分与预设评分阈值进行比较;
S03:在所述信用评分高于或等于预设评分阈值时,获取与所述出行路线关联的驾驶路段对应的当前道路状况,并获取与所述驾驶路段上所有智能汽车的信用评分;
其中,预设评分阈值可以为90分、95分等。
具体地,在接收到驾驶员发送的包含出行路线的预出行指令时,获取与所述驾驶员的智能汽车对应的信用评分之后,将信用评分与预设评分阈值进行比较,在信用评分高于或等于预设评分阈值时,获取与预出行指令中出行路线关联的驾驶路段对应的当前道路状况(如是否拥堵、是否容易打滑、天气情况等),并获取与驾驶路段上所有智能汽车的信用评分。
示例性地,在信用评分高于或等于预设评分阈值时,监测节点可以通过拍摄关联的驾驶路段的当前道路状况,并通过拍摄驾驶路段上的所有智能汽车的图像,以对该图像中所有智能汽车进行车牌图像识别,以得到与驾驶路段上所有智能汽车对应的车牌号,进而自汽车道路监测数据处理***中的信用账本中,获取与各车牌号对应的信用评分。
S04:根据所述当前道路状况以及所有所述智能汽车的信用评分,确定所述出行路线 的推荐值,并将所述推荐值发送至所述驾驶员,以令所述驾驶员根据所述推荐值确定是否继续以所述出行路线出行。
具体地,在获取与所述出行路线关联的驾驶路段对应的当前道路状况,并获取与所述驾驶路段上所有智能汽车的信用评分之后,监测节点根据当前道路状况以及所有智能汽车的信用评分,确定出行路线的推荐值。
示例性地,监测节点可以根据当前道路状况,如天气晴朗对应的推荐值较高,下雨天气的推荐值较低,道路若出现拥堵情况时推荐值较低,道路通畅时推荐值较低;且在获取所有智能汽车的信用评分之后,将各智能汽车的信用评分与预设评分阈值进行比较,若超过预设数量(如在驾驶路段上的所有智能汽车数量的80%等)的智能汽车的信用评分低于预设评分阈值,则表征该出行线路上的驾驶员存在较多的违规次数,可能发生意外的几率较高,进而根据当前道路状况以及所有所述智能汽车的信用评分,确定所述出行路线的推荐值,并将推荐值发送至驾驶员,以令驾驶员根据所述推荐值确定是否继续以所述出行路线出行。如,在推荐值较低时(如推荐值为50分、60分等),则监测节点也可以根据该驾驶员的出行线路中的出发地以及目的地,为驾驶员推荐一条适合的出行线路,以供驾驶员确定是否以出行路线出行。
在一实施例中,如图5所示,提出一种汽车道路监测数据处理方法,以该方法应用在图1中的云服务端节点为例进行说明,包括如下步骤:
S90:响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
可以理解地,信用评分指的是针对每一个智能汽车的驾驶记录的评分,也即一个智能汽车关联一个信用评分,且一个智能汽车关联一个车辆信用账号,所有智能汽车的车辆信用账号均存储在基于区块链构建的汽车道路监测数据处理***的信用账本上;示例性地,假设智能汽车的驾驶员存在违反交通规则的行为,则会降低该智能汽车的信用评分;亦或者智能汽车的驾驶员在一段时间内(如30天)连续没有存在违反交通规则的行为,则可以提高该智能汽车的信用评分;因此,无论是提高智能汽车的信用评分或者是降低智能汽车的信用评分,均视为智能汽车的信用评分发生改变。
具体地,在监测节点检测到智能汽车的信用评分发生改变时,通过监测到智能汽车的信用评分发生改变的监测节点向所有其它监测节点发送数据迭代指令,以令所有监测节点获取与其对应的驾驶路段的实时道路状况,以及与该智能汽车对应的驾驶记录。可以理解地,针对所有监测节点,在每一个监测节点中均存在一个独立建模的监测模型,也即在监测智能汽车的信用评分发生改变之后,每一监测节点均进行本地建模,得到监测模型;获取对应的实时道路状况以及智能汽车的驾驶记录之后,将实时道路状况以及智能汽车的驾驶记录输入至监测模型中进行学习,得到监测模型输出的模型梯度以及模型损失;进一步地,每一个监测节点获取到的实时道路状况以及智能汽车的驾驶记录可能是不同的,因此,每一个监测节点的监测模型输出的模型梯度以及模型损失也可能是不同的。
进一步地,在各监测节点确定对应的模型梯度以及模型损失之后,各监测节点通过如同态加密算法、均匀加密算法(如均匀哈希算法)等加密算法对模型梯度以及模型损失进行加密,进而得到加密信息,并将该加密信息发送至云服务端节点。
S100:根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
具体地,在接收各所述监测节点发送的加密信息之后,对各加密信息进行联邦聚合处理,并根据联邦聚合处理后的加密信息构建云服务端节点的云服务模型;对云服务模型进行联邦拆分,如对各监测节点的模型梯度以及模型损失进行加权平均后,确定更新参数, 并将更新参数发送至各监测节点中,以令各监测节点确定更新参数的有效性。可以理解地,对于各监测节点的加密信息,是根据各监测节点获取的对应驾驶路段的当前道路状况以及驾驶记录确定的,也即对于各监测节点的加密信息是不同的,但是各监测节点的数据特征相同(如各监测节点均采集当前道路状况以及驾驶记录),因此本实施例中,采用横向联邦学习方式确定更新参数。
S110:在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
优选地,预设数量为所有监测节点的2/3,示例性地的假设监测节点共有9个,则需要超过6个监测节点(如7个、8个等)确定更新参数的有效性之后,生成一个存储更新参数的参数区块。
可以理解地,一个监测节点关联一个区块链,也即每一个监测节点中均关联一个包含多个区块的区块链,该区块链的最后一个区块表征联邦学习迭代数,也即每接收一次云服务端节点发送的更新参数,且在超过预设数量的监测节点确定所述更新参数的有效性时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,进而最后一个区块表征最新的迭代后的更新参数,以确定智能汽车的信用评分改变有效。
在一实施例中,所述接收各所述监测节点发送的加密信息之前,包括:
构建基于区块链的汽车道路监测数据处理***;所述汽车道路监测数据处理***包括用于存储车辆信用账号的信用账本;一个所述车辆信用账号关联一辆智能汽车;一辆智能汽车关联一个信用评分;
可以理解地,信用账本为基于区块链的分布式存储数据库,该信用账本中建立了与智能汽车关联的车辆信用账号,并且一个车辆信用账号关联一个信用评分,该信用评分会根据驾驶员的驾驶习惯进行调整,如违规驾驶习惯会降低信用评分,良好驾驶习惯可以提高信用评分。
进一步地,智能汽车关联的信用评分发生改变时可以通过下述表达式确定:
信用评分下降表达式:s
i-(σ
i+β
i)*μ*α
T;
其中,s
i为信用账本中第i个车辆信用账号关联的信用评分;σ
i指的是第i个车辆信用账号对应的标准化错误代价(该标准化错误代价可以根据历史上第i个车辆信用账号,违反标注化规则的次数确定);β
i指的是第第i个车辆信用账号对应的伤害指数(该伤害指数可以根据历史上第i个车辆信用账号,违反交通规则带来的损耗进行确定,如对本智能汽车的损耗,以及对其它智能汽车的损耗等);μ指的是当前交通状况;α
T指的是在T时间段内,第i个车辆信用账号的信用评分下降次数。
信用评分上升表达式:s
p+log(t+29);
其中,s
p为信用账本中第p个车辆信用账号关联的信用评分;t为第p个车辆信用账号关联的信用评分未发生信用评分降低的天数。
在所述汽车道路监测数据处理***中部署所述监测节点;一个所述监测节点关联一个所述驾驶路段。
可以理解地,一个监测节点关联一个驾驶路段,也即各监测节点负责监控的驾驶路段是不同的。此外,一个监测节点关联一个路边监测仪器进而该路边监测仪器可以获取与其 对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录。进一步地,一个监测节点还关联一个包含多个区块的区块链,在各监测节点接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效性时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
在一实施例中,提供一种汽车道路监测数据处理装置,应用于终端的监测节点中,该汽车道路监测数据处理装置与上述实施例中汽车道路监测数据处理方法一一对应。如图6所示,该汽车道路监测数据处理装置包括信用评分监测模块10、数据获取模块20、模型梯度损失确定模块30、加密信息发送模块40和更新参数接收模块50。各功能模块详细说明如下:
信用评分监测模块10,用于实时监测各监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的监测节点记录为迭代提出节点;
第一数据获取模块20,用于令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的所述监测节点中,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
模型梯度损失确定模块30,用于根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
加密信息发送模块40,用于对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
更新参数接收模块50,用于接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
优选地,汽车道路监测数据处理装置还包括:
节点迭代数获取模块,用于获取与各所述监测节点关联的区块链分别对应的各节点迭代数;
区块链记录模块,用于在各节点迭代数存在不同时,将数值最大的节点迭代数对应的区块链记录为替换区块链,将除所述替换区块链之外的其他区块链记录为待替换区块链;
区块链替换模块,用于将所有所述待替换区块链均替换为所述替换区块链。
优选地,汽车道路监测数据处理装置还包括:
信用评分获取模块,用于在接收到驾驶员发送的包含出行路线的预出行指令时,获取与所述驾驶员的智能汽车对应的信用评分;
信用评分比较模块,用于将所述信用评分与预设评分阈值进行比较;
第二数据获取模块,用于在所述信用评分高于或等于所述预设评分阈值时,获取与所述出行路线关联的驾驶路段对应的当前道路状况,并获取与所述驾驶路段上所有智能汽车的信用评分;
推荐值确定模块,用于根据所述当前道路状况以及所有所述智能汽车的信用评分,确定所述出行路线的推荐值,并将所述推荐值发送至所述驾驶员,以令所述驾驶员根据所述推荐值确定是否继续以所述出行路线出行。
在一实施例中,如图7所示,提出另一种汽车道路监测数据处理装置,应用于云服务端节点中,包括:
加密信息接收模块90,用于响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
更新参数发送模块100,用于根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
信用评分变更确认模块110,用于在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效性之后,确定所述智能汽车的信用评分改变有效。
优选地,汽车道路监测数据处理装置还包括:
***构建模块,用于构建基于区块链的汽车道路监测数据处理***;所述汽车道路监测数据处理***包括用于存储车辆信用账号的信用账本;一个所述车辆信用账号关联一辆智能汽车;一辆智能汽车关联一个信用评分;
监测节点部署模块,用于在所述汽车道路监测数据处理***中部署所述监测节点;一个所述监测节点关联一个所述驾驶路段。
优选地,加密信息接收模块包括:
联邦聚合单元,用于对各所述加密信息进行联邦聚合处理,以构建云服务模型;
联邦拆分单元,用于对所述云服务模型进行联邦拆分,确定所述云服务模型的更新参数。
关于汽车道路监测数据处理装置的具体限定可以参见上文中对于汽车道路监测数据处理方法的限定,在此不再赘述。上述汽车道路监测数据处理装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中,也可以以软件形式存储于计算机设备中的存储器中,以便于处理器调用执行以上各个模块对应的操作。
在一个实施例中,提供了一种计算机设备,该计算机设备可以是服务器,其内部结构图可以如图8所示。该计算机设备包括通过***总线连接的处理器、存储器、网络接口和数据库。其中,该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括可读存储介质、内存储器。该可读存储介质存储有操作***、计算机可读指令和数据库。该内存储器为可读存储介质中的操作***和计算机可读指令的运行提供环境。该计算机设备的数据库用于存储上述实施例中汽车道路监测数据处理所使用到的数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种汽车道路监测数据处理方法。本实施例所提供的可读存储介质包括非易失性可读存储介质和易失性可读存储介质。
在一个实施例中,提供了一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述处理器执行所述计算机可读指令时实现如下步骤:
实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;
令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息 发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
在一个实施例中,提供了另一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,应用于云服务端节点中,其中,所述处理器执行所述计算机可读指令时实现如下步骤:
响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
在一个实施例中,提供了一个或多个存储有计算机可读指令的可读存储介质,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行如下步骤:
实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;
令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;
根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;
对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;
接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
在一个实施例中,提供了另一个或多个存储有计算机可读指令的可读存储介质,应用于云服务端节点中,其中,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行如下步骤:
响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;
根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;
在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过 计算机可读指令来指令相关的硬件来完成,所述的计算机可读指令可存储于一非易失性计算机可读取存储介质或者易失性计算机可读取存储介质中,该计算机可读指令在执行时,可包括如上述各方法的实施例的流程。其中,本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用,均可包括非易失性和/或易失性存储器。
以上所述实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述实施例对本申请进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围,均应包含在本申请的保护范围之内。
Claims (20)
- 一种汽车道路监测数据处理方法,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述方法包括:实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
- 如权利要求1所述的汽车道路监测数据处理方法,其中,所述将所述参数区块加入至各所述监测节点关联的区块链的尾端之后,包括:获取与各所述监测节点关联的区块链分别对应的各节点迭代数;在各节点迭代数存在不同时,将数值最大的节点迭代数对应的区块链记录为替换区块链,将除所述替换区块链之外的其他区块链记录为待替换区块链;将所有所述待替换区块链均替换为所述替换区块链。
- 如权利要求1所述的汽车道路监测数据处理方法,其中,所述确定所述智能汽车的信用评分改变有效之后,还包括:响应于驾驶员发送的包含出行路线的预出行指令,获取与所述驾驶员的智能汽车对应的信用评分;将所述信用评分与预设评分阈值进行比较;在所述信用评分高于或等于所述预设评分阈值时,获取与所述出行路线关联的驾驶路段对应的当前道路状况,并获取与所述驾驶路段上所有智能汽车的信用评分;根据所述当前道路状况以及所有所述智能汽车的信用评分,确定所述出行路线的推荐值,并将所述推荐值发送至所述驾驶员,以令所述驾驶员根据所述推荐值确定是否继续以所述出行路线出行。
- 一种汽车道路监测数据处理方法,所述汽车道路监测数据处理方法应用于云服务端节点中,其中,所述方法包括:响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
- 如权利要求4所述的汽车道路监测数据处理方法,其中,所述接收各所述监测节点 发送的加密信息之前,包括:构建基于区块链的汽车道路监测数据处理***;所述汽车道路监测数据处理***包括用于存储车辆信用账号的信用账本;一个所述车辆信用账号关联一辆智能汽车;一辆智能汽车关联一个信用评分;在所述汽车道路监测数据处理***中部署所述监测节点;一个所述监测节点关联一个所述驾驶路段。
- 如权利要求4所述的汽车道路监测数据处理方法,其中,所述根据各所述监测节点的加密信息确定更新参数,包括:对各所述加密信息进行联邦聚合处理,以构建云服务模型;对所述云服务模型进行联邦拆分,确定所述云服务模型的更新参数。
- 一种汽车道路监测数据处理装置,应用于终端的监测节点中,所述终端的监测节点为多个,其中,所述装置包括:信用评分监测模块,用于实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;数据获取模块,用于令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的所有所述监测节点中,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;模型梯度损失确定模块,用于根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;加密信息发送模块,用于对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;更新参数接收模块,用于接收所述云服务端节点发送的更新参数,并在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
- 一种汽车道路监测数据处理装置,应用于云服务端节点中,其中,所述装置包括:加密信息接收模块,用于响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;更新参数发送模块,用于根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;信用评分变更确认模块,用于在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效性之后,确定所述智能汽车的信用评分改变有效。
- 一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述处理器执行所述计算机可读指令时实现如下步骤:实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
- 如权利要求9所述的计算机设备,其中,所述将所述参数区块加入至各所述监测节点关联的区块链的尾端之后,所述处理器执行所述计算机可读指令时还实现如下步骤:获取与各所述监测节点关联的区块链分别对应的各节点迭代数;在各节点迭代数存在不同时,将数值最大的节点迭代数对应的区块链记录为替换区块链,将除所述替换区块链之外的其他区块链记录为待替换区块链;将所有所述待替换区块链均替换为所述替换区块链。
- 如权利要求9所述的计算机设备,其中,所述确定所述智能汽车的信用评分改变有效之后,所述处理器执行所述计算机可读指令时还实现如下步骤:响应于驾驶员发送的包含出行路线的预出行指令,获取与所述驾驶员的智能汽车对应的信用评分;将所述信用评分与预设评分阈值进行比较;在所述信用评分高于或等于所述预设评分阈值时,获取与所述出行路线关联的驾驶路段对应的当前道路状况,并获取与所述驾驶路段上所有智能汽车的信用评分;根据所述当前道路状况以及所有所述智能汽车的信用评分,确定所述出行路线的推荐值,并将所述推荐值发送至所述驾驶员,以令所述驾驶员根据所述推荐值确定是否继续以所述出行路线出行。
- 一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,应用于云服务端节点中,其中,所述处理器执行所述计算机可读指令时实现如下步骤:响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
- 如权利要求12所述的计算机设备,其中,所述接收各所述监测节点发送的加密信息之前,所述处理器执行所述计算机可读指令时还实现如下步骤:构建基于区块链的汽车道路监测数据处理***;所述汽车道路监测数据处理***包括用于存储车辆信用账号的信用账本;一个所述车辆信用账号关联一辆智能汽车;一辆智能汽车关联一个信用评分;在所述汽车道路监测数据处理***中部署所述监测节点;一个所述监测节点关联一个所述驾驶路段。
- 如权利要求12所述的计算机设备,其中,所述根据各所述监测节点的加密信息确定更新参数,包括:对各所述加密信息进行联邦聚合处理,以构建云服务模型;对所述云服务模型进行联邦拆分,确定所述云服务模型的更新参数。
- 一个或多个存储有计算机可读指令的可读存储介质,应用于终端的监测节点,所述终端的监测节点为多个,其中,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行如下步骤:实时监测各所述监测节点范围内的智能汽车的信用评分是否发生改变,响应于所述智能汽车的信用评分的改变,将监测到所述智能汽车的信用评分发生改变的所述监测节点记录为迭代提出节点;令所述迭代提出节点发送数据迭代指令至除所述迭代提出节点外的其余所述监测节点,以获取所有所述监测节点对应的驾驶路段的实时道路状况,以及所述智能汽车的驾驶记录;根据所述实时道路状况以及所述智能汽车的驾驶记录,确定各所述监测节点的模型梯度以及模型损失;对所述模型梯度以及所述模型损失进行加密处理,得到加密信息,并将所述加密信息发送至云服务端节点,以令所述云服务端节点根据各所述监测节点的加密信息确定更新参数;接收所述云服务端节点发送的更新参数,在超过预设数量的监测节点确定所述更新参数的有效时,生成一个存储所述更新参数的参数区块,并将所述参数区块加入至各所述监测节点关联的区块链的尾端,以确定所述智能汽车的信用评分改变有效。
- 如权利要求15所述的可读存储介质,其中,所述将所述参数区块加入至各所述监测节点关联的区块链的尾端之后,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器还执行如下步骤:获取与各所述监测节点关联的区块链分别对应的各节点迭代数;在各节点迭代数存在不同时,将数值最大的节点迭代数对应的区块链记录为替换区块链,将除所述替换区块链之外的其他区块链记录为待替换区块链;将所有所述待替换区块链均替换为所述替换区块链。
- 如权利要求15所述的可读存储介质,其中,所述确定所述智能汽车的信用评分改变有效之后,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器还执行如下步骤:响应于驾驶员发送的包含出行路线的预出行指令,获取与所述驾驶员的智能汽车对应的信用评分;将所述信用评分与预设评分阈值进行比较;在所述信用评分高于或等于所述预设评分阈值时,获取与所述出行路线关联的驾驶路段对应的当前道路状况,并获取与所述驾驶路段上所有智能汽车的信用评分;根据所述当前道路状况以及所有所述智能汽车的信用评分,确定所述出行路线的推荐值,并将所述推荐值发送至所述驾驶员,以令所述驾驶员根据所述推荐值确定是否继续以所述出行路线出行。
- 一个或多个存储有计算机可读指令的可读存储介质,应用于云服务端节点中,其中,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行如下步骤:响应于智能汽车信用评分的改变,接收各所述监测节点发送的加密信息;所述加密信息是在各所述监测节点接收到数据迭代指令之后,根据与各所述监测节点对应的模型梯度以及模型损失确定的;所述模型梯度以及所述模型损失根据各所述监测节点对应的驾驶路段的实时道路状况以及所述智能汽车的驾驶记录确定;根据各所述监测节点的加密信息确定更新参数,并将所述更新参数发送至各所述监测节点;在各所述监测节点接收到所述更新参数,且超过预设数量的监测节点确定所述更新参数的有效之后,确定所述智能汽车的信用评分改变有效。
- 如权利要求18所述的可读存储介质,其中,所述接收各所述监测节点发送的加密信息之前,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器还执行如下步骤:构建基于区块链的汽车道路监测数据处理***;所述汽车道路监测数据处理***包括用于存储车辆信用账号的信用账本;一个所述车辆信用账号关联一辆智能汽车;一辆智能汽车关联一个信用评分;在所述汽车道路监测数据处理***中部署所述监测节点;一个所述监测节点关联一个所述驾驶路段。
- 如权利要求18所述的可读存储介质,其中,所述根据各所述监测节点的加密信息确定更新参数,包括:对各所述加密信息进行联邦聚合处理,以构建云服务模型;对所述云服务模型进行联邦拆分,确定所述云服务模型的更新参数。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011640125.2 | 2020-12-31 | ||
CN202011640125.2A CN112863175B (zh) | 2020-12-31 | 2020-12-31 | 汽车道路监测数据处理方法、装置、设备及存储介质 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022141865A1 true WO2022141865A1 (zh) | 2022-07-07 |
Family
ID=76000752
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/083954 WO2022141865A1 (zh) | 2020-12-31 | 2021-03-30 | 汽车道路监测数据处理方法、装置、设备及存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN112863175B (zh) |
WO (1) | WO2022141865A1 (zh) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109086619A (zh) * | 2018-07-05 | 2018-12-25 | 广东天泽汇通科技有限公司 | 基于区块链的停车信用调整方法及其装置、电子设备 |
US20190026657A1 (en) * | 2016-03-26 | 2019-01-24 | Alibaba Group Holding Limited | Distributed Cluster Training Method and Apparatus |
CN109711737A (zh) * | 2018-12-28 | 2019-05-03 | 英华达(南京)科技有限公司 | 一种威胁车辆检测方法、装置及*** |
CN110706371A (zh) * | 2019-10-17 | 2020-01-17 | 腾讯科技(深圳)有限公司 | 基于区块链的行车安全管理方法、***及存储介质 |
Family Cites Families (39)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2975087A1 (en) * | 2015-01-28 | 2016-08-04 | Allstate Insurance Company | Road segment safety rating |
US10249194B2 (en) * | 2016-08-30 | 2019-04-02 | International Business Machines Corporation | Modifying behavior of autonomous vehicle based on advanced predicted behavior analysis of nearby drivers |
US10490066B2 (en) * | 2016-12-29 | 2019-11-26 | X Development Llc | Dynamic traffic control |
CN107292528A (zh) * | 2017-06-30 | 2017-10-24 | 阿里巴巴集团控股有限公司 | 车险风险预测方法、装置及服务器 |
CN109754105B (zh) * | 2017-11-07 | 2024-01-05 | 华为技术有限公司 | 一种预测方法及终端、服务器 |
EP3503012A1 (en) * | 2017-12-20 | 2019-06-26 | Accenture Global Solutions Limited | Analytics engine for multiple blockchain nodes |
CN108063826B (zh) * | 2017-12-27 | 2020-08-14 | 上海唯链信息科技有限公司 | 一种基于区块链技术的车联网的共享与追溯*** |
CN111861433B (zh) * | 2018-03-30 | 2024-04-02 | 创新先进技术有限公司 | 基于区块链的业务执行方法及装置、电子设备 |
US10769869B2 (en) * | 2018-06-27 | 2020-09-08 | International Business Machines Corporation | Self-driving vehicle integrity management on a blockchain |
WO2020065708A1 (ja) * | 2018-09-25 | 2020-04-02 | 株式会社ウフル | コンピュータシステム、危険運転車両通知方法及びプログラム |
CN109543627B (zh) * | 2018-11-27 | 2023-08-01 | 西安电子科技大学 | 一种判断驾驶行为类别的方法、装置、及计算机设备 |
CN111833480B (zh) * | 2019-04-12 | 2022-04-15 | 比亚迪股份有限公司 | 驾驶行为检测方法、装置和车辆 |
US10839682B1 (en) * | 2019-04-26 | 2020-11-17 | Blackberry Limited | Method and system for traffic behavior detection and warnings |
US11823568B2 (en) * | 2019-05-02 | 2023-11-21 | Toyota Motor North America, Inc. | Dynamic speed limit for vehicles and autonomous vehicles |
US11694110B2 (en) * | 2019-06-12 | 2023-07-04 | International Business Machines Corporation | Aggregated machine learning verification for database |
CN110519286B (zh) * | 2019-09-01 | 2021-12-24 | 江西理工大学 | 一种基于联盟区块链的智能交通数据安全访问方法 |
CN110648244A (zh) * | 2019-09-05 | 2020-01-03 | 广州亚美信息科技有限公司 | 基于区块链的车险方案生成方法、装置和行车数据处理*** |
CN110659678B (zh) * | 2019-09-09 | 2023-11-17 | 腾讯科技(深圳)有限公司 | 一种用户行为分类方法、***及存储介质 |
US20200026289A1 (en) * | 2019-09-28 | 2020-01-23 | Ignacio J. Alvarez | Distributed traffic safety consensus |
CN110958301A (zh) * | 2019-11-04 | 2020-04-03 | 广州亚美信息科技有限公司 | 基于联盟链的车辆档案数据处理方法、装置和*** |
CN110990870A (zh) * | 2019-11-29 | 2020-04-10 | 上海能塔智能科技有限公司 | 运维、使用模型库的处理方法、装置、设备与介质 |
CN111212110B (zh) * | 2019-12-13 | 2022-06-03 | 清华大学深圳国际研究生院 | 一种基于区块链的联邦学习***及方法 |
CN111211893B (zh) * | 2020-01-22 | 2022-03-04 | 西安电子科技大学 | 一种应用区块链的自动驾驶汽车模型共享方法 |
CN111339554B (zh) * | 2020-02-17 | 2022-04-01 | 电子科技大学 | 基于移动边缘计算的用户数据隐私保护方法 |
CN111368326B (zh) * | 2020-02-25 | 2023-01-10 | 百度在线网络技术(北京)有限公司 | 车辆数据处理方法、装置、电子设备和存储介质 |
CN111368319B (zh) * | 2020-03-04 | 2022-04-19 | 西安电子科技大学 | 一种联邦学习环境下基于区块链的数据安全访问方法 |
CN111476139B (zh) * | 2020-04-01 | 2023-05-02 | 同济大学 | 基于联邦迁移学习的驾驶员行为云边协同学习*** |
CN111159305B (zh) * | 2020-04-02 | 2020-08-04 | 支付宝(杭州)信息技术有限公司 | 基于区块链的车辆数据处理方法、装置及*** |
CN111565174A (zh) * | 2020-04-20 | 2020-08-21 | 中移雄安信息通信科技有限公司 | 车辆质量信息确定模型训练方法及车辆质量信息确定方法 |
CN111753885B (zh) * | 2020-06-09 | 2023-09-01 | 华侨大学 | 一种基于深度学习的隐私增强数据处理方法和*** |
CN111539731A (zh) * | 2020-06-19 | 2020-08-14 | 支付宝(杭州)信息技术有限公司 | 基于区块链的联邦学习方法及装置和电子设备 |
CN111756829A (zh) * | 2020-06-19 | 2020-10-09 | 深圳市蔚链科技有限公司 | 账本数据同步方法、装置、设备和存储介质 |
CN111723946A (zh) * | 2020-06-19 | 2020-09-29 | 深圳前海微众银行股份有限公司 | 一种应用于区块链的联邦学习方法及装置 |
CN111966698B (zh) * | 2020-07-03 | 2023-06-13 | 华南师范大学 | 一种基于区块链的可信联邦学习方法、***、装置及介质 |
CN111856934B (zh) * | 2020-07-16 | 2022-11-15 | 南京大量数控科技有限公司 | 一种同构智能车间之间的联邦学习数据处理算法 |
CN111832503B (zh) * | 2020-07-20 | 2023-12-19 | 中国人民武装警察部队工程大学 | 一种基于联邦学习的分心驾驶识别方法 |
CN112132277A (zh) * | 2020-09-21 | 2020-12-25 | 平安科技(深圳)有限公司 | 联邦学习模型训练方法、装置、终端设备及存储介质 |
AU2020102543A4 (en) * | 2020-09-30 | 2020-11-19 | Aggarwal, Vibha DR | Iot based trust modeling approach for cooperative autonomous driving and intelligent transport systems (its) |
CN112132293A (zh) * | 2020-09-30 | 2020-12-25 | 腾讯科技(深圳)有限公司 | 纵向联邦学习中的计算方法、装置、设备及介质 |
-
2020
- 2020-12-31 CN CN202011640125.2A patent/CN112863175B/zh active Active
-
2021
- 2021-03-30 WO PCT/CN2021/083954 patent/WO2022141865A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190026657A1 (en) * | 2016-03-26 | 2019-01-24 | Alibaba Group Holding Limited | Distributed Cluster Training Method and Apparatus |
CN109086619A (zh) * | 2018-07-05 | 2018-12-25 | 广东天泽汇通科技有限公司 | 基于区块链的停车信用调整方法及其装置、电子设备 |
CN109711737A (zh) * | 2018-12-28 | 2019-05-03 | 英华达(南京)科技有限公司 | 一种威胁车辆检测方法、装置及*** |
CN110706371A (zh) * | 2019-10-17 | 2020-01-17 | 腾讯科技(深圳)有限公司 | 基于区块链的行车安全管理方法、***及存储介质 |
Non-Patent Citations (1)
Title |
---|
LU YUNLONG; HUANG XIAOHONG; ZHANG KE; MAHARJAN SABITA; ZHANG YAN: "Blockchain Empowered Asynchronous Federated Learning for Secure Data Sharing in Internet of Vehicles", IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, vol. 69, no. 4, 13 February 2020 (2020-02-13), USA, pages 4298 - 4311, XP011783958, ISSN: 0018-9545, DOI: 10.1109/TVT.2020.2973651 * |
Also Published As
Publication number | Publication date |
---|---|
CN112863175A (zh) | 2021-05-28 |
CN112863175B (zh) | 2022-11-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111506980B (zh) | 对虚拟驾驶环境生成交通场景的方法及装置 | |
US20220332335A1 (en) | Vehicle-data analytics | |
Hanselmann et al. | CANet: An unsupervised intrusion detection system for high dimensional CAN bus data | |
US11748474B2 (en) | Security system and methods for identification of in-vehicle attack originator | |
US12044539B2 (en) | Data processing system communicating with a map data processing system to determine or alter a navigation path based on one or more road segments | |
CN111095955B (zh) | 用于联网车辆网络安全的***和方法 | |
Boddupalli et al. | Resilient cooperative adaptive cruise control for autonomous vehicles using machine learning | |
CN111508101A (zh) | 通过检测驾驶场景来评估驾驶员的驾驶习惯的方法和装置 | |
US20180204119A1 (en) | Method and Apparatus for Driver Identification Leveraging Telematics Data | |
EP3895950B1 (en) | Methods and systems for automated driving system monitoring and management | |
US12001553B2 (en) | Detecting vehicle malfunctions and cyber attacks using machine learning | |
CN107045794B (zh) | 路况处理方法及装置 | |
WO2022078353A1 (zh) | 车辆行使状态判断方法、装置、计算机设备及存储介质 | |
US11572077B2 (en) | Method for distributed data analysis | |
WO2022222647A1 (zh) | 车辆意图预测方法、装置、设备及存储介质 | |
KR102196027B1 (ko) | Lstm 기반 조향 행동 모니터링 장치 및 그 방법 | |
WO2022141865A1 (zh) | 汽车道路监测数据处理方法、装置、设备及存储介质 | |
US11688207B2 (en) | System and method for contextually monitoring vehicle state | |
CN116583891A (zh) | 用于车辆核查和验证的危急场景识别 | |
CN112277956B (zh) | 一种用方向盘检测酒驾行为的方法和装置 | |
CN108900634A (zh) | 一种基于v2x的车辆远程控制方法及*** | |
CN115640828A (zh) | 一种基于对抗生成网络的车载数字孪生欺骗检测方法 | |
CN116601619A (zh) | 安全控制器区域网络(can)收发器 | |
JP7279685B2 (ja) | 情報処理システム | |
US20240061420A1 (en) | Contract testing for autonomous vehicles |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21912635 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21912635 Country of ref document: EP Kind code of ref document: A1 |