CN117246345A - Method, device, equipment and medium for controlling generating type vehicle - Google Patents
Method, device, equipment and medium for controlling generating type vehicle Download PDFInfo
- Publication number
- CN117246345A CN117246345A CN202311462621.7A CN202311462621A CN117246345A CN 117246345 A CN117246345 A CN 117246345A CN 202311462621 A CN202311462621 A CN 202311462621A CN 117246345 A CN117246345 A CN 117246345A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- data
- model
- simulation
- vehicle control
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000004088 simulation Methods 0.000 claims abstract description 115
- 238000011217 control strategy Methods 0.000 claims abstract description 99
- 238000012549 training Methods 0.000 claims abstract description 59
- 230000008569 process Effects 0.000 claims abstract description 25
- 230000003993 interaction Effects 0.000 claims description 43
- 230000015654 memory Effects 0.000 claims description 26
- 238000005457 optimization Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- 230000007547 defect Effects 0.000 abstract description 4
- 238000013461 design Methods 0.000 abstract description 4
- 238000012795 verification Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/0098—Details of control systems ensuring comfort, safety or stability not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/08—Interaction between the driver and the control system
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0062—Adapting control system settings
- B60W2050/0075—Automatic parameter input, automatic initialising or calibrating means
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of vehicle control, and discloses a method, a device, equipment and a medium for generating vehicle control, wherein the method comprises the following steps: setting up an online simulation training environment, and performing simulated driving based on the online simulation training environment to generate simulated driving data; constructing an initial generation type model, training the initial generation type model according to driving simulation data, and taking the initial generation type model for generating an optimal vehicle control strategy as a final generation type instruction control model; and acquiring actual driving data in the process of running the vehicle by a user, inputting the actual driving data into a generated instruction control model deployed on the vehicle to generate an optimal vehicle control strategy, and controlling the vehicle. The invention can solve the inflexibility of the rule-based vehicle control logic, prevent the occurrence of design defects caused by insufficient knowledge of designers or the situation that some special cases in reality cannot be processed based on limited rules, and improve the vehicle automation control capability.
Description
Technical Field
The invention relates to the technical field of vehicle control, in particular to a method, a device, equipment and a medium for generating vehicle control.
Background
The traditional vehicle electronic control method realizes the control of the vehicle through specific interaction rules based on pre-designed signals or services. Along with the improvement of the electric control capability of the vehicle, the increase of the control complexity and the diversification of the human-computer interaction requirements. The traditional vehicle electronic control method is limited in that the cognition of a designer can only meet the most basic vehicle control requirement, and a user is required to participate in a large amount in the control process to consume the energy of the user, so that the fatigue of the user in the use of the vehicle is caused to influence the user experience.
In recent years, there has been proposed a vehicle control method by means of scene, which solves the above problems, edits a custom scene by opening a signal or service that can be freely combined by a user, and performs a series of actions when a scene trigger condition is satisfied, thereby achieving an automatic vehicle control purpose. The method can bring the creativity of users into play to a certain extent and expand the types of vehicle control. However, the vehicle control method for scene is still a regularized control method designed in advance, and a user needs to master a certain related knowledge of the vehicle to define the scene.
Disclosure of Invention
In view of the above, the present invention provides a method, apparatus, device and medium for generating vehicle control, so as to solve the problem of inflexible vehicle control.
In a first aspect, the present invention provides a method for generating vehicle control, the method comprising:
setting up an online simulation training environment, and performing simulated driving based on the online simulation training environment to generate simulated driving data;
constructing an initial generation type model, training the initial generation type model according to driving simulation data, and taking the initial generation type model for generating an optimal vehicle control strategy as a final generation type instruction control model;
and acquiring actual driving data in the process of running the vehicle by a user, inputting the actual driving data into a generated instruction control model deployed on the vehicle to generate an optimal vehicle control strategy, and controlling the vehicle.
According to the generated vehicle control method provided by the embodiment of the invention, the on-line simulation training environment is built for simulated driving, the generated command control model is trained according to the generated simulation driving data, and in the process that a user runs the vehicle, the generated command control model controls the vehicle according to the acquired actual driving data and the optimal vehicle control strategy. According to the invention, the optimal vehicle control strategy is automatically generated according to the current vehicle driving scene by training the generating model, so that the inflexibility of the rule-based vehicle control logic can be solved, the situation that design defects caused by insufficient knowledge of designers or special cases in reality cannot be processed based on limited rules can be prevented, and the vehicle automatic control capability is improved.
In an alternative embodiment, an online simulation training environment comprises: a vehicle simulation model, a traffic environment model and a human-vehicle interaction model; corresponding simulated driving data, comprising: vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data; corresponding actual driving data, including: vehicle operation data, traffic environment data and human-vehicle interaction data.
According to the invention, the vehicle operation data, the traffic environment data and the human-vehicle interaction data related to the vehicle operation scene are simulated or collected, so that the related parameters which can influence the vehicle control strategy can be obtained to the greatest extent, and the generated vehicle control strategy is more accurate and reliable.
In an alternative embodiment, the training process for the initially generated model according to the driving simulation data includes: constructing a preset scene, and filtering and classifying vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data of a vehicle control strategy based on the preset scene of the vehicle control strategy; inputting a vehicle control strategy preset scene and vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data of a vehicle control strategy corresponding to the vehicle control strategy preset scene into a vehicle control strategy initial generation model for training, and generating a vehicle control strategy; and comparing the vehicle control strategy with the vehicle operation simulation data of the vehicle control strategy, and iteratively adjusting the vehicle control strategy initial generation model according to the comparison result until the vehicle control strategy comparison result reaches a preset condition.
According to the invention, the traditional rule-based vehicle control strategy is replaced by the creatively generated model, so that the vehicle automatically generates the vehicle control strategy according to the current running state and the traffic environment, the constraint of rules on vehicle control can be reduced, and the automatic control capability of the vehicle is expanded.
In an alternative embodiment, after training the generated instruction control model, the method further comprises: deploying the trained generated instruction control model in a preset real vehicle environment; performing simulated driving based on a preset real vehicle environment, and controlling the real vehicle through an optimal vehicle control strategy generated by a generated instruction control model; and collecting real vehicle operation data, traffic environment data and human-vehicle interaction data in the real vehicle operation process, uploading the real vehicle operation data, the traffic environment data and the human-vehicle interaction data to an on-line simulation training environment, and optimizing the generated instruction control model.
According to the method, before the trained generated instruction control model is applied to the user vehicle, the training generated instruction control model is deployed in the preset real vehicle environment to carry out real vehicle verification, and errors generated according to the vehicle control strategy generated by the model in the real vehicle environment are corrected in time, so that the model is optimized, the accuracy of the vehicle control strategy generated by the model can be further improved, accidents are reduced, and therefore safety of the user is guaranteed.
In an alternative embodiment, after acquiring actual driving data during the running of the vehicle by the user, the method further comprises: optimizing a generated instruction control model according to vehicle operation data, traffic environment data and human-vehicle interaction data; uploading the model optimization result to an on-line simulation training environment and a preset real vehicle environment, and further optimizing and verifying the generated instruction control model.
According to the invention, the model optimization is carried out by acquiring the actual driving data of the user in the actual vehicle using process, so that more real sample data can be acquired, and the accuracy of the vehicle control strategy generated by the generated command control model is further improved.
In an alternative embodiment, after inputting each data to the generated command control model deployed on the vehicle to generate the optimal vehicle control strategy, the method further comprises: judging an optimal vehicle control strategy according to a preset safety rule; if the optimal vehicle control strategy does not accord with the preset safety rule, correcting the optimal vehicle control strategy; and if the optimal vehicle control strategy accords with the preset safety rule, controlling the vehicle according to the optimal vehicle control strategy.
In an alternative embodiment, the preset security rules include: at least one of a maximum driving distance, a minimum following distance, a minimum lane change space or preset traffic regulation requirement conditions.
According to the invention, the optimal vehicle control strategy generated by the generated command control model is subjected to safety inspection and reasonable adjustment according to the safety rule of the vehicle, so that possible safety problems of the generated command control model can be considered, unsafe control strategy generation can be prevented based on safety consideration, and the personal safety of a user is further ensured.
In a second aspect, the present invention provides a generator-type vehicle control apparatus, the apparatus comprising:
the simulation environment building module is used for building an online simulation training environment and performing simulated driving based on the online simulation training environment to generate simulation driving data;
the system comprises a generating model training module, a driving simulation module and a vehicle control module, wherein the generating model training module is used for constructing an initial generating model, training the initial generating model according to driving simulation data and taking the initial generating model for generating an optimal vehicle control strategy as a final generating instruction control model;
and the vehicle control module is used for acquiring actual driving data in the process of running the vehicle by a user, inputting the actual driving data into a generated instruction control model deployed on the vehicle to generate an optimal vehicle control strategy, and controlling the vehicle.
According to the generated vehicle control device provided by the embodiment of the invention, the on-line simulation training environment is built for simulated driving, the generated instruction control model is trained according to the generated simulation driving data, and in the process that a user runs the vehicle, the generated instruction control model controls the vehicle according to the acquired actual driving data and the optimal vehicle control strategy. According to the invention, the optimal vehicle control strategy is automatically generated according to the current vehicle driving scene by training the generating model, so that the inflexibility of the rule-based vehicle control logic can be solved, the situation that design defects caused by insufficient knowledge of designers or special cases in reality cannot be processed based on limited rules can be prevented, and the vehicle automatic control capability is improved.
In a third aspect, the present invention provides a computer device comprising: the system comprises a memory and a processor, wherein the memory and the processor are in communication connection, the memory stores computer instructions, and the processor executes the computer instructions, so that the generating vehicle control method of the first aspect or any implementation mode corresponding to the first aspect is executed.
In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method of generating a vehicle control of the first aspect or any of its corresponding embodiments.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method of generating vehicle control according to an embodiment of the invention;
FIG. 2 is an environment setup schematic of a generative vehicle control method according to an embodiment of the invention;
fig. 3 is a block diagram of a construction of a generator-type vehicle control apparatus according to an embodiment of the invention;
fig. 4 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment of the invention is suitable for the scene that the vehicle automatically generates the vehicle control strategy according to the simple instruction of the user or the surrounding environment. The embodiment of the invention provides a generating type vehicle control method, which automatically generates an optimal vehicle control strategy through a generating type command control model so as to achieve the effect of automatic and flexible control of a vehicle. It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
In this embodiment, a method for controlling a generating vehicle is provided, which may be used in the computer described above, and fig. 1 is a flowchart of a method for controlling a generating vehicle according to an embodiment of the present invention, as shown in fig. 1, where the flowchart includes the following steps:
step S101, an online simulation training environment is built, and simulated driving is performed based on the online simulation training environment to generate simulated driving data.
Specifically, in the embodiment of the invention, an online simulation training environment including a vehicle simulation model, a traffic environment model and a human-vehicle interaction model is built at a cloud end, then simulated driving is performed based on the cloud end environment, and simulated driving data is generated in the simulated driving process. The vehicle simulation model simulates and generates vehicle operation simulation data, such as vehicle speed, steering, vehicle window state, vehicle lamp state and the like; the traffic environment simulation data simulate to generate traffic environment data such as pedestrians, other vehicles, traffic lights and the like; the human-vehicle interaction model simulates and generates human-vehicle interaction data, such as the language, behavior, etc. of the driver, which is given by way of example only and not limitation.
Step S102, an initial generation type model is built, training is carried out on the initial generation type model according to driving simulation data, and the initial generation type model for generating the optimal vehicle control strategy is used as a final generation type command control model.
Specifically, in the embodiment of the present invention, the employed generative model is a technology which has been attracting attention in the field of artificial intelligence in recent years, and can be applied not only to the fields of speech synthesis, image generation, and the like, but also to the fields of natural language processing, machine translation, and the like. The generative model is a method for generating data by using a probability model. It may be seen as a process of extracting samples from a priori distribution, so that new data may be generated, rather than just classifying existing data. According to the embodiment of the invention, the vehicle operation simulation data, the traffic environment simulation data and the human-vehicle interaction simulation data obtained by simulating the driving are input into the initial generation type model for training, and finally the initial generation type model for generating the optimal vehicle control strategy is used as the final generation type instruction control model.
In an optional implementation manner, the process of inputting the vehicle operation simulation data, the traffic environment simulation data and the human-vehicle interaction simulation data obtained by the simulation driving into the initial generation model for training according to the embodiment of the invention comprises the following steps:
(1) And constructing a preset scene, and filtering and classifying vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data of the vehicle control strategy based on the preset scene of the vehicle control strategy. In the embodiment of the invention, firstly, the acquired vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data are preprocessed, and the data are divided into a training data set and a test data set which meet model training. Meanwhile, a scene is preset, and a series of input data sets aiming at specific scenes are constructed by filtering and classifying the scenes from vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data.
(2) And inputting the vehicle control strategy preset scene and vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data of the vehicle control strategy corresponding to the vehicle control strategy preset scene into a vehicle control strategy initial generation model for training, so as to generate a vehicle control strategy. In the embodiment of the invention, the input data set is imported into the initial generation type model for learning and training, the initial generation type model can learn the relation between a scene and vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data, and then a vehicle control strategy is generated.
(3) And comparing the vehicle control strategy with the vehicle operation simulation data of the vehicle control strategy, and iteratively adjusting the vehicle control strategy initial generation model according to the comparison result until the vehicle control strategy comparison result reaches a preset condition. The embodiment of the invention compares the generated vehicle control strategy with the vehicle operation simulation data, judges whether the initial generation type model accords with the preset condition, and optimizes the initial generation type model by iteratively adjusting the data set or adjusting the parameters of the initial generation type model if the initial generation type model does not accord with the preset condition until the initial generation type model accords with the preset condition, and takes the initial generation type model for generating the optimal vehicle control strategy as the final generation type command control model.
In an alternative embodiment, for example, to obtain a generated model of "sudden situation in front of vehicle, the rear vehicle needs to be pre-warned", a simulation environment needs to be built in advance, the vehicle model (including an electronic and electric model or signal) and complete vehicle environment data are input into the simulation environment, and the vehicle is operated in the simulation environment to simulate various situations of "sudden situation in front of vehicle", and the operations are performed for various situations, such as: deceleration, continuous point braking, double flashing, defensive driving and whistling. And recording simulation operation data in the simulation process, and training a generated model.
Step S103, acquiring actual driving data in the process of running the vehicle by a user, inputting the actual driving data into a generated command control model deployed on the vehicle to generate an optimal vehicle control strategy, and controlling the vehicle.
Specifically, in the embodiment of the present invention, as shown in fig. 2, after the generated instruction control model is trained, the trained generated instruction control model is deployed in a preset real vehicle environment, where the real vehicle environment includes a central controller of a vehicle, each wheel, a vehicle lamp, a wiper, steering, accelerating, braking, etc., where the generated instruction control model is disposed in the central controller, and the above is only for example and not limited thereto. And performing simulated driving based on a preset real vehicle environment, controlling the real vehicle through an optimal vehicle control strategy generated by the generated command control model, verifying the generated command control model, pushing the model to a vehicle of a user for application if the model is verified, and adjusting simulation or retraining the real vehicle environment if the model is verified. In addition, real vehicle operation data, traffic environment data and human-vehicle interaction data are collected in the real vehicle operation process, and the real vehicle operation data, the traffic environment data and the human-vehicle interaction data are uploaded to an online simulation training environment to optimize the generated instruction control model.
In an optional implementation manner, the embodiment of the invention deploys the optimized generated instruction control model to the user vehicle for practical application, and the vehicle acquires actual driving data including vehicle running data, traffic environment data and human-vehicle interaction data in the process of running the vehicle by the user, wherein the human-vehicle interaction data includes simple voice input or simple actions of the user. The vehicle inputs the acquired vehicle operation data, traffic environment data and human-vehicle interaction data into an on-premise generated command control model to generate an optimal vehicle control strategy.
In an alternative embodiment, after generating the optimal vehicle control strategy, the vehicle further determines the optimal vehicle control strategy according to a preset safety rule: if the optimal vehicle control strategy does not accord with the preset safety rule, correcting the optimal vehicle control strategy; and if the optimal vehicle control strategy accords with the preset safety rule, controlling the vehicle according to the optimal vehicle control strategy. According to the embodiment of the invention, the control strategy output by the generated instruction model is subjected to safety verification based on safety consideration, so that unsafe control strategy generation can be prevented. Wherein the preset security rules include at least one of: the maximum driving speed is used for avoiding exceeding the safe driving speed; the minimum following distance is used for avoiding exceeding the minimum following distance; the minimum lane change space is used for avoiding the lane change exceeding the safety range; the preset traffic regulation requirement conditions, such as the regulation requirement of prohibiting parking of road sections, are only used as distances, and are not limited to the distances.
In an alternative embodiment, after the actual driving data is obtained in the process of running the vehicle by the user, the generated instruction control model is optimized according to the vehicle running data, the traffic environment data and the human-vehicle interaction data, and the model optimization result is uploaded to an on-line simulation training environment and a preset real vehicle environment to further optimize and verify the generated instruction control model.
In an alternative embodiment, as shown in fig. 2, the cloud online simulation training environment further includes a data system and a management system, which are used for storing and managing the real vehicle data and the acquired vehicle data. The data management system provides a storage space for storing data, a data read-write system for reading or writing the data, and a data recording system for recording basic attributes of the data, but is not limited to the above.
According to the generated vehicle control method provided by the embodiment of the invention, the on-line simulation training environment is built for simulated driving, the generated command control model is trained according to the generated simulation driving data, and in the process that a user runs the vehicle, the generated command control model controls the vehicle according to the acquired actual driving data and the optimal vehicle control strategy. According to the invention, the optimal vehicle control strategy is automatically generated according to the current vehicle driving scene by training the generating model, so that the inflexibility of the rule-based vehicle control logic can be solved, the situation that design defects caused by insufficient knowledge of designers or special cases in reality cannot be processed based on limited rules can be prevented, and the vehicle automatic control capability is improved.
The present embodiment provides a generation type vehicle control apparatus, as shown in fig. 3, including:
the simulation environment construction module 301 is configured to construct an online simulation training environment, and perform simulated driving based on the online simulation training environment to generate simulated driving data;
the generative model training module 302 is configured to construct an initial generative model, train the initial generative model according to driving simulation data, and take the initial generative model for generating the optimal vehicle control strategy as a final generative instruction control model;
the vehicle control module 303 is configured to obtain actual driving data during a user running the vehicle, and input the actual driving data to a generated command control model deployed on the vehicle to generate an optimal vehicle control strategy, so as to control the vehicle.
In some alternative embodiments, the simulation environment setup module 301 includes: a vehicle simulation model, a traffic environment model and a human-vehicle interaction model; corresponding simulated driving data, comprising: vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data; corresponding actual driving data, including: vehicle operation data, traffic environment data and human-vehicle interaction data.
In some alternative embodiments, generative model training module 302 comprises a model training unit, the process of which comprises: constructing a preset scene, and filtering and classifying vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data of a vehicle control strategy based on the preset scene of the vehicle control strategy; inputting a vehicle control strategy preset scene and vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data of a vehicle control strategy corresponding to the vehicle control strategy preset scene into a vehicle control strategy initial generation model for training, and generating a vehicle control strategy; and comparing the vehicle control strategy with the vehicle operation simulation data of the vehicle control strategy, and iteratively adjusting the vehicle control strategy initial generation model according to the comparison result until the vehicle control strategy comparison result reaches a preset condition.
In some optional embodiments, the system further includes a real vehicle verification module, configured to deploy the trained generated instruction control model in a preset real vehicle environment; performing simulated driving based on a preset real vehicle environment, and controlling the real vehicle through an optimal vehicle control strategy generated by a generated instruction control model; and collecting real vehicle operation data, traffic environment data and human-vehicle interaction data in the real vehicle operation process, uploading the real vehicle operation data, the traffic environment data and the human-vehicle interaction data to an on-line simulation training environment, and optimizing the generated instruction control model.
In some alternative embodiments, the vehicle control module 303 includes a model optimization unit for optimizing the generated command control model based on vehicle operation data, traffic environment data, and human-vehicle interaction data; uploading the model optimization result to an on-line simulation training environment and a preset real vehicle environment, and further optimizing and verifying the generated instruction control model.
In some optional embodiments, the system further comprises a safety verification module, configured to determine an optimal vehicle control policy according to a preset safety rule; if the optimal vehicle control strategy does not accord with the preset safety rule, correcting the optimal vehicle control strategy; and if the optimal vehicle control strategy accords with the preset safety rule, controlling the vehicle according to the optimal vehicle control strategy.
In some alternative embodiments, the preset security rules of the security verification module include: at least one of a maximum driving distance, a minimum following distance, a minimum lane change space or preset traffic regulation requirement conditions.
Further functional descriptions of the above respective modules and units are the same as those of the above corresponding embodiments, and are not repeated here.
The resulting vehicle control apparatus in this embodiment is presented in the form of functional units, herein referred to as FPGA (Field Programmable Gate Array ) circuits, processors and memory executing one or more software or fixed programs, and/or other devices that can provide the functionality described above.
The embodiment of the invention also provides computer equipment, which is provided with the generating type vehicle control device shown in the figure 3.
Referring to fig. 4, fig. 4 is a schematic structural diagram of a computer device according to an alternative embodiment of the present invention, as shown in fig. 4, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components are communicatively coupled to each other using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions executing within the computer device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In some alternative embodiments, multiple processors and/or multiple buses may be used, if desired, along with multiple memories and multiple memories. Also, multiple computer devices may be connected, each providing a portion of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). One processor 10 is illustrated in fig. 4.
The processor 10 may be a central processor, a network processor, or a combination thereof. The processor 10 may further include a hardware chip, among others. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general-purpose array logic, or any combination thereof.
Wherein the memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform a method for implementing the embodiments described above.
The memory 20 may include a storage program area that may store an operating system, at least one application program required for functions, and a storage data area; the storage data area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, memory 20 may optionally include memory located remotely from processor 10, which may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid state disk; the memory 20 may also comprise a combination of the above types of memories.
The computer device also includes a communication interface 30 for the computer device to communicate with other devices or communication networks.
The embodiments of the present invention also provide a computer readable storage medium, and the method according to the embodiments of the present invention described above may be implemented in hardware, firmware, or as a computer code which may be recorded on a storage medium, or as original stored in a remote storage medium or a non-transitory machine readable storage medium downloaded through a network and to be stored in a local storage medium, so that the method described herein may be stored on such software process on a storage medium using a general purpose computer, a special purpose processor, or programmable or special purpose hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, a solid state disk or the like; further, the storage medium may also comprise a combination of memories of the kind described above. It will be appreciated that a computer, processor, microprocessor controller or programmable hardware includes a storage element that can store or receive software or computer code that, when accessed and executed by the computer, processor or hardware, implements the methods illustrated by the above embodiments.
Although embodiments of the present invention have been described in connection with the accompanying drawings, various modifications and variations may be made by those skilled in the art without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope of the invention as defined by the appended claims.
Claims (10)
1. A generation type vehicle control method, characterized by comprising:
building an online simulation training environment, and performing simulated driving based on the online simulation training environment to generate simulated driving data;
constructing an initial generation type model, training the initial generation type model according to the driving simulation data, and taking the initial generation type model for generating the optimal vehicle control strategy as a final generation type instruction control model;
and acquiring actual driving data in the process of running the vehicle by a user, inputting the actual driving data into the generated command control model deployed on the vehicle to generate an optimal vehicle control strategy, and controlling the vehicle.
2. The method of claim 1, wherein the online simulation training environment comprises: a vehicle simulation model, a traffic environment model and a human-vehicle interaction model;
corresponding simulated driving data, comprising: vehicle operation simulation data, traffic environment simulation data and human-vehicle interaction simulation data;
corresponding actual driving data, including: vehicle operation data, traffic environment data and human-vehicle interaction data.
3. The method of claim 2, wherein the training the initially generated model from the driving simulation data comprises:
constructing a preset scene, and filtering and classifying the vehicle operation simulation data, the traffic environment simulation data and the human-vehicle interaction simulation data based on the preset scene;
inputting the preset scene, the vehicle operation simulation data, the traffic environment simulation data and the human-vehicle interaction simulation data corresponding to the preset scene into the initial generation model for training, and generating a vehicle control strategy;
and comparing the vehicle control strategy with the vehicle operation simulation data, and iteratively adjusting the initial generation model according to a comparison result until the comparison result reaches a preset condition.
4. The method of claim 3, further comprising, after training the generated command control model:
deploying the trained generated instruction control model in a preset real vehicle environment;
performing simulated driving based on the preset real vehicle environment, and controlling the real vehicle through an optimal vehicle control strategy generated by the generated instruction control model;
and acquiring real vehicle operation data, traffic environment data and human-vehicle interaction data in the real vehicle operation process, uploading the real vehicle operation data, the traffic environment data and the human-vehicle interaction data to the online simulation training environment, and optimizing the generated instruction control model.
5. The method of claim 4, further comprising, after acquiring actual driving data during operation of the vehicle by the user:
optimizing the generated instruction control model according to the vehicle operation data, the traffic environment data and the human-vehicle interaction data;
and uploading a model optimization result to the online simulation training environment and the preset real vehicle environment, and further optimizing and verifying the generated instruction control model.
6. The method of claim 1, wherein after inputting each data to the generated command control model deployed on the vehicle to generate an optimal vehicle control strategy, further comprising: judging the optimal vehicle control strategy according to a preset safety rule;
if the optimal vehicle control strategy does not accord with the preset safety rule, correcting the optimal vehicle control strategy;
and if the optimal vehicle control strategy accords with the preset safety rule, controlling the vehicle according to the optimal vehicle control strategy.
7. The method of claim 6, wherein the preset security rules comprise: at least one of a maximum driving distance, a minimum following distance, a minimum lane change space or preset traffic regulation requirement conditions.
8. A generation type vehicle control apparatus, characterized by comprising:
the simulation environment building module is used for building an online simulation training environment and performing simulated driving based on the online simulation training environment to generate simulation driving data;
the generating model training module is used for constructing an initial generating model, training the initial generating model according to the driving simulation data and taking the initial generating model for generating the optimal vehicle control strategy as a final generating instruction control model;
and the vehicle control module is used for acquiring actual driving data in the process of running the vehicle by a user, inputting the actual driving data into the generated command control model deployed on the vehicle to generate an optimal vehicle control strategy, and controlling the vehicle.
9. A computer device, comprising:
a memory and a processor communicatively coupled to each other, the memory having stored therein computer instructions that, upon execution, perform the method of generating vehicle control of any of claims 1-7.
10. A computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the generated vehicle control method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311462621.7A CN117246345A (en) | 2023-11-06 | 2023-11-06 | Method, device, equipment and medium for controlling generating type vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311462621.7A CN117246345A (en) | 2023-11-06 | 2023-11-06 | Method, device, equipment and medium for controlling generating type vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117246345A true CN117246345A (en) | 2023-12-19 |
Family
ID=89129750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311462621.7A Pending CN117246345A (en) | 2023-11-06 | 2023-11-06 | Method, device, equipment and medium for controlling generating type vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117246345A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107506830A (en) * | 2017-06-20 | 2017-12-22 | 同济大学 | Towards the artificial intelligence training platform of intelligent automobile programmed decision-making module |
| DE102017007136A1 (en) * | 2017-07-27 | 2019-01-31 | Opel Automobile Gmbh | Method and device for training self-learning algorithms for an automated mobile vehicle |
| DE102018217004A1 (en) * | 2017-10-12 | 2019-04-18 | Honda Motor Co., Ltd. | Autonomous vehicle strategy generation |
| US20200033866A1 (en) * | 2018-07-24 | 2020-01-30 | Pony.ai, Inc. | Generative adversarial network enriched driving simulation |
| CN110991095A (en) * | 2020-03-05 | 2020-04-10 | 北京三快在线科技有限公司 | Training method and device for vehicle driving decision model |
| CN112906126A (en) * | 2021-01-15 | 2021-06-04 | 北京航空航天大学 | Vehicle hardware in-loop simulation training system and method based on deep reinforcement learning |
| CN114162146A (en) * | 2022-02-09 | 2022-03-11 | 苏州浪潮智能科技有限公司 | Driving strategy model training method and automatic driving control method |
| CN114667545A (en) * | 2019-10-31 | 2022-06-24 | 标致雪铁龙汽车股份有限公司 | Method for training at least one algorithm for a control unit of a motor vehicle, computer program product and motor vehicle |
| CN115257819A (en) * | 2022-09-02 | 2022-11-01 | 东南大学 | Decision-making method for safe driving of large-scale commercial vehicle in urban low-speed environment |
| CN115303297A (en) * | 2022-07-25 | 2022-11-08 | 武汉理工大学 | Method and device for controlling end-to-end automatic driving under urban market scene based on attention mechanism and graph model reinforcement learning |
| CN116620327A (en) * | 2023-06-20 | 2023-08-22 | 浙江大学 | Lane changing decision method for realizing automatic driving high-speed scene based on PPO and Lattice |
| CN116894395A (en) * | 2023-07-26 | 2023-10-17 | 重庆长安汽车股份有限公司 | Automatic driving test scene generation method, system and storage medium |
-
2023
- 2023-11-06 CN CN202311462621.7A patent/CN117246345A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107506830A (en) * | 2017-06-20 | 2017-12-22 | 同济大学 | Towards the artificial intelligence training platform of intelligent automobile programmed decision-making module |
| DE102017007136A1 (en) * | 2017-07-27 | 2019-01-31 | Opel Automobile Gmbh | Method and device for training self-learning algorithms for an automated mobile vehicle |
| DE102018217004A1 (en) * | 2017-10-12 | 2019-04-18 | Honda Motor Co., Ltd. | Autonomous vehicle strategy generation |
| US20200033866A1 (en) * | 2018-07-24 | 2020-01-30 | Pony.ai, Inc. | Generative adversarial network enriched driving simulation |
| CN114667545A (en) * | 2019-10-31 | 2022-06-24 | 标致雪铁龙汽车股份有限公司 | Method for training at least one algorithm for a control unit of a motor vehicle, computer program product and motor vehicle |
| CN110991095A (en) * | 2020-03-05 | 2020-04-10 | 北京三快在线科技有限公司 | Training method and device for vehicle driving decision model |
| CN112906126A (en) * | 2021-01-15 | 2021-06-04 | 北京航空航天大学 | Vehicle hardware in-loop simulation training system and method based on deep reinforcement learning |
| CN114162146A (en) * | 2022-02-09 | 2022-03-11 | 苏州浪潮智能科技有限公司 | Driving strategy model training method and automatic driving control method |
| CN115303297A (en) * | 2022-07-25 | 2022-11-08 | 武汉理工大学 | Method and device for controlling end-to-end automatic driving under urban market scene based on attention mechanism and graph model reinforcement learning |
| CN115257819A (en) * | 2022-09-02 | 2022-11-01 | 东南大学 | Decision-making method for safe driving of large-scale commercial vehicle in urban low-speed environment |
| CN116620327A (en) * | 2023-06-20 | 2023-08-22 | 浙江大学 | Lane changing decision method for realizing automatic driving high-speed scene based on PPO and Lattice |
| CN116894395A (en) * | 2023-07-26 | 2023-10-17 | 重庆长安汽车股份有限公司 | Automatic driving test scene generation method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109991987B (en) | Automatic driving decision-making method and device | |
| CN114117740A (en) | Simulation test scene generation method and device based on automatic driving | |
| CN114358128B (en) | Method for training end-to-end automatic driving strategy | |
| CN110245406B (en) | Driving simulation method, device and storage medium | |
| US20230415787A1 (en) | Method and apparatus for generating a simulation scene, electronic device and storage medium | |
| US20220258752A1 (en) | Autonomous cps self-evolution framework based on federated reinforcement learning for performance self-evolution of autonomous cps and performance self-evolution method for autonomous cps using the same | |
| CN113022581A (en) | Method and apparatus for determining a configuration for an autonomous vehicle | |
| CN112382165A (en) | Driving strategy generation method, device, medium, equipment and simulation system | |
| KR102206684B1 (en) | Learning method for analyzing driving situation and driving style, and an apparatus for performing the same | |
| CN116403174A (en) | End-to-end automatic driving method, system, simulation system and storage medium | |
| US8751094B2 (en) | Method for validation of a graphically based executable control specification using model extraction | |
| CN112418574A (en) | Artificial intelligence-based urban rail transit operation simulation system and method | |
| CN117246345A (en) | Method, device, equipment and medium for controlling generating type vehicle | |
| CN115053209A (en) | Method and system for generating engineering drawings in engineering system | |
| CN116461507A (en) | Vehicle driving decision method, device, equipment and storage medium | |
| US8751093B2 (en) | Hierarchical accumulated validation system and method | |
| CN115795900A (en) | Software in-loop simulation test method and device for automatic driving system and electronic equipment | |
| CN113450559B (en) | Method and device for auditing control scheme of traffic light | |
| CN114104005A (en) | Decision-making method, device and equipment of automatic driving equipment and readable storage medium | |
| DE102020204979A1 (en) | Method and device for traffic control | |
| CN114582124B (en) | Scene editing method, device, medium and electronic equipment | |
| CN112100836B (en) | Method and device for acquiring vehicle boundary parameters | |
| JP7505386B2 (en) | Remote support device, remote support system, remote support method and program | |
| CN114708744B (en) | Vehicle starting optimization control method and device based on fusion traffic information | |
| EP4198800A1 (en) | System and method for generating and simulating vehicle events and data |
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 |