Vehicle positioning method and system in weak GNSS environment

Abstract

A vehicle positioning method and system in a weak GNSS environment. The vehicle positioning method in a weak GNSS environment comprises: preprocessing system operation data information of an autonomous vehicle, wherein the system operation data information includes longitude and latitude data acquired from a GNSS, and a three-axis acceleration, a three-axis angular velocity, and heading information which are acquired from an INS; using the preprocessed system operation data information acquired from the INS to train a position prediction network, to finally output supervision information of the position prediction network; training a position correction network on the basis of said supervision information, the preprocessed system operation data information, and supervision information of the position correction network, to finally output a predicted value for correction; and in a weak GNSS environment or during GNSS interruption, outputting a final predicted value by means of the position prediction network and the position correction network. The vehicle positioning performance of a GNSS / INS integrated navigation system under a weak GNSS condition is improved by only using data of the GNSS / INS integrated navigation system, without adding an additional sensor.

Description

Vehicle positioning method and system in weak GNSS environment

[0001] Cross-reference to Related Applications

[0002] The present application claims priority to the Chinese patent application No. 202410409727.9, filed on April 7, 2024, and entitled “Vehicle positioning method and system in weak GNSS environment”, the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present application belongs to the field of electrical test, and particularly relates to a vehicle positioning method and system in weak GNSS environment. BACKGROUND

[0004] An automatic driving car positioning system aims to provide position information meeting certain accuracy requirements for a vehicle to assist the vehicle in making intelligent decisions. Vehicle positioning technology is a basic technology that integrates many high technologies such as sensor technology, computer technology, communication technology, information processing technology, and artificial intelligence technology. Global Navigation Satellite System (GNSS) and Inertial Navigation System (INS) are two commonly used systems in ground vehicle positioning.

[0005] In some scenarios (such as tunnels, urban canyons, etc.), GNSS signals cannot be received by the corresponding signal receivers due to shielding, and INS has the defect of cumulative positioning error. The combination of the two can improve the performance of the positioning system to some extent, but it is still difficult to meet the positioning requirements in complex environments.

[0006] To solve the problem of how to continuously and stably provide position information meeting accuracy requirements under weak GNSS signal conditions, one solution is to add more vehicle-mounted sensors for auxiliary positioning, but this way increases the hardware cost and complexity of the system, and also puts higher requirements on multi-sensor data fusion algorithms. SUMMARY

[0007] The present application provides a vehicle positioning method and system in weak GNSS environment to solve the problem of how to improve the vehicle positioning performance of GNSS / INS integrated navigation system under weak GNSS conditions without adding additional sensors, by training position prediction network and position correction network in stages using only the data of GNSS / INS integrated navigation system.

[0008] To solve the above technical problems, the present application provides a vehicle positioning method in weak GNSS environment, comprising:

[0009] Step S1: preprocessing system running data information in normal driving state of the autonomous vehicle, wherein the system running data information comprises longitude and latitude data obtained from a GNSS system and three-axis acceleration, three-axis angular velocity and heading information obtained from an INS system;

[0010] Step S2: training a position prediction network with the preprocessed three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system; taking the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system as input of the trained prediction network, and taking the difference between the GNSS system measurement value and the INS system measurement value as the supervision information of the prediction network;

[0011] Step S3: training a position correction network according to the supervision information in step S2 and the preprocessed three-axis acceleration, three-axis angular velocity, heading information obtained from the INS system and the supervision information of the position correction network; similarly taking the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system as input of the trained position correction network, and taking the error between the GNSS system position prediction value and the measurement value as the input of the position correction network to correct the prediction value of the position prediction network;

[0012] Step S4: in a weak GNSS environment or GNSS interruption case, taking the first-order difference of the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system and the attitude information of the autonomous vehicle as input of the trained position prediction network and position correction network to further correct the prediction value of the position prediction network to obtain the final prediction value of the position information of the autonomous vehicle.

[0013] Optionally, preprocessing the system running data information in the normal driving state of the autonomous vehicle comprises:

[0014] Optionally, the linear interpolation method is used to complete the full error or incomplete field in the system running data information.

[0015] Optionally, the supervision information of the prediction network is the difference between the position measurement value of the GNSS system and the position measurement value of the INS system in the same coordinate system.

[0016] Optionally, the supervision information of the position correction network is the difference between the position prediction value of the GNSS system and the position prediction value of the position prediction network in the same coordinate system.

[0017] Optionally, the position prediction value of the position prediction network is the sum of the supervision information of the prediction network and the corrected three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system.

[0018] Optionally, the LSTM network hidden layer number of the position prediction network is set to 2 layers, the network input layer has 7 neurons, the length of the input time series data is 30, and the output layer contains three neurons.

[0019] Optionally, the LSTM network hidden layer number of the position correction network is set to 2 layers, and the output layer contains three neurons.

[0020] The application also provides a vehicle positioning system in a weak GNSS environment, comprising:

[0021] A data acquisition module is configured to preprocess system running data information of the autonomous vehicle in a normal driving state, wherein the system running data information comprises latitude and longitude data obtained from a GNSS system and three-axis acceleration, three-axis angular velocity and heading information obtained from an INS system.

[0022] A position prediction network module is configured to train a position prediction network using the preprocessed three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system; the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system are used as the input of the trained prediction network, and the difference between the GNSS system measurement value and the INS system measurement value is used as the supervision information of the prediction network.

[0023] A position correction network module is configured to train a position correction network using the supervision information in step S2 and the preprocessed three-axis acceleration, three-axis angular velocity, heading information obtained from the INS system and the supervision information of the position correction network; similarly, the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system are used as the input of the trained position correction network, and the error between the position prediction value of the GNSS system and the measurement value is used to correct the prediction value of the position prediction network.

[0024] An output module is configured to use the first-order difference of the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system and the attitude information of the autonomous vehicle as the input of the trained position prediction network and position correction network, and further correct the prediction value of the position prediction network to obtain the final prediction value of the autonomous vehicle position information.

[0025] The application also provides a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the vehicle positioning method in a weak GNSS environment according to any one of the above embodiments when executing the computer program.

[0026] The application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the vehicle positioning method in a weak GNSS environment according to any one of the above embodiments.

[0027] Advantages of the application:

[0028] The application aims to provide a vehicle positioning method and system in a weak GNSS environment without adding other vehicle-mounted sensors. The method trains a position prediction neural network and a position correction neural network in stages, and improves the vehicle positioning performance of a GNSS / INS integrated navigation system in a weak GNSS condition. The two neural networks both take the output of an inertial navigation system as network input. In the first stage, the position prediction network is trained by taking the error of a GNSS system and an INS system as supervision information; in the second stage, the position correction network is trained by taking the residual error of a GNSS measurement value and a GNSS prediction value as supervision information. The trained prediction network and correction network are used simultaneously in a weak GNSS environment to predict the vehicle position. BRIEF DESCRIPTION OF DRAWINGS

[0029] Fig. 1 is a step flowchart of a vehicle positioning method in a weak GNSS environment provided by the application;

[0030] Fig. 2 is an installation schematic diagram of a GNSS / INS micro-electro-mechanical system, a GNSS antenna and other related devices provided by the application;

[0031] Fig. 3 is a training process schematic diagram of a position prediction network and a position correction network provided by the application;

[0032] Fig. 4 is a use schematic diagram of a position prediction network and a correction network in a weak GNSS environment or a GNSS interruption condition provided by the application;

[0033] Fig. 5 is a schematic diagram of a process from training to use of a position prediction network and a position correction network provided by the application;

[0034] Fig. 6 is a schematic diagram of a vehicle positioning system in a weak GNSS environment provided by the application. DETAILED DESCRIPTION

[0035] In order to enable persons skilled in the art to better understand the application scheme, the technical solutions in the embodiments of the application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, but not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by persons skilled in the art without creative labor should belong to the protection scope of the application.

[0036] It should be noted that the terms "first", "second", and the like in the description and in the claims of the present application and above-described accompanying drawings are used to distinguish similar objects, and are not necessarily used to describe a particular chronological or sequential order. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that 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 apparatus including a series of steps or units does not necessarily have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to such a process, method, product, or apparatus.

[0037] The present application will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0038] The present application provides a vehicle positioning method in a weak GNSS environment, as shown in FIG. 1. It should be noted that in the present embodiment, a GNSS / INS combined navigation system is taken as an example, and the data acquisition device used includes a high-precision micro-electromechanical combined navigation system installed in the vehicle and a set of GNSS antennas for receiving satellite signals, an industrial computer for recording the GNSS log of the combined navigation system, and a power supply device matched with the above-mentioned devices. The device installation is shown in FIG. 2, and the calibrated combined navigation system and industrial computer are placed fixedly at the rear of the vehicle compartment. The GNSS antennas are respectively screwed onto two strong magnetic suction discs and are respectively fixed and placed in the forward and backward directions of the autonomous vehicle. They are placed as high as possible in the autonomous vehicle to ensure that good GNSS signals can be received, and at the same time, the line formed by the phase centers of the two GNSS antennas is consistent or parallel with the central axis direction of the test carrier.

[0039] In combination with FIG. 1, the method includes:

[0040] Step S1: pre-processing the system running data information of the autonomous vehicle in a normal driving state, wherein the system running data information includes latitude and longitude data obtained from the GNSS system and three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system;

[0041] In one embodiment, the system running data information of the autonomous vehicle in a normal driving state is obtained from the autonomous driving sensors of the autonomous vehicle, and the GNSS data and INS data of the autonomous vehicle during operation are extracted from the obtained system running data information, including: lat,lon,ω x ,ω y ,ω z ,a x ,a ya z ,β

[0042] wherein lat represents latitude data obtained from the GNSS system, lon represents longitude data obtained from the GNSS system (unit: degree), ω x ,ω y ,ω z respectively represent three-axis angular velocities (unit: degree / second) in x-axis, y-axis, z-axis directions obtained from the INS system, a x ,a y ,a z respectively represent three-axis accelerations (unit: gravity acceleration g) in x-axis, y-axis, z-axis directions obtained from the INS system, and β represents a heading.

[0043] Step S2: training the position prediction network with the preprocessed three-axis accelerations, three-axis angular velocities and heading information obtained from the INS system; taking the three-axis accelerations, three-axis angular velocities and heading information obtained from the INS system as inputs of the trained prediction network, and outputting the supervision information of the prediction network composed of the difference between the GNSS system measurement value and the INS system measurement value;

[0044] In an embodiment, the latitude and longitude data obtained from the GNSS system is projected by Mercator projection to obtain position information p GNSS , and the position information obtained by mathematical integration of the three-axis accelerations, three-axis angular velocities and heading information obtained from the INS system is denoted as p INS .

[0045] The three-axis accelerations, three-axis angular velocities and heading information obtained from the INS system are taken as the vehicle attitude information ψ, including ω x ,ω y ,ω z ,a x ,a y ,a z . The vehicle attitude information, the first-order difference Δψ of the attitude information, and the heading information β are taken as inputs of the position prediction network,

[0046] The difference Δp between p GNSS and p INS in the same coordinate system is taken as the supervision information of the position prediction network, and the position prediction network is trained.

[0047] It is noted that the input information of the network is all taken from the INS system, and after the training is completed, the position prediction network will have the ability to take the INS system information as input to predict the GNSS position p' GNSS , and output the supervision information of the prediction network composed of the difference between the GNSS system measurement value and the INS system measurement value for the position correction network training.

[0048] Step S3: training the position correction network according to the supervision information in step S2 and the pre-processed three-axis acceleration, three-axis angular velocity, heading information and supervision information of the position correction network obtained from the INS system; similarly, taking the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system as the input of the trained position correction network, outputting the error between the position prediction value of the GNSS system and the measured value, and correcting the prediction value of the position prediction network;

[0049] In an embodiment, the network still takes the vehicle attitude information ψ, the first-order differential Δψ of the attitude information, and the heading information β as the input of the position correction network, and takes the error e GNSS between the GNSS system measured value p GNSS and the prediction value p' p of the prediction network as the supervision information of the position correction network to train the network. The input information of the network is also taken from the INS system, and after the training is completed, the position correction network outputs the error between the position prediction value of the GNSS system and the measured value, and corrects the prediction value of the position prediction network.

[0050] It should be noted that, as shown in FIG. 3, the training processes of steps S2 and S3 include two stages, stage 1, the position prediction network training process; and stage 2, the position correction network training process.

[0051] Step S4: in the weak GNSS environment or GNSS interruption case, taking the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system and the first-order differential of the attitude information of the autonomous vehicle as the input of the trained position prediction network and position correction network, further correcting the prediction value output by the position prediction network to obtain the final prediction value of the position information of the autonomous vehicle.

[0052] In an embodiment, as shown in FIG. 4, in the weak GNSS environment or GNSS interruption case, the positioning system cannot obtain the position information from the GNSS system, and the input of the above-mentioned position prediction network and correction network is taken from the INS system, so it can be used in the weak GNSS environment or GNSS interruption case, as shown in FIG. 4. When used, the vehicle attitude information ψ, the first-order differential Δψ of the attitude information, and the heading information β are simultaneously input into the two networks, at this time, the position prediction network will predict the vehicle position according to the information given by the INS system, and combine p INS to obtain p' GNSS . At the same time, the position correction network will output error information e p to further correct the prediction value p' GNSS given by the prediction network to obtain the final output value p pred of the vehicle position information.

[0053] In the embodiment, as shown in FIG. 5, the application provides a vehicle positioning method based on a weak GNSS environment. The method trains a position prediction network and a position correction network in two stages. In the first stage, the position prediction network is trained with the error of the GNSS system and the INS system as supervision information. After the training is completed, the position prediction network will have the ability to predict the vehicle position using the INS system information as input. In the second stage, the position correction network is trained with the residual error of the GNSS measurement value and the GNSS prediction value as supervision information. After the training is completed, the position correction network will be able to further correct the prediction value given by the prediction network. Both networks use the INS system data as input, ensuring the availability of the positioning function in a weak GNSS environment or in the case of GNSS interruption.

[0054] Optionally, the pre-processing of the system running data information of the autonomous vehicle in a normal driving state comprises: using a linear interpolation method to complete the full error or incomplete field in the system running data information.

[0055] In an embodiment, in step S1, the directly obtained data has errors or incomplete fields. In this case, the system running data information of the autonomous vehicle in a normal driving state needs to be pre-processed. There are still problems of incomplete, missing and repeated records in the extracted data information file, which cannot be directly used as input to the model for training. In the application, one of the repeated records is retained. For samples with incomplete fields or field errors, the linear interpolation of the corresponding fields of the two adjacent samples before and after the sample is used to complete the error or incomplete field. The pre-processed data is used as the training data of the neural network. The error or incomplete field is removed, so that the accuracy of the network is improved in the training process.

[0056] Optionally, the supervision information of the prediction network is the difference between the position measurement value of the GNSS system and the position measurement value of the INS system in the same coordinate system.

[0057] Optionally, the supervision information of the position correction network is the difference between the position prediction value of the GNSS system and the position prediction value of the position prediction network in the same coordinate system.

[0058] Optionally, the position prediction value of the position prediction network is the sum of the supervision information of the prediction network and the corrected three-axis acceleration, three-axis angular velocity and heading information obtained by the INS system.

[0059] Optionally, the number of hidden layers of the LSTM network of the position prediction network is set to 2 layers, each layer contains 16 memory cells; the input layer of the network has 7 neurons, corresponding to 7 input variables: three-axis acceleration, three-axis angular velocity, heading, the length of the input time series data is 30, and the output layer contains three neurons corresponding to p GNSS and pINS a three-dimensional spatial position difference between the GNSS system measurement value p and the predicted value p'.

[0060] Optionally, the number of LSTM network hidden layers of the position correction network is set to 2 layers, and the output layer contains three neurons, corresponding to the three-dimensional spatial position difference between the GNSS system measurement value p and the predicted value p'. GNSS and the predicted value p' GNSS a three-dimensional spatial position difference between the GNSS system measurement value p and the predicted value p'. It should be noted that the above network hyperparameters can be adjusted according to the actual training situation.

[0061] The application also provides a vehicle positioning system in a weak GNSS environment, as shown in FIG. 6, which comprises:

[0062] a data acquisition module, configured to preprocess system running data information of the autonomous vehicle in a normal driving state, wherein the system running data information comprises latitude and longitude data obtained from a GNSS system and three-axis acceleration, three-axis angular velocity and heading information obtained from an INS system;

[0063] a position prediction network module, configured to train a position prediction network by using the preprocessed three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system; input the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system into the trained position prediction network, and output the supervision information of the position prediction network composed of the difference between the GNSS system measurement value and the INS system measurement value;

[0064] a position correction network module, configured to train a position correction network according to the supervision information in step S2 and the preprocessed three-axis acceleration, three-axis angular velocity, heading information obtained from the INS system and the supervision information of the position correction network; similarly, input the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system into the trained position correction network, and output the error between the position prediction value of the GNSS system and the measurement value, and correct the prediction value of the position prediction network;

[0065] an output module, configured to input the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system and the first-order difference of the attitude information of the autonomous vehicle into the trained position prediction network and position correction network as inputs, further correct the prediction value output by the position prediction network to obtain the final prediction value of the position information of the autonomous vehicle in a weak GNSS environment or GNSS interruption.

[0066] It should be noted that the system can implement the vehicle positioning method in a weak GNSS environment in the above method embodiment, which will not be described here.

[0067] The application further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the vehicle positioning method in a weak GNSS environment according to any one of the above embodiments when executing the computer program.

[0068] In another embodiment of the application, a terminal device is provided, comprising a processor and a memory, wherein the memory is configured to store a computer program, and the computer program comprises program instructions; and the processor is configured to execute the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor is the computing core and control core of the terminal, and is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement a corresponding method flow or a corresponding function. The processor in the embodiments of the application can be used for the operation of the vehicle positioning method in a weak GNSS environment.

[0069] The application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the vehicle positioning method in a weak GNSS environment according to any one of the above embodiments.

[0070] In still another embodiment of the present application, the present application also provides a storage medium, specifically a computer readable storage medium (Memory), which is a memory device in the terminal device, used for storing programs and data. It can be understood that the computer readable storage medium herein can include the built-in storage medium in the terminal device, and of course can also include the expansion storage medium supported by the terminal device. The computer readable storage medium provides a storage space, which stores the operating system of the terminal. Moreover, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory. The one or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to realize the corresponding steps of the vehicle positioning method in the weak GNSS environment in the above-mentioned embodiments.

[0071] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0072] The present application is described with reference to flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a device that implements the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0073] These computer program instructions can also be stored in a computer readable memory capable of directing the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer readable memory produce a product including instruction devices, which implement the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0074] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks.

[0075] Finally, it should be noted that the above examples are merely used to illustrate the technical solutions of the present application, rather than to limit the present application. Although the present application has been described in detail with reference to the above examples, those skilled in the art should understand that the specific implementation of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A vehicle positioning method in a weak GNSS environment, characterized in that: include: Step S1: Preprocessing system operation data information of the autonomous driving vehicle in a normal driving state, wherein the system operation data information includes latitude and longitude data obtained from the GNSS system and three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system; Step S2: training a position prediction network using the pre-processed three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system; using the three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system as input to the trained prediction network, and outputting supervision information of the prediction network consisting of the difference between the GNSS system measurement value and the INS system measurement value; Step S3: training a position correction network based on the supervision information in step S2 and the pre-processed three-axis acceleration, three-axis angular velocity, heading information obtained from the INS system and the supervision information of the position correction network; similarly using the three-axis acceleration, three-axis angular velocity and heading information obtained by the INS system as inputs to the trained position correction network, outputting the error between the position prediction value and the measured value of the GNSS system, and correcting the predicted value of the position prediction network; Step S4: In a weak GNSS environment or GNSS outage, the three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system and the first-order difference of the attitude information of the autonomous vehicle are simultaneously used as inputs to the trained position prediction network and the position correction network, and the predicted value output by the position prediction network is further corrected to obtain the final predicted value of the autonomous vehicle position information; Preprocessing of system operation data information of an autonomous driving vehicle under normal driving conditions includes: Using linear interpolation to complete all erroneous or incomplete fields in the system operation data information; The supervision information of the prediction network is the difference between the position measurement value of the GNSS system and the position measurement value of the INS system in the same coordinate system.

2. The vehicle positioning method in a weak GNSS environment according to claim 1, characterized in that: The supervision information of the position correction network is the difference between the position prediction value of the GNSS system and the position prediction value of the position prediction network in the same coordinate system.

3. The vehicle positioning method in a weak GNSS environment according to claim 2, characterized in that: The position prediction value of the position prediction network is the sum of the supervision information of the prediction network and the three-axis acceleration, three-axis angular velocity and heading information obtained by the corrected INS system.

4. The vehicle positioning method in a weak GNSS environment according to claim 1, characterized in that: The number of hidden layers of the LSTM network of the position prediction network is set to 2, the network input layer has 7 neurons, the length of the input time series data is 30, and the output layer contains three neurons.

5. The vehicle positioning method in a weak GNSS environment according to claim 1, characterized in that: The number of hidden layers of the LSTM network of the position correction network is set to 2, and the output layer contains three neurons.

6. A vehicle positioning system in a weak GNSS environment, characterized by: include: A data acquisition module, configured to pre-process system operation data information of the autonomous driving vehicle under normal driving conditions, wherein the system operation data information includes latitude and longitude data obtained from the GNSS system and three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system; a position prediction network module, configured to train a position prediction network using the preprocessed three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system; use the three-axis acceleration, three-axis angular velocity, and heading information obtained from the INS system as input to the trained prediction network, and output supervision information of the prediction network consisting of the difference between the GNSS system measurement value and the INS system measurement value; a position correction network module, configured to train a position correction network based on the supervision information in step S2 and the pre-processed three-axis acceleration, three-axis angular velocity, heading information obtained from the INS system, and the supervision information of the position correction network; similarly using the three-axis acceleration, three-axis angular velocity, and heading information obtained by the INS system as input to the trained position correction network, outputting the error between the position prediction value and the measured value of the GNSS system, and correcting the predicted value of the position prediction network; An output module is used to use the three-axis acceleration, three-axis angular velocity and heading information obtained from the INS system and the first-order difference of the attitude information of the autonomous driving vehicle as inputs to the trained position prediction network and the position correction network in a weak GNSS environment or GNSS interruption, and further correct the predicted value output by the position prediction network to obtain the final predicted value of the autonomous driving vehicle position information.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the vehicle positioning method in a weak GNSS environment according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the vehicle positioning method in a weak GNSS environment according to any one of claims 1 to 5 are implemented.

Images