Behavior decision model optimization, behavior decision method, apparatus, device, and medium

Abstract

The application discloses a behavior decision model optimization, a behavior decision method, a device, equipment and a medium. The optimization method of the vehicle behavior decision model comprises the following steps: obtaining a vehicle behavior decision model to be optimized and a standard verification sample set matched with the vehicle behavior decision model, wherein the standard verification sample set comprises a plurality of vehicle environment description characteristic values and a standard behavior decision result; performing data preprocessing on the standard verification sample set to obtain a normalized sample set; setting parameters based on a bat algorithm, and performing multi-round iteration processing on a bat colony by using the bat algorithm; when a bat iteration end condition is met, a global optimal bat is obtained; and the global optimal bat is used to update the vehicle behavior decision model to obtain an optimized vehicle behavior decision model. By using the above technical solution, the vehicle behavior decision model can be efficiently optimized, and the accuracy of the vehicle behavior decision model can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle automatic driving, and in particular to a behavior decision model optimization, a behavior decision method, device, equipment and medium. BACKGROUND

[0002] In the process of vehicle automatic driving, in order to ensure the fluency and safety of vehicle driving, the current behavior of automatic driving needs to be decided according to the current driving state of the vehicle and the surrounding environment state of the vehicle.

[0003] In the prior art, the classifier can be used for event decision in various scenarios, but the existing classifier cannot be directly used to obtain vehicle behavior decision according to vehicle related information, so it is important to optimize the existing classifier to obtain the optimal vehicle behavior decision. SUMMARY

[0004] The present application provides a behavior decision model optimization, a behavior decision method, device, equipment and medium, which can efficiently optimize the vehicle behavior decision model and improve the accuracy of the vehicle behavior decision model.

[0005] According to one aspect of the present application, a vehicle behavior decision model optimization method is provided, comprising:

[0006] obtaining a vehicle behavior decision model to be optimized and a standard verification sample set matched with the vehicle behavior decision model, the standard verification sample set including a plurality of vehicle environment description feature values and a standard behavior decision result;

[0007] data preprocessing is performed on the standard verification sample set to obtain a normalized sample set;

[0008] parameters are set based on the bat algorithm, and the bat algorithm is used for multi-round iteration processing of the bat colony;

[0009] In each iteration process, the standard verification sample set is input into each updated vehicle behavior decision model, and the fitness of each updated vehicle behavior decision model is used as an evaluation index, the optimal bat in the current round is identified among the bats, and the optimal bat in the current round is used for iteration optimization of the bat colony in the current round, the bats including a penalty factor and a Gaussian kernel function;

[0010] When the bat iteration end condition is met, the global optimal bat is obtained;

[0011] The global optimal bat is used to update the vehicle behavior decision model to obtain an optimized vehicle behavior decision model.

[0012] According to another aspect of the present application, a vehicle behavior decision method is provided, which is executed by a car machine system and comprises:

[0013] acquire a plurality of vehicle environment description characteristic values corresponding to at least one associated vehicle related to the vehicle behavior decision through information acquisition results of a plurality of sensors arranged in the vehicle;

[0014] The vehicle environment description characteristic values include current vehicle speeds of the vehicle and each associated vehicle, current vehicle accelerations of the vehicle and each associated vehicle, current vehicle left front coordinates of the vehicle and each associated vehicle, current vehicle front longitudinal coordinates of the vehicle and each associated vehicle, current vehicle widths of the vehicle and each associated vehicle, current vehicle lengths of the vehicle and each associated vehicle, a distance between the vehicle and at least one front vehicle, and a distance between the vehicle and at least one rear vehicle.

[0015] The plurality of vehicle environment description characteristic values are input into a plurality of optimized vehicle behavior decision models obtained by optimizing the method in any one of the embodiments one and two, and accuracy rates corresponding to each vehicle behavior decision output by the plurality of optimized vehicle behavior decision models are acquired.

[0016] The vehicle behavior decision with the highest accuracy rate is selected as a vehicle behavior decision result, and a vehicle control operation matched with the vehicle behavior decision result is performed.

[0017] According to another aspect of the present application, a device for optimizing a vehicle behavior decision model is provided, comprising:

[0018] a standard verification sample set acquisition module configured to acquire a vehicle behavior decision model to be optimized and a standard verification sample set matched with the vehicle behavior decision model, the standard verification sample set including a plurality of vehicle environment description characteristic values and a standard behavior decision result;

[0019] a normalized sample set acquisition module configured to perform data preprocessing on the standard verification sample set to obtain a normalized sample set;

[0020] a bat iteration module configured to set parameters based on a bat algorithm, and perform multi-round iteration processing of a bat colony by using the bat algorithm;

[0021] In each iteration round, the vehicle behavior decision model is updated by using each bat matched with the current iteration round, the normalized sample set is input into each updated vehicle behavior decision model, the fitness of each updated vehicle behavior decision model is used as an evaluation index, the best bat in the current iteration round is identified from the bats, and the bat colony in the current iteration round is optimized by using the best bat in the current iteration round, the bats including a penalty factor and a Gaussian kernel function;

[0022] a global best bat acquisition module configured to acquire a global best bat when a bat iteration end condition is met;

[0023] The vehicle behavior decision model updating module is configured to update the vehicle behavior decision model by using the global optimal bat to obtain an optimized vehicle behavior decision model.

[0024] According to another aspect of the present application, a vehicle behavior decision device is provided, comprising:

[0025] The environment description feature value acquisition module is configured to acquire a plurality of vehicle environment description feature values corresponding to at least one associated vehicle related to the vehicle behavior decision of the host vehicle through information acquisition results of a plurality of sensors arranged in the vehicle.

[0026] The vehicle environment description feature values include current vehicle speeds of the host vehicle and each associated vehicle, current vehicle accelerations of the host vehicle and each associated vehicle, current vehicle left front coordinates of the host vehicle and each associated vehicle, current vehicle front longitudinal coordinates of the host vehicle and each associated vehicle, current vehicle widths of the host vehicle and each associated vehicle, current vehicle lengths of the host vehicle and each associated vehicle, a distance between the host vehicle and at least one front vehicle, and a distance between the host vehicle and at least one rear vehicle.

[0027] The behavior decision accuracy rate acquisition module is configured to input the plurality of vehicle environment description feature values into the plurality of optimized vehicle behavior decision models optimized by the method of any one of the first and second embodiments to acquire accuracy rates corresponding to each vehicle behavior decision output by the plurality of optimized vehicle behavior decision models.

[0028] The vehicle control module is configured to select a vehicle behavior decision with the highest accuracy rate as a vehicle behavior decision result and perform a vehicle control operation matched with the vehicle behavior decision result.

[0029] According to another aspect of the present application, an electronic device is provided, comprising:

[0030] at least one processor; and

[0031] a memory in communication with the at least one processor; wherein

[0032] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the vehicle behavior decision model optimization method according to the first and second embodiments of the present application, or to implement the vehicle behavior decision method according to the third embodiment of the present application.

[0033] According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the optimization method of the vehicle behavior decision model according to the first and second embodiments of the present application, or the vehicle behavior decision method according to the third embodiment of the present application.

[0034] The technical solution of the embodiments of the present application can effectively improve the accuracy of the optimized vehicle behavior decision model by performing data preprocessing on the standard verification sample set to obtain a normalized sample set, and optimizing the vehicle behavior decision model based on the normalized sample set and the bat algorithm.

[0035] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0037] Figure 1 is a flowchart of the optimization method of the vehicle behavior decision model according to the first embodiment of the present application;

[0038] Figure 2 is a flowchart of another optimization method of the vehicle behavior decision model according to the second embodiment of the present application;

[0039] Figure 3a is a flowchart of the vehicle behavior decision method according to the third embodiment of the present application;

[0040] Figure 3b is a vehicle position relationship diagram according to the third embodiment of the present application;

[0041] Figure 4 is a structural schematic diagram of an optimization device of the vehicle behavior decision model according to the fourth embodiment of the present application;

[0042] Figure 5 is a structural schematic diagram of a vehicle behavior decision device according to the fifth embodiment of the present application;

[0043] Figure 6 is a structural schematic diagram of an electronic device implementing the optimization method of the vehicle behavior decision model and the vehicle behavior decision method according to the embodiments of the present application. DETAILED DESCRIPTION

[0044] In order to make the personnel in the technical field better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor should belong to the protection scope of the present application.

[0045] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0046] Embodiment one

[0047] Figure 1 A flowchart of a method for optimizing a vehicle behavior decision model is provided for the first embodiment of the present application. The present embodiment can be applicable to the case where a normalized sample set is obtained by data preprocessing on a standard verification sample set, and the vehicle behavior decision model is optimized based on the normalized sample set and the bat algorithm. The method can be executed by a vehicle behavior decision model optimization device, which can be realized in the form of hardware and / or software, and can be configured in a computer with data processing function. As shown in the figure, the method comprises: Figure 1

[0048] S110, obtaining a vehicle behavior decision model to be optimized and a standard verification sample set matched with the vehicle behavior decision model, the standard verification sample set including a plurality of vehicle environment description feature values and a standard behavior decision result.

[0049] Preferably, in the embodiments of the present application, SVM (Support Vector Machine) is selected as the vehicle behavior decision model, and SVM is a support vector machine capable of solving binary classification problems.

[0050] ​The vehicle environment description feature value is input data of the vehicle behavior decision model, and after the vehicle environment description feature is input to the vehicle behavior decision model, the vehicle behavior decision model can output a corresponding decision result.

[0051] In S120, data preprocessing is performed on the standard verification sample set to obtain a normalized sample set.

[0052] In the normalized sample set, each normalized sample can include a plurality of vehicle environment description feature values and standard behavior decision results. The vehicle environment description feature values can include the vehicle type of the current vehicle and each related vehicle, the current vehicle speed of the current vehicle and each related vehicle, the current vehicle acceleration of the current vehicle and each related vehicle, the current vehicle left front coordinate of the current vehicle and each related vehicle, the current vehicle front longitudinal coordinate of the current vehicle and each related vehicle, the current vehicle width of the current vehicle and each related vehicle, the current vehicle length of the current vehicle and each related vehicle, the distance between the current vehicle and at least one front vehicle, and the distance between the current vehicle and at least one rear vehicle. The standard behavior decision results can include the behavior accuracy of each vehicle.

[0053] The data preprocessing can include normalization processing of the standard verification sample set, unifying the data standard, and making each feature in the normalized sample set more balanced.

[0054] In S130, parameters are set based on the bat algorithm, and the bat algorithm is used for multi-round iteration processing of the bat colony.

[0055] The bat algorithm parameter setting can include the number of bat colonies, the maximum number of iterations, the initial position of the bat, the initial speed of the bat, and the fitness function.

[0056] In each iteration process, the fitness of the normalized sample set input to each updated vehicle behavior decision model after updating the vehicle behavior decision model by each bat matching the current iteration round is used as an evaluation index, the best bat is identified among the bats, and the best bat is used for iteration optimization of the bat colony in the current round. The bat includes a penalty factor and a Gaussian kernel function.

[0057] It can be understood that the bat proposed in the embodiment of the application includes a penalty factor and a Gaussian kernel function, that is, the bat coordinates are composed of the penalty factor and the Gaussian kernel function. The iteration process of the bat colony using the bat algorithm can be understood as the optimization process of the penalty factor and the Gaussian kernel function. Each bat colony can include a plurality of bats, that is, each bat colony can represent a plurality of different penalty factors and Gaussian kernel functions.

[0058] In S140, when the bat iteration end condition is met, the global optimal bat is obtained.

[0059] In a specific embodiment, when the preset maximum number of iterations has been reached and there is a fitness value greater than the preset minimum fitness value in the bat population, the bat iteration end condition can be considered to be met. Alternatively, a random number can be generated, and the relationship between the random number and the pulse emission rate can be used as one of the bat iteration end conditions.

[0060] S150, updating the vehicle behavior decision model using the global optimal bat to obtain an optimized vehicle behavior decision model.

[0061] When the global optimal bat is obtained, the optimal penalty factor and the optimal Gaussian kernel function can be obtained, and the optimal penalty factor and the optimal Gaussian kernel function can be input into the vehicle behavior decision model to obtain the optimized vehicle behavior decision model.

[0062] The technical scheme of the embodiment of the application can effectively improve the accuracy of the optimized vehicle behavior decision model by performing data preprocessing on the standard verification sample set to obtain a normalized sample set, and optimizing the vehicle behavior decision model based on the bat algorithm using the normalized sample set.

[0063] Embodiment Two

[0064] Figure 2 A flowchart of a vehicle behavior decision model optimization method provided by Embodiment Two of the application is provided, and the vehicle behavior decision model optimization method is specifically described based on the above-mentioned embodiments. As shown in Figure 2 The method comprises the following steps:

[0065] S210, obtaining a vehicle behavior decision model to be optimized and a standard verification sample set matched with the vehicle behavior decision model, the standard verification sample set comprising a plurality of vehicle environment description feature values and standard behavior decision results.

[0066] S220, performing data normalization processing on each vehicle environment description feature value in each standard verification sample to obtain a normalized sample set subjected to data normalization processing.

[0067] S230, updating the vehicle behavior decision model to be optimized using a bat initial position, inputting the normalized data into the updated vehicle behavior decision model, obtaining a current optimal bat through a preset fitness function, and saving information related to the current optimal bat.

[0068] The information related to the optimal bat can include the position of the optimal bat, the current iteration speed of the optimal bat, etc.

[0069] S240, iterating each bat in the current bat population according to the information related to the current optimal bat.

[0070] S250, calculate the optimal flight position of the current iteration according to the optimal bat-related information in the last iteration round, and update the position of each bat after the last iteration according to the optimal flight position and the initial speed of the bat.

[0071] The calculation of the optimal flight position of the current iteration can specifically include: calculating the optimal flight position of the current iteration according to the position of the optimal bat in the last iteration round and a preset position iteration function.

[0072] S260, compare the fitness value of each bat after the current iteration with the fitness value of each bat after the last iteration, save the position of each bat with a higher fitness value in the two iterations, update the related parameters required for the next iteration, and perform the next iteration until the bat iteration stopping condition is met, and stop the bat iteration.

[0073] It should be noted that the bat iteration stopping condition needs to be detected after the first iteration, and in this embodiment, the step of judging whether the bat iteration stopping condition is met is mentioned only to facilitate understanding of the entire bat iteration process, and does not limit the position of the bat iteration stopping condition.

[0074] S270, when the bat iteration ending condition is met, calculate the final fitness value of each bat for the behavior decision model according to the normalized sample set.

[0075] It should be noted that the fitness value of the bat is related to the accuracy of the decision result of the vehicle behavior decision model, that is, the higher the fitness value of the bat, the higher the accuracy of the decision result of the vehicle behavior decision model.

[0076] S280, obtain the bat with the highest fitness value, and take the bat with the highest fitness value as the global optimal bat.

[0077] S290, update the vehicle behavior decision model using the global optimal bat to obtain an optimized vehicle behavior decision model.

[0078] The technical scheme of the embodiment of the application, by means of data normalization processing of the vehicle environment description characteristic value in the standard verification sample set, makes each characteristic value used for optimizing the vehicle behavior decision model more balanced, and improves the decision accuracy of the vehicle behavior decision model.

[0079] Embodiment three

[0080] Figure 3aA flowchart of a vehicle behavior decision method provided for Embodiment Three of the present application, the embodiment can be applicable to collect a plurality of vehicle environment description characteristic values related to vehicle behavior decision, and control the vehicle according to the behavior decision result output by the optimized vehicle behavior decision model. The method can be executed by a vehicle behavior decision device, which can be realized in the form of hardware and / or software, and can be configured in a vehicle machine system with data processing function. As shown in Figure 3a The method comprises the following steps:

[0081] S310, acquiring a plurality of vehicle environment description characteristic values corresponding to at least one associated vehicle related to the vehicle and vehicle behavior decision through the information collection results of a plurality of sensors arranged in the vehicle.

[0082] The vehicle environment description characteristic values include the vehicle types of the vehicle and each associated vehicle, the current vehicle speeds of the vehicle and each associated vehicle, the current vehicle accelerations of the vehicle and each associated vehicle, the current vehicle left front coordinates of the vehicle and each associated vehicle, the current vehicle front longitudinal coordinates of the vehicle and each associated vehicle, the current vehicle widths of the vehicle and each associated vehicle, the current vehicle lengths of the vehicle and each associated vehicle, the distances between the vehicle and at least one front vehicle, and the distances between the vehicle and at least one rear vehicle.

[0083] In order to facilitate understanding of the positional relationship between the vehicle and the associated vehicles in actual application, Figure 3b a positional relationship diagram of the vehicle and the associated vehicles is shown. As shown in Figure 3b Vehicle M is the vehicle, vehicle MP is the front vehicle of the vehicle, vehicle MB is the rear vehicle of the vehicle, vehicle TP is the front vehicle of the adjacent lane of the vehicle, and vehicle TB is the rear vehicle of the adjacent lane of the vehicle.

[0084] Based on the vehicle positional relationship diagram as shown in Figure 3b The vehicle environment description characteristic values can include the vehicle types of vehicle M, vehicle MP, vehicle MB, vehicle TP and vehicle TB; the speeds of vehicle M, vehicle MP, vehicle MB, vehicle TP and vehicle TB; the accelerations of vehicle M, vehicle MP, vehicle MB, vehicle TP and vehicle TB; the lengths of vehicle M, vehicle MP, vehicle MB, vehicle TP and vehicle TB; the widths of vehicle M, vehicle MP, vehicle MB, vehicle TP and vehicle TB; the current vehicle left front coordinates of vehicle M, vehicle TP and vehicle TB; the current vehicle front longitudinal coordinates of vehicle M, vehicle TP and vehicle TB; the distance between vehicle M and vehicle MP; and the distance between vehicle M and vehicle MB.

[0085] The type of the vehicle M is a car type, and the types of the vehicle MP, the vehicle MB, the vehicle TP, and the vehicle TB can be obtained according to information collection results of sensors, and can include a truck type, a car type, and a motorcycle type.

[0086] S320, inputting the plurality of vehicle environment description feature values into a plurality of optimized vehicle behavior decision models to obtain accuracy rates of vehicle behavior decisions corresponding to each vehicle behavior decision output by the plurality of optimized vehicle behavior decision models.

[0087] In order to realize simultaneous decision of a plurality of vehicle behaviors, the present application creatively proposes to obtain accuracy rates of vehicle behavior decisions through a plurality of vehicle behavior decision models, and selects a behavior decision with the highest accuracy rate for vehicle control.

[0088] Optionally, the vehicle behavior can include free driving, vehicle following, and lane changing, and more vehicle behaviors can be simultaneously judged, and the embodiment is only illustrative and does not specifically limit this.

[0089] S330, screening a vehicle behavior decision with the highest accuracy rate as a vehicle behavior decision result, and performing a vehicle control operation matched with the vehicle behavior decision result.

[0090] The technical scheme of the embodiment of the present application can make the vehicle system accurately perform automatic behavior decision according to an external vehicle environment, and can obtain a current optimal vehicle behavior in a plurality of optional vehicle behaviors, and perform a vehicle control operation through the optimal vehicle behavior, thereby effectively improving the intelligentization of the vehicle system.

[0091] Embodiment Four

[0092] Figure 4 A structural schematic diagram of a vehicle behavior decision model optimization device provided by Embodiment Three of the present application. As shown in the figure, the device includes a standard verification sample set acquisition module 410, a normalized sample set acquisition module 420, a bat iteration module 430, a global optimal bat acquisition module 440, and a vehicle behavior decision model updating module 450. Figure 4

[0093] The standard verification sample set acquisition module 410 is configured to acquire a vehicle behavior decision model to be optimized and a standard verification sample set matched with the vehicle behavior decision model, and the standard verification sample includes a plurality of vehicle environment description feature values and a standard behavior decision result.

[0094] The normalized sample set acquisition module 420 is configured to perform data preprocessing on the standard verification sample set to obtain a normalized sample set.​

[0095] a bat iteration module 430, configured to perform multi-round iteration processing on the bat colony based on bat algorithm setting parameters and by using the bat algorithm;

[0096] In each iteration round, the vehicle behavior decision model is updated by using each bat matched with the current iteration round, and then the standard verification sample set is input into the updated vehicle behavior decision model, and the fitness is used as an evaluation index, the optimal bat in the current round is identified in the bats, and the optimal bat in the current round is used to perform iteration optimization of the bat colony in the current round, and the bats include a penalty factor and a Gaussian kernel function.

[0097] a global optimal bat obtaining module 440, configured to obtain the global optimal bat when the bat iteration end condition is met.

[0098] a vehicle behavior decision model updating module 450, configured to update the vehicle behavior decision model by using the global optimal bat to obtain an optimized vehicle behavior decision model.

[0099] The technical scheme of the embodiment of the application can effectively improve the accuracy of the optimized vehicle behavior decision model by performing data preprocessing on the standard verification sample set to obtain a normalized sample set, and optimizing the vehicle behavior decision model based on the bat algorithm and the normalized sample set.

[0100] On the basis of the above embodiments, the normalized sample set obtaining module 420 can be specifically configured to perform data normalization processing on each vehicle environment description feature value in each standard verification sample to obtain a normalized sample set subjected to data normalization processing.

[0101] On the basis of the above embodiments, the bat algorithm setting parameters can include the number of bat colonies, the maximum number of iterations, the initial position of the bat, the initial speed of the bat, and the fitness function.

[0102] On the basis of the above embodiments, the bat iteration module 430 can be specifically configured to:

[0103] The initial position of the bat is used to update the vehicle behavior decision model to be optimized, the normalized data is input into the updated vehicle behavior decision model, the current optimal bat is obtained by using a preset fitness function, and the information related to the current optimal bat is saved;

[0104] According to the information related to the current optimal bat, each bat in the current bat colony is iterated;

[0105] The optimal flight position in the current iteration round is calculated according to the information related to the optimal bat in the last iteration round, and each bat position after the last iteration is updated according to the optimal flight position and the initial speed of the bat.

[0106] The fitness value of each bat after the current iteration is compared with the fitness value of each bat after the last iteration, the positions of the bats with higher fitness values in the two iterations are saved, and the related parameters required for the next iteration are updated, and the next iteration is performed until the bat iteration stopping condition is met, and the bat iteration is stopped.

[0107] On the basis of the above embodiments, the global optimal bat obtaining module 440 can be specifically used for:

[0108] When the bat iteration end condition is met, the final fitness value of each bat for the behavior decision model is calculated according to the normalized sample set;

[0109] The bat with the highest fitness value is obtained, and the bat with the highest fitness value is taken as the global optimal bat.

[0110] The optimization device for the vehicle behavior decision model provided in the embodiments of the present application can execute the optimization method for the vehicle behavior decision model provided in any embodiment of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0111] Embodiment five

[0112] Figure 5 A structural schematic diagram of a vehicle behavior decision device provided in embodiment five of the present application is shown in FIG. 5. Figure 5 As shown in the figure, the device includes an environment description feature value obtaining module 510, a behavior decision accuracy rate obtaining module 520, and a vehicle control module 530.

[0113] The environment description feature value obtaining module 510 is used to obtain a plurality of vehicle environment description feature values corresponding to at least one associated vehicle related to the vehicle behavior decision of the vehicle through the information collection results of a plurality of sensors arranged in the vehicle;

[0114] The vehicle environment description feature values include the vehicle types of the vehicle and each associated vehicle, the current vehicle speeds of the vehicle and each associated vehicle, the current vehicle accelerations of the vehicle and each associated vehicle, the current vehicle left front coordinates of the vehicle and each associated vehicle, the current vehicle front longitudinal coordinates of the vehicle and each associated vehicle, the current vehicle widths of the vehicle and each associated vehicle, the current vehicle lengths of the vehicle and each associated vehicle, the distances between the vehicle and at least one front vehicle, and the distances between the vehicle and at least one rear vehicle.

[0115] The behavior decision accuracy rate obtaining module 520 is used to input the plurality of vehicle environment description feature values into a plurality of optimized vehicle behavior decision models optimized by the method in any one of the embodiments one and two, and obtain the accuracy rates corresponding to each vehicle behavior decision output by the plurality of optimized vehicle behavior decision models.

[0116] The vehicle control module 530 is configured to screen out the vehicle behavior decision with the highest accuracy as a vehicle behavior decision result, and perform a vehicle control operation matched with the vehicle behavior decision result.

[0117] The sensor collects the vehicle environment description characteristic value, and inputs the vehicle environment description characteristic value into the optimized vehicle behavior decision model to obtain a behavior decision result, so that the vehicle control system can accurately make automatic behavior decisions according to the external vehicle environment, and can obtain the current optimal vehicle behavior in a plurality of optional vehicle behaviors, and perform a vehicle control operation through the optimal vehicle behavior, thereby effectively improving the intelligence of the vehicle control system.

[0118] The vehicle behavior decision device provided in the embodiments of the present application can execute the vehicle behavior decision method provided in the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0119] Embodiment six

[0120] Figure 6 A structural schematic diagram of an electronic device 60 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart phones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0121] As shown in Figure 6 The electronic device 60 includes at least one processor 61, and a memory, such as a read-only memory (ROM) 62, a random access memory (RAM) 63, etc., which are communicatively connected to the at least one processor 61, wherein the memory stores a computer program executable by the at least one processor. The processor 61 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 62 or loaded from the storage unit 68 into the random access memory (RAM) 63. In the RAM 63, various programs and data required for the operation of the electronic device 60 can also be stored. The processor 61, the ROM 62, and the RAM 63 are connected to each other through a bus 64. An input / output (I / O) interface 65 is also connected to the bus 64.

[0122] A plurality of components in the electronic device 60 are connected to the I / O interface 65, including: an input unit 66, such as a keyboard, a mouse, etc.; an output unit 67, such as various types of displays, speakers, etc.; a storage unit 68, such as a magnetic disk, an optical disk, etc.; and a communication unit 69, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 69 allows the electronic device 60 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0123] The processor 61 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the processor 61 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 61 performs various methods and processes described above, such as the optimization method of a vehicle behavior decision model and the vehicle behavior decision method as described in embodiments of the present application.

[0124] In some embodiments, the optimization method of a vehicle behavior decision model and the vehicle behavior decision method can be implemented as a computer program tangibly embodied in a computer readable storage medium, such as the storage unit 68. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 60 via the ROM 62 and / or the communication unit 69. When the computer program is loaded onto the RAM 63 and executed by the processor 61, one or more steps of the optimization method of a vehicle behavior decision model and the vehicle behavior decision method described above can be performed. Alternatively, in other embodiments, the processor 61 can be configured to perform the optimization method of a vehicle behavior decision model and the vehicle behavior decision method by any other appropriate means, such as by means of firmware.

[0125] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0126] Computer programs for implementing the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be implemented on general purpose computers, special purpose computers, or other programmable data processing apparatus to produce the functions / acts specified in the flow diagrams and / or block diagrams. Computer programs can be applied to input data to perform the functions of the present application and to generate output information. The output information can be applied to one or more output devices such as a display screen, printer, storage, etc. These functions / acts performed by the computer programs are referred to as instructions.

[0127] In the context of the present application, a computer-readable storage medium can be any tangible medium that can contain or store program instructions for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0128] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0129] The systems and techniques described herein can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described herein, or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0130] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0131] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be performed in parallel, in series, or in a different order, and the present disclosure is not limited in this regard.

[0132] The specific embodiments described above are not intended to limit the scope of the present disclosure. Those skilled in the art will understand that various modifications, combinations, sub-combinations, and alternatives can be made to the specific embodiments without departing from the spirit and principles of the present disclosure. Any further modifications, equivalents, and / or alternatives come within the scope of the present disclosure as recited by the claims.

Claims

1. A method for optimizing a vehicle behavior decision model, characterized in that: include: Obtaining a vehicle behavior decision model to be optimized and a standard verification sample set that matches the vehicle behavior decision model, the standard verification sample including a plurality of vehicle environment description feature values ​​and a standard behavior decision result; wherein the vehicle environment description feature values ​​include vehicle type of the current vehicle and each related vehicle, current vehicle speed of the current vehicle and each related vehicle, current vehicle acceleration of the current vehicle and each related vehicle, current vehicle left front coordinates of the current vehicle and each related vehicle, current vehicle front longitudinal coordinates of the current vehicle and each related vehicle, current vehicle width of the current vehicle and each related vehicle, current vehicle length of the current vehicle and each related vehicle, distance between the current vehicle and at least one vehicle ahead, and distance between the current vehicle and at least one vehicle behind; and the standard behavior decision result including the accuracy rate of each vehicle behavior; Perform data preprocessing on the standard validation sample set to obtain a normalized sample set; Parameters are set based on the bat algorithm, and multiple rounds of iterative processing of bat groups are performed using the bat algorithm; In each round of iteration, the vehicle behavior decision model is updated using the bats that match the current iteration. The fitness of the standard verification sample set input into each updated vehicle behavior decision model is used as the evaluation indicator. The optimal bat of the current round is identified among the bats, and the optimal bat of the current round is used to iteratively optimize the bat group of the current round. The bat includes a penalty factor and a Gaussian kernel function. When the bat iteration end condition is met, the global optimal bat is obtained; The vehicle behavior decision model is updated using the global optimal bat to obtain the optimized vehicle behavior decision model; Among them, the bat algorithm setting parameters include the number of bat groups, the maximum number of iterations, the initial position of the bats, the initial speed of the bats and the fitness function; The method of setting parameters based on the bat algorithm and performing multiple rounds of iterative processing of the bat group using the bat algorithm includes: Using the initial position of the bat to update the vehicle behavior decision model to be optimized, inputting the normalized data into the updated vehicle behavior decision model, obtaining the current optimal bat through a preset fitness function, and saving the relevant information of the current optimal bat; Iterate each bat in the current bat group according to the current optimal bat related information; The optimal flight position of this iteration is calculated based on the optimal bat-related information in the previous iteration, and the positions of each bat after the previous iteration are updated iteratively based on the optimal flight position and the initial speed of the bat; Compare the fitness values ​​of each bat after this round of iteration with the fitness values ​​of each bat after the previous round of iteration, save the positions of the bats with higher fitness values ​​in the two rounds of iteration, and update the relevant parameters required for the next round of iteration. Perform the next round of iteration until the bat iteration stopping condition is met, and then stop the bat iteration. The optimal flight position of this iteration is calculated, including: calculating the optimal flight position of this iteration according to the position of the optimal bat in the previous iteration and a preset position iteration function.

2. The method according to claim 1, characterized in that Perform data preprocessing on the standard validation sample set to obtain a normalized sample set, including: Data normalization processing is performed on each vehicle environment description feature value in each standard verification sample to obtain a normalized sample set that has undergone data normalization processing.

3. The method according to claim 1, characterized in that When the bat iteration end condition is met, obtaining the global optimal bat includes: When the bat iteration end condition is met, the final fitness value of the behavior decision model for each bat is calculated based on the normalized sample set; Get the bat with the highest fitness value and take it as the global optimal bat.

4. A vehicle behavior decision method, executed by a vehicle computer system, characterized in that: include: Acquire, through information collection results from a plurality of sensors disposed in the vehicle, a plurality of vehicle environment description feature values ​​corresponding to at least one associated vehicle related to the vehicle and the vehicle behavior decision; The vehicle environment description feature values ​​include the vehicle type of the vehicle and each related vehicle, the current vehicle speed of the vehicle and each related vehicle, the current vehicle acceleration of the vehicle and each related vehicle, the current vehicle left front coordinate of the vehicle and each related vehicle, the current vehicle front longitudinal coordinate of the vehicle and each related vehicle, the current vehicle width of the vehicle and each related vehicle, the current vehicle length of the vehicle and each related vehicle, the distance between the vehicle and at least one vehicle in front, and the distance between the vehicle and at least one vehicle behind; Inputting the plurality of vehicle environment description feature values ​​into a plurality of optimized vehicle behavior decision models optimized by the method according to any one of claims 1 to 3, and obtaining the accuracy corresponding to each vehicle behavior decision output by the plurality of optimized vehicle behavior decision models; The vehicle behavior decision with the highest accuracy is screened out as the vehicle behavior decision result, and a vehicle control operation matching the vehicle behavior decision result is executed.

5. An optimization device for a vehicle behavior decision model, characterized in that: include: a standard verification sample set acquisition module, configured to acquire a vehicle behavior decision model to be optimized and a standard verification sample set matching the vehicle behavior decision model, wherein the standard verification sample includes a plurality of vehicle environment description feature values ​​and a standard behavior decision result; wherein the vehicle environment description feature values ​​include the vehicle type of the current vehicle and each related vehicle, the current vehicle speed of the current vehicle and each related vehicle, the current vehicle acceleration of the current vehicle and each related vehicle, the current vehicle left front coordinate of the current vehicle and each related vehicle, the current vehicle front longitudinal coordinate of the current vehicle and each related vehicle, the current vehicle width of the current vehicle and each related vehicle, the current vehicle length of the current vehicle and each related vehicle, the distance between the current vehicle and at least one vehicle ahead, and the distance between the current vehicle and at least one vehicle behind; and the standard behavior decision result includes the accuracy rate of each vehicle behavior; The normalized sample set acquisition module is used to perform data preprocessing on the standard verification sample set to obtain a normalized sample set; The bat iteration module is used to set parameters based on the bat algorithm and use the bat algorithm to perform multiple rounds of iterative processing of bat groups; In each round of iteration, the vehicle behavior decision model is updated using the bats that match the current iteration. The fitness of the standard verification sample set input into each updated vehicle behavior decision model is used as the evaluation indicator. The optimal bat of the current round is identified among the bats, and the optimal bat of the current round is used to iteratively optimize the bat group of the current round. The bat includes a penalty factor and a Gaussian kernel function. The global optimal bat acquisition module is used to obtain the global optimal bat when the bat iteration end condition is met; The vehicle behavior decision model update module is used to update the vehicle behavior decision model using the global optimal bat to obtain the optimized vehicle behavior decision model; Among them, the bat algorithm setting parameters include the number of bat groups, the maximum number of iterations, the initial position of the bats, the initial speed of the bats and the fitness function; Among them, the bat iteration module is specifically used for: Using the initial position of the bat to update the vehicle behavior decision model to be optimized, inputting the normalized data into the updated vehicle behavior decision model, obtaining the current optimal bat through a preset fitness function, and saving the relevant information of the current optimal bat; Iterate each bat in the current bat group according to the current optimal bat related information; The optimal flight position of this iteration is calculated based on the optimal bat-related information in the previous iteration, and the positions of each bat after the previous iteration are updated iteratively based on the optimal flight position and the initial speed of the bat; Compare the fitness values ​​of each bat after this round of iteration with the fitness values ​​of each bat after the previous round of iteration, save the positions of the bats with higher fitness values ​​in the two rounds of iteration, and update the relevant parameters required for the next round of iteration. Perform the next round of iteration until the bat iteration stopping condition is met, and then stop the bat iteration. The optimal flight position of this iteration is calculated, including: calculating the optimal flight position of this iteration according to the position of the optimal bat in the previous iteration and a preset position iteration function.

6. A vehicle behavior decision-making device, characterized in that: include: An environment description feature value acquisition module is used to acquire multiple vehicle environment description feature values ​​corresponding to at least one associated vehicle related to the vehicle and the vehicle behavior decision based on information collection results of multiple sensors provided in the vehicle; The vehicle environment description feature values ​​include the vehicle type of the vehicle and each related vehicle, the current vehicle speed of the vehicle and each related vehicle, the current vehicle acceleration of the vehicle and each related vehicle, the current vehicle left front coordinate of the vehicle and each related vehicle, the current vehicle front longitudinal coordinate of the vehicle and each related vehicle, the current vehicle width of the vehicle and each related vehicle, the current vehicle length of the vehicle and each related vehicle, the distance between the vehicle and at least one vehicle in front, and the distance between the vehicle and at least one vehicle behind; a behavior decision accuracy acquisition module, configured to input the plurality of vehicle environment description feature values ​​into a plurality of optimized vehicle behavior decision models optimized using the method according to any one of claims 1 to 3, and obtain the accuracy corresponding to each vehicle behavior decision output by the plurality of optimized vehicle behavior decision models; The vehicle control module is used to screen out the vehicle behavior decision with the highest accuracy as the vehicle behavior decision result, and execute the vehicle control operation matching the vehicle behavior decision result.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the optimization method of the vehicle behavior decision model described in any one of claims 1 to 3, or execute the vehicle behavior decision method described in claim 4.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, which are used to enable a processor to implement the optimization method of the vehicle behavior decision model described in any one of claims 1 to 3, or to execute the vehicle behavior decision method described in claim 4 when executed.

Images