Vehicle positioning method and device, vehicle and storage medium

Abstract

The invention relates to the technical field of vehicles, in particular to a vehicle positioning method and device, a vehicle and a storage medium, and the method comprises the steps that positioning information of one or more reference targets in a current state and a previous state is acquired, and the positioning information comprises initial transverse and longitudinal distances between the vehicle and the reference targets; the positioning information of the current state and the positioning information of the previous state are input into corresponding target error models, the target error models output positioning errors of the reference target, and the positioning errors are transverse and longitudinal errors and error change rates between the current state and the previous state; and determining an optimal reference target in the plurality of reference targets according to the positioning error, and determining the relative position between the vehicle and the surrounding targets according to the positioning information of the optimal reference target. Therefore, the problems of relatively low stability and accuracy, relatively poor applicability and the like of sensor detection positioning information in related technologies are solved.

Description

Technical Field

[0001] This application relates to the technical field of vehicles, and particularly relates to a vehicle positioning method, device, vehicle, and storage medium. Background Art

[0002] During the operation of an autonomous vehicle or mobile equipment, in order to ensure accurate tracking of the planned path, it is necessary to strictly guarantee the real-time positioning accuracy in both the transverse and longitudinal directions.

[0003] In related technologies, the common practice is to use more sensors, combine the odometer characteristics of the sensors, detect the vehicle and the environment, and determine the positional relationship of the vehicle or its surroundings through multi-sensor fusion.

[0004] Since the distance to the vehicle is relatively close and each sensor can detect well, it is easier to achieve accurate confirmation in the longitudinal direction. However, as the longitudinal position increases with the distance, the detection ability of the sensors decreases, ultimately resulting in a decrease in the stability and accuracy of the positioning information, and poor applicability. Summary of the Invention

[0005] This application provides a vehicle positioning method, device, vehicle, and storage medium to solve the problems of low stability and accuracy of sensor detection and positioning information, and poor applicability in related technologies.

[0006] In a first aspect embodiment of this application, a vehicle positioning method is provided, including the following steps: obtaining the positioning information of one or more reference targets in the current state and the previous state, where the positioning information includes the initial transverse and longitudinal distances between the vehicle and the reference target; inputting the positioning information of the current state and the previous state into the corresponding target error model, and the target error model outputs the positioning error of the reference target, where the positioning error is the transverse and longitudinal error and the error change rate between the current state and the previous state; determining the best reference target among the multiple reference targets according to the positioning error, and determining the relative position between the vehicle and the surrounding targets according to the positioning information of the best reference target.

[0007] Optionally, the target error model includes an error model of the vehicle pose, an error model of the target type, and an error model of the scene. Among them, the error model of the vehicle pose calculates the error and the error change rate based on the positioning information determined by the vehicle's own motion data, the error model of the target type calculates the error and the error change rate based on the positioning information determined by the geodetic coordinates, and the error model of the scene calculates the error and the error change rate based on the positioning information determined by the sensor data.

[0008] Optionally, determining the optimal reference target among the multiple reference targets according to the positioning error includes: determining the confidence level of each reference target according to the positioning error, where the confidence level represents the accuracy of the positioning information of the reference target; using the reference target with the highest confidence level as the optimal reference target.

[0009] Optionally, before using the reference target with the highest confidence level as the optimal reference target, it further includes: if the highest confidence level is less than a preset value, obtaining the cost value of each reference target; correcting the confidence level of each reference target according to the cost value.

[0010] Optionally, obtaining the cost value of each reference target includes: calculating the variance between the current state and the initial lateral distance of the previous state; determining the cost value of each reference target according to the variance.

[0011] Optionally, determining the relative position between the vehicle and the surrounding targets according to the positioning information of the optimal reference target includes: calculating the optimal longitudinal distance according to the initial lateral distance of the optimal reference target; determining the relative position between the vehicle and the surrounding targets according to the initial lateral distance and the optimal longitudinal distance of the optimal reference target.

[0012] An embodiment of the second aspect of the present application provides a positioning device for a vehicle, including: a positioning device for a vehicle, characterized by including: an acquisition module, configured to acquire the positioning information of one or more reference targets in the current state and the previous state, where the positioning information includes the initial lateral and longitudinal distances between the vehicle and the reference target; an input module, configured to input the positioning information of the current state and the previous state into a corresponding target error model, and the target error model outputs the positioning error of the reference target, where the positioning error is the lateral and longitudinal error and the error change rate between the current state and the previous state; a determination module, configured to determine the optimal reference target among the multiple reference targets according to the positioning error, and determine the relative position between the vehicle and the surrounding targets according to the positioning information of the optimal reference target.

[0013] Optionally, the target error model includes an error model of the vehicle pose, an error model of the target type, and an error model of the scene. Among them, the error model of the vehicle pose calculates the error and the error change rate based on the positioning information determined by the vehicle's own motion data, the error model of the target type calculates the error and the error change rate based on the positioning information determined by the geodetic coordinates, and the error model of the scene calculates the error and the error change rate based on the positioning information determined by the sensor data.

[0014] Optionally, the determining module is further configured to: determine the confidence level of each reference target according to the positioning error, where the confidence level represents the accuracy of the positioning information of the reference target; and use the reference target with the highest confidence level as the optimal reference target.

[0015] Optionally, the determining module is further configured to: if the highest confidence level is less than a preset value, obtain the cost value of each reference target; and correct the confidence level of each reference target according to the cost value.

[0016] Optionally, the determining module is further configured to: calculate the variance between the current state and the initial lateral distance of the previous state; and determine the cost value of each reference target according to the variance.

[0017] Optionally, the determining module is further configured to: calculate the optimal longitudinal distance according to the initial lateral distance of the optimal reference target; and determine the relative position between the vehicle and the surrounding targets according to the initial lateral distance and the optimal longitudinal distance of the optimal reference target.

[0018] An embodiment of the third aspect of the present application provides a vehicle, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the program to implement the vehicle positioning method as described in the above embodiments.

[0019] An embodiment of the third aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor to implement the vehicle positioning method as described in the above embodiments.

[0020] Therefore, the present application has at least the following beneficial effects:

[0021] In the embodiment of the present application, the current state of one or more reference targets and the positioning information of the previous state are input into the target error model to output the lateral and longitudinal errors and the error change rate, that is, the positioning error, between the current state and the previous state of the reference target. Based on the positioning error, the optimal reference target among multiple reference targets is determined, and the relative position between the vehicle and the surrounding targets is deduced according to the positioning information of the optimal reference target. Without additional detection means, the accuracy and stability of positioning can be improved, more real-time positioning scenarios can be covered, and the applicability is higher.

[0022] The additional aspects and advantages of the present application will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present application. Description of the Drawings

[0023] The above and / or additional aspects and advantages of the present application will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, where:

[0024] Figure 1 It is a flowchart of a vehicle positioning method provided according to an embodiment of the present application;

[0025] Figure 2 It is an example diagram of a vehicle positioning device according to an embodiment of the present application;

[0026] Figure 3 It is a schematic structural diagram of a vehicle according to an embodiment of the present application. Detailed implementation manners

[0027] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application, and should not be construed as limiting the present application.

[0028] During the process of autonomous driving, for the lateral position of a vehicle or mobile equipment, since it is relatively close to the vehicle and each sensor can detect well, it is easier to accurately confirm the lateral position than the longitudinal position. As the distance increases in the longitudinal direction, the detection ability of the sensor decreases, ultimately resulting in a decrease in the certainty and accuracy of longitudinal long-distance positioning. Therefore, navigation has to rely on high-precision positioning devices or other inertial auxiliary positioning; for example, in the case of optical star positioning, the positioning of the target star generally uses the position of the remote sensing satellite as a reference, and estimates the distance and position of the target star through the lateral and longitudinal positioning accuracy and the conversion of the earth coordinates.

[0029] During the operation of an autonomous vehicle or mobile equipment, in order to ensure accurate tracking of the planned path, it is necessary to strictly ensure the real-time positioning accuracy in both the horizontal and vertical directions. The usual approach is to use more sensors, combine the odometer characteristics of the sensors, detect the vehicle and the environment, and determine the position relationship of the vehicle or its surroundings through multi-sensor fusion. Generally speaking, the higher the accuracy of a single sensor and the more sensors are fused, the higher the accuracy of the sensor, and the higher the final positioning accuracy obtained. In these applications, a decoupled accuracy processing method for the horizontal and vertical directions is usually adopted; it is impossible to achieve accurate longitudinal positioning of obstacles under different working conditions; the accurate longitudinal positioning of the vehicle itself in scenarios with multiple lanes and large vehicle flows, as well as the rapid positioning and lane change in the high-speed diversion area, resulting in low applicability.

[0030] The vehicle positioning method, device, vehicle and storage medium of the embodiments of the present application will be described below with reference to the accompanying drawings. Specifically, Figure 1 It is a schematic flowchart of a vehicle positioning method provided by an embodiment of the present application.

[0031] As Figure 1As shown, the vehicle positioning method includes the following steps:

[0032] In step S101, obtain the positioning information of one or more reference targets in the current state and the previous state, where the positioning information includes the initial lateral and longitudinal distances between the vehicle and the reference targets.

[0033] Among them, the reference target can be a lane facility, or a more complex scene semantics, or a self-vehicle positioning reference, without specific limitation.

[0034] It can be understood that in the embodiments of the present application, the initial lateral and longitudinal distances between the vehicle and one or more reference targets in the current state and the previous state can be obtained through the vehicle's own positioning sensors or cameras, etc., so as to determine the best reference target from the positioning information.

[0035] In step S102, input the positioning information of the current state and the previous state into the corresponding target error model, and the target error model outputs the positioning error of the reference target, where the positioning error is the lateral and longitudinal error and the error change rate between the current state and the previous state.

[0036] Among them, the target error model includes an error model of the vehicle pose, an error model of the target type, and an error model of the scene. Among them, the error model of the vehicle pose calculates the error and the error change rate based on the positioning information determined by the vehicle's own motion data, the error model of the target type calculates the error and the error change rate based on the positioning information determined by the geodetic coordinates, and the error model of the scene calculates the error and the error change rate based on the positioning information determined by the sensor data.

[0037] It can be understood that since there is one or more reference targets during the vehicle's driving process in the present application, the positioning information of each reference target can be input into the corresponding target error model to obtain the output error, so as to subsequently determine the best reference target among multiple reference targets according to the positioning error.

[0038] Since different error models process different data, the embodiments of the present application classify the collected data and respectively belong to the spaces of three types of error models, namely, the error model of the vehicle pose, the error model of the target type, and the error model of the scene, so as to perform data preparation and filtering before reconstruction. For example, through measurement systems such as in-vehicle sensors, geodetic coordinates, and other out-of-vehicle facilities, the position and state data of the vehicle and the environmental reference are collected. Among them, the measurement systems include but are not limited to distance sensors, vehicle speed sensors, cameras, high-precision maps, vehicle pose sensors, and angle sensors, and the collected data includes but is not limited to position coordinates, size, motion speed, and orientation, or the positioning information at the current moment is obtained through other out-of-vehicle facilities.

[0039] Specifically, according to the surrounding scenarios and operating conditions in which the vehicle operates, three error models are established respectively in accordance with the error model rules and theoretical mathematical models: the vehicle pose error model, the error model of the vehicle pose, the error model of the target type, and the error model of the scenario. The details are as follows:

[0040] (1) The error model of the vehicle pose is based on the positioning information determined by the vehicle's own motion, and forms the error and error change rate between the positioning information of the previous state. It depends on time transfer or is determined with the help of measuring devices. The specific process of constructing the error model of the vehicle pose is as follows:

[0041] Based on the vehicle dynamics and kinematics equations, establish the dynamic equations of the error and error change rate between the positioning information in the longitudinal and lateral directions of the vehicle and that of the previous state. At the same time, set the input verification rules and the confidence threshold list for the input of vehicle attitude parameter indicators. The input verification rules and the confidence threshold list can determine the situations where the analysis results of each model are not ideal. Use the static and dynamic true value data sequences of the vehicle to initialize and verify the model and the verification rules. Among them, the dynamic true value data such as longitudinal and lateral positions, heading angle, steering angle, yaw angular velocity, etc. When the model output result has a large difference from the actual data, that is, the model analysis result is not ideal, then input the more accurate lateral correction into the error model for correction. For example, input the information of the local high-precision map or V2X roadside facilities into the error model for correction.

[0042] (2) The error model of the target type is similar to the error and error change rate between the positioning information of different reference targets determined based on different target types such as geodetic coordinates, infrastructure, and marking signs and the positioning information of the previous state. Among them, the error distribution refers to the position and attitude error fluctuation range of the dynamic target object, and its error central value follows the kinematic law. For the static target object, the error central value is 0. The specific process of constructing the error model of the target type is as follows:

[0043] Based on the kinematic equations, establish the dynamic equations of the error and error change rate of the position and attitude of the target object in the longitudinal and lateral directions. At the same time, set the input verification rules and the confidence threshold list for the input of the target object attitude parameter indicators. The input verification rules and the confidence threshold list can determine the situations where the analysis results of each model are not ideal. Use the static and dynamic true value data sequences of the target object to initialize and verify the model and the verification rules. Among them, the dynamic true value data such as longitudinal and lateral positions, orientation angle, speed, etc. When the model output result has a large difference from the actual data, that is, the model analysis result is not ideal, then input the more accurate lateral correction data into the error model for correction. For example, use the information of the local high-precision map or V2X roadside facilities to input into the error model for correction.

[0044] (3) The error model of the scene is based on the error and error change rate between the relative deviation of the dynamic and static state caused by different interference factors of the sensor and the positioning information of the posture based on the positioning information of the previous state, where the relative deviation and posture refer to the positioning parameters detected by the sensor and their range of variation. The construction process of the scene error model is as follows:

[0045] Based on the kinematic equation, the error and error change rate dynamic equation of the position and attitude of the sensor in the lateral and longitudinal directions relative to the surrounding facilities are established, and the input verification rules and confidence threshold list of the vehicle attitude parameter index input are set. The input verification rules and confidence threshold list can determine the situation where the analysis results of each model are not ideal, and the model and verification rules are initialized and verified using the static true value data sequence of the sensor, where the sensor is static relative to the earth, and the true value data is such as the current absolute position, direction angle, speed, acceleration, etc. When the model output result is greatly different from the actual data, that is, the model analysis result is not ideal, a more accurate lateral correction is input into the error model for correction, such as inputting the information of local high-precision maps, high-precision grade sensors, local slam mapping or V2X roadside facilities into the error model for correction.

[0046] In summary, the input of the vehicle posture error model and the target type error model focuses on data, such as the positioning information statistics of lane signs, standard targets, and basic facilities. The input of the scene error model focuses more on the abstract and semantic information based on sensor signal disturbance errors and continuous contextual environment positioning information, so as to make more accurate and stable error inferences.

[0047] In step S103, the best reference target among multiple reference targets is determined according to the positioning error, and the relative position between the vehicle and the surrounding targets is determined according to the positioning information of the best reference target.

[0048] It can be understood that the embodiment of the present application determines the best reference target among multiple reference targets based on the positioning error, and infers the relative position between the vehicle and the surrounding targets based on the positioning information of the best reference target. Without additional detection means, the accuracy and stability of positioning can be improved, and more real-time positioning scenarios can be covered, with higher applicability.

[0049] In an embodiment of the present application, determining the best reference target among multiple reference targets based on the positioning error includes: determining the confidence of each reference target based on the positioning error; and taking the reference target with the highest confidence as the best reference target.

[0050] It can be understood that, in the embodiments of the present application, the confidence level of each reference target is determined based on the positioning error, and the reference target with the highest confidence level is used as the best reference target to improve the accuracy of the positioning information. Among them, the confidence level represents the accuracy degree of the positioning information of the reference target. For example, the larger the positioning error is, the smaller the corresponding confidence level is, and the corresponding positioning information is inaccurate. If the positioning error is smaller, the corresponding confidence level is higher, and the corresponding positioning information is accurate.

[0051] In the embodiments of the present application, before using the reference target with the highest confidence level as the best reference target, it further includes: if the highest confidence level is less than a preset value, calculate the cost value of each reference target according to the initial lateral distance; correct the confidence level of each reference target according to the cost value.

[0052] Among them, the preset value can be set or calibrated according to the actual situation, and is used to limit whether the positioning information is available. If the highest confidence level is less than the preset value, it means that the positioning errors output by all error models are relatively large, that is, all are untrustworthy; the cost value represents the credibility of the initial lateral distance. The higher the cost value is, the higher the corresponding credibility is, that is, the more accurate the lateral positioning of the reference target is. On the contrary, the lower the cost value is, the lower the corresponding credibility is, that is, the less accurate the lateral positioning of the reference target is.

[0053] It can be understood that, since the present application mainly uses the lateral positioning of the reference target to realize the estimation of the longitudinal positioning after determining the best reference target, therefore, when the positioning errors of all error models are relatively large, the reference target with relatively accurate lateral positioning can be determined according to the cost value, and the best reference target can be determined according to the lateral positioning. In specific implementation, the confidence level is corrected through the cost value, and the best reference target is selected through the corrected confidence level, effectively improving the applicability of the solution and the accuracy of the selection of the best reference target.

[0054] It should be noted that, in the embodiments of the present application, the cost value of each reference target can be calculated according to the initial lateral distance in various ways. Therefore, the cost function for calculating the cost value can be specifically selected according to actual needs. For example, the cost function can be the least squares method, etc., without specific limitation. A possible implementation manner will be described below.

[0055] As a possible implementation manner, calculating the cost value of each reference target according to the initial lateral distance includes: calculating the variance between the initial lateral distance of the current state and that of the previous state; determining the cost value of each reference target according to the variance.

[0056] It can be understood that, in the embodiments of the present application, the cost value of each reference target can be determined by calculating the variance of the initial lateral distance between the current state and the previous state, so as to correct the confidence level of each reference target and improve the accuracy of the positioning information.

[0057] In the embodiment of the present application, determining the relative position between the vehicle and surrounding objects according to the positioning information of the best reference object includes: calculating the best longitudinal distance according to the initial lateral distance of the best reference object; determining the relative position between the vehicle and surrounding objects according to the initial lateral distance and the best longitudinal distance of the best reference object.

[0058] It can be understood that in the embodiment of the present application, the best longitudinal distance can be calculated according to the initial lateral distance of the best reference object, and the relative position between the vehicle and surrounding objects can be determined based on the initial lateral distance and the best longitudinal distance of the best reference object, improving the accuracy and stability of the positioning information.

[0059] Specifically, for the accurate longitudinal positioning obtained in this cycle, it can be used as a position reference to calculate the relative positions of various target objects of interest during the vehicle's driving. For example, the relative position can be calculated through the fusion positioning of measured positioning and inferred positioning, such as the longitudinal positioning of target objects and the vehicle itself in scenarios such as automatic lane change, highway diversion area, and multi-vehicle driving.

[0060] According to the vehicle positioning method proposed in the embodiment of the present application, a dynamic inference method and a construction system for horizontal and vertical accurate positioning based on an error model can be used. From data-driven and real-time scenario reference, using state space, constructing an error model, and an optimization strategy, through the correlation and corresponding compensation of horizontal and vertical positioning errors, accurate inference of the longitudinal positioning during vehicle operation can be performed. On the one hand, the relative relationship between the vehicle and scenario elements is fully utilized, and the perception and surrounding data information are fully explored without adding additional detection means; on the other hand, based on the original field data and multiple reference objects, applying multiple error model combinations and real-time compensation strategies can cover more real-time positioning scenarios.

[0061] Next, a vehicle positioning device according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0062] Figure 2 It is a block diagram of the vehicle positioning device according to the embodiment of the present application.

[0063] As Figure 2 shown, the vehicle positioning device 10 includes: an acquisition module 100, an input module 200, and a determination module 300.

[0064] Among them, the acquisition module 100 is used to acquire the positioning information of one or more reference targets in the current state and the previous state. The positioning information includes the initial lateral and longitudinal distances between the vehicle and the reference target. The input module 200 is used to input the positioning information of the current state and the previous state into the corresponding target error model. The target error model outputs the positioning error of the reference target. The positioning error is the lateral and longitudinal error and the error change rate between the current state and the previous state. The determination module 300 is used to determine the best reference target among the multiple reference targets according to the positioning error, and determine the relative position between the vehicle and the surrounding targets according to the positioning information of the best reference target.

[0065] In the embodiment of the present application, the target error model includes an error model of the vehicle pose, an error model of the target type, and an error model of the scene. Among them, the error model of the vehicle pose calculates the error and the error change rate based on the positioning information determined by the vehicle's own motion data. The error model of the target type calculates the error and the error change rate based on the positioning information determined by the geodetic coordinates. The error model of the scene calculates the error and the error change rate based on the positioning information determined by the sensor data.

[0066] In the embodiment of the present application, the determination module 300 is further used to: determine the confidence level of each reference target according to the positioning error. The confidence level represents the accuracy of the positioning information of the reference target. The reference target with the highest confidence level is used as the best reference target.

[0067] In the embodiment of the present application, the determination module 300 is further used to: if the highest confidence level is less than the preset value, obtain the cost value of each reference target. The confidence level of each reference target is corrected according to the cost value.

[0068] In the embodiment of the present application, the determination module 300 is further used to: calculate the variance between the initial lateral distances of the current state and the previous state. The cost value of each reference target is determined according to the variance.

[0069] In the embodiment of the present application, the determination module 300 is further used to: calculate the best longitudinal distance according to the initial lateral distance of the best reference target. The relative position between the vehicle and the surrounding targets is determined according to the initial lateral distance and the best longitudinal distance of the best reference target.

[0070] It should be noted that the foregoing explanation of the embodiment of the vehicle positioning method also applies to the vehicle positioning device of this embodiment, and will not be repeated here.

[0071] According to the positioning device of a vehicle proposed in an embodiment of the present application, the current state of one or more reference targets and the positioning information of the previous state are input into a target error model to output the lateral and longitudinal errors and the error change rate, that is, the positioning error, between the current state and the previous state of the reference target. Based on the positioning error, the best reference target among multiple reference targets is determined, and the relative position between the vehicle and surrounding targets is deduced according to the positioning information of the best reference target. Without additional detection means, the accuracy and stability of positioning can be improved, more real-time positioning scenarios can be covered, and the applicability is higher.

[0072] Figure 3 The structural schematic diagram of the vehicle provided by the embodiment of the present application. The vehicle may include:

[0073] A memory 301, a processor 302, and a computer program stored on the memory 301 and executable on the processor 302.

[0074] When the processor 302 executes the program, it implements the vehicle positioning method provided in the above embodiment.

[0075] Further, the vehicle further includes:

[0076] A communication interface 303 for communication between the memory 301 and the processor 302.

[0077] The memory 301 is used to store a computer program executable on the processor 302.

[0078] The memory 301 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0079] If the memory 301, the processor 302, and the communication interface 303 are implemented independently, the communication interface 303, the memory 301, and the processor 302 can be interconnected through a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 3 only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0080] Optionally, in a specific implementation, if the memory 301, the processor 302, and the communication interface 303 are integrated on a single chip, the memory 301, the processor 302, and the communication interface 303 can communicate with each other through an internal interface.

[0081] The processor 302 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0082] The embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor to implement the vehicle positioning method as described in the above embodiments.

[0083] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples.

[0084] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0085] Any process or method description shown in a flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or N executable instructions for implementing a customized logic function or process, and the scope of the preferred embodiments of the present application includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art of the embodiments of the present application.

[0086] It should be understood that each part of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by a combination of any one or more of the following techniques well known in the art: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays, field programmable gate arrays, and the like.

[0087] Those of ordinary skill in the art can understand that all or part of the steps carried by the methods in the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0088] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A positioning method for a vehicle, characterized in that, It includes the following steps: Obtain the positioning information of one or more reference targets in the current state and the previous state, where the positioning information includes the initial lateral and longitudinal distances between the vehicle and the reference target; Input the positioning information of the current state and the previous state into the corresponding target error model, and the target error model outputs the positioning error of the reference target, where the positioning error is the lateral and longitudinal error and the error change rate between the current state and the previous state; Determine the best reference target among the multiple reference targets according to the positioning error, and determine the relative position between the vehicle and the surrounding targets according to the positioning information of the best reference target.

2. The positioning method for a vehicle according to claim 1, characterized in that, The target error model includes an error model of the vehicle pose, an error model of the target type, and an error model of the scene. Among them, the error model of the vehicle pose calculates the error and the error change rate based on the positioning information determined by the vehicle's own motion data, the error model of the target type calculates the error and the error change rate based on the positioning information determined by the geodetic coordinates, and the error model of the scene calculates the error and the error change rate based on the positioning information determined by the sensor data.

3. The positioning method for a vehicle according to claim 1, characterized in that, The determining the best reference target among the multiple reference targets according to the positioning error includes: Determine the confidence level of each reference target according to the positioning error, where the confidence level represents the accuracy of the positioning information of the reference target; Take the reference target with the highest confidence level as the best reference target.

4. The positioning method for a vehicle according to claim 3, characterized in that, Before taking the reference target with the highest confidence level as the best reference target, it further includes: If the highest confidence level is less than the preset value, calculate the cost value of each reference target according to the initial lateral distance, where the cost value represents the credibility of the initial lateral distance; Correct the confidence level of each reference target according to the cost value.

5. The positioning method for a vehicle according to claim 4, characterized in that, The calculating the cost value of each reference target according to the initial lateral distance includes: Calculate the variance between the initial lateral distances of the current state and the previous state; Determine the cost value of each reference target according to the variance.

6. The positioning method for a vehicle according to claim 1, characterized in that, The determining the relative position between the vehicle and the surrounding targets according to the positioning information of the best reference target includes: Calculate the best longitudinal distance according to the initial lateral distance of the best reference target; Determine the relative position between the vehicle and the surrounding targets according to the initial lateral distance and the best longitudinal distance of the best reference target.

7. A positioning device for a vehicle, characterized in that, It includes: An acquisition module for obtaining the positioning information of one or more reference targets in the current state and the previous state, where the positioning information includes the initial lateral and longitudinal distances between the vehicle and the reference target; An input module for inputting the positioning information of the current state and the previous state into the corresponding target error model, and the target error model outputs the positioning error of the reference target, where the positioning error is the lateral and longitudinal error and the error change rate between the current state and the previous state; A determination module for determining the best reference target among the multiple reference targets according to the positioning error, and determining the relative position between the vehicle and the surrounding targets according to the positioning information of the best reference target.

8. The positioning device for a vehicle according to claim 7, characterized in that, The target error model includes an error model of vehicle pose, an error model of target type, and an error model of the scene. Among them, the error model of vehicle pose calculates the error and error change rate based on the positioning information determined by the vehicle's own motion data; the error model of target type calculates the error and error change rate based on the positioning information determined by the geodetic coordinates; and the error model of the scene calculates the error and error change rate based on the positioning information determined by the sensor data.

9. A vehicle, characterized in that, It includes: A memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the program to implement the vehicle positioning method according to any one of claims 1-6.

10. A computer-readable storage medium, on which a computer program is stored, characterized in that, The program is executed by the processor to implement the vehicle positioning method according to any one of claims 1-6.

Images