Autopilot Vehicle Positioning Method, Device, Electronic Device, and Storage Medium

Abstract

The present application discloses a positioning method, device, electronic device, and storage medium for an autonomous driving vehicle. The method includes: obtaining the visual lateral correction information and heading angle of the autonomous driving vehicle at the previous moment; determining the turning state of the autonomous driving vehicle according to the visual lateral correction information and heading angle at the previous moment; determining the visual lateral correction information in the turning state according to the visual lateral correction information and heading angle at the previous moment, and the turning state; using the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment; and using the corrected fusion positioning information as measurement information for fusion positioning to obtain the final fusion positioning result. The present application takes into account the problem that accurate visual lateral correction information cannot be stably obtained in scenarios such as turning. By analyzing the turning state of the vehicle, the visual lateral correction information of the vehicle in different turning states is determined and used for correction, thereby improving the positioning accuracy and stability.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular, to a positioning method, device, electronic device, and storage medium for autonomous driving vehicles. Background Art

[0002] In the autonomous driving scenario, high-precision positioning of autonomous driving vehicles needs to be achieved. Currently, a multi-sensor fusion positioning method is usually adopted, that is, the positioning information collected by multiple sensors is fused through a Kalman filter to achieve high-precision positioning of the vehicle. For example, a fusion positioning scheme in the prior art is a fusion positioning scheme based on IMU (Inertial Measurement Unit) and RTK (Real-time kinematic). However, in this scheme, when the autonomous driving vehicle encounters working conditions such as cities, canyons, or tunnels, RTK will be interfered with or have no signal and cannot work. Especially in the long tunnel working condition, high-precision positioning information cannot be obtained, and the autonomous driving vehicle cannot perform lane keeping due to the above reasons.

[0003] Based on this, a scheme for correcting the above fusion positioning result based on the visual lateral correction information output by the image processing subsystem of the autonomous driving vehicle is proposed in the prior art, which can improve the fusion positioning accuracy when the positioning signals such as RTK are poor and perform lane keeping.

[0004] However, the visual lateral correction information depends on the recognition effect of the deep learning model. It cannot stably output the visual lateral correction information when the vehicle turns, changes lanes, or the light is not good, and misrecognition will occur, resulting in inaccurate visual lateral correction information output, which in turn affects the fusion positioning accuracy and stability. Summary of the Invention

[0005] Embodiments of the present application provide a positioning method, device, electronic device, and storage medium for autonomous driving vehicles to improve the positioning accuracy and positioning stability of autonomous driving vehicles in scenarios such as turning.

[0006] Embodiments of the present application adopt the following technical solutions:

[0007] In a first aspect, embodiments of the present application provide a positioning method for autonomous driving vehicles, where the method includes:

[0008] Obtain the visual lateral correction information and heading angle of the autonomous driving vehicle at the previous moment;

[0009] Determine the turning state of the autonomous driving vehicle according to the visual lateral correction information and heading angle at the previous moment;

[0010] Determine the visual lateral correction information in the turning state based on the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous vehicle;

[0011] Use the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information;

[0012] Use the corrected fusion positioning information as measurement information for fusion positioning to obtain the final fusion positioning result of the autonomous vehicle.

[0013] Optionally, the visual lateral correction information at the previous moment includes the position offset in the navigation coordinate system, the position offset includes the x-axis position offset and the y-axis position offset, and determining the turning state of the autonomous vehicle according to the visual lateral correction information and the heading angle at the previous moment includes:

[0014] Compare the absolute value of the x-axis position offset with the absolute value of the y-axis position offset;

[0015] If the absolute value of the x-axis position offset is greater than the absolute value of the y-axis position offset, determine the turning state of the autonomous vehicle according to the x-axis position offset and the heading angle;

[0016] If the absolute value of the x-axis position offset is not greater than the absolute value of the y-axis position offset, determine the turning state of the autonomous vehicle according to the y-axis position offset and the heading angle.

[0017] Optionally, the visual lateral correction information at the previous moment includes the position offset in the navigation coordinate system, and determining the visual lateral correction information in the turning state according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous vehicle includes:

[0018] Determine the absolute value of visual lateral correction according to the position offset in the navigation coordinate system;

[0019] Determine the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle, and the turning state of the autonomous vehicle.

[0020] Optionally, after determining the absolute value of visual lateral correction, the method further includes:

[0021] Compare the absolute value of visual lateral correction with a preset absolute value threshold;

[0022] If the absolute value of the visual lateral correction is not greater than the preset absolute value threshold, determine the visual lateral correction information in the turning state according to the absolute value of the visual lateral correction, the heading angle, and the turning state of the autonomous vehicle;

[0023] If the absolute value of the visual lateral correction is greater than the preset absolute value threshold, discard the absolute value of the visual lateral correction.

[0024] Optionally, the correcting the fusion positioning information at the current moment by using the visual lateral correction information in the turning state includes:

[0025] Determine the current cumulative time, which is calculated starting from the time when the visual lateral correction information in the turning state is obtained;

[0026] If the current cumulative time is less than the first cumulative time threshold, directly correct the fusion positioning information at the current moment by using the visual lateral correction information in the turning state;

[0027] If the current cumulative time is not less than the first cumulative time threshold but less than the second cumulative time threshold, perform attenuation processing on the visual lateral correction information in the turning state, and correct the fusion positioning information at the current moment by using the attenuated visual lateral correction information;

[0028] If the current cumulative time is not less than the second cumulative time threshold, set the attenuated visual lateral correction information to zero.

[0029] Optionally, the performing attenuation processing on the visual lateral correction information in the turning state includes:

[0030] Determine an attenuation factor according to the first cumulative time threshold and the second cumulative time threshold;

[0031] Perform attenuation processing on the visual lateral correction information in the turning state according to the attenuation factor to obtain the attenuated visual lateral correction information.

[0032] Optionally, after obtaining the visual lateral correction information and the heading angle of the autonomous vehicle at the previous moment, the method further includes:

[0033] Determine whether new visual lateral correction information is obtained;

[0034] If obtained, correct the fusion positioning information at the current moment by using the new visual lateral correction information.

[0035] In a second aspect, an embodiment of the present application further provides a positioning device for an autonomous vehicle, where the device includes:

[0036] An acquisition unit, configured to acquire the visual lateral correction information and the heading angle of the autonomous vehicle at the previous moment;

[0037] A first determination unit, configured to determine the turning state of the autonomous vehicle according to the visual lateral correction information and the heading angle at the previous moment;

[0038] A second determination unit, configured to determine the visual lateral correction information in the turning state according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous vehicle;

[0039] A first correction unit, configured to correct the fused positioning information at the current moment by using the visual lateral correction information in the turning state to obtain the corrected fused positioning information;

[0040] A fused positioning unit, configured to perform fused positioning by using the corrected fused positioning information as measurement information to obtain the final fused positioning result of the autonomous vehicle.

[0041] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0042] A processor; and

[0043] A memory arranged to store computer-executable instructions, where the executable instructions, when executed, cause the processor to execute any one of the foregoing methods.

[0044] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device is caused to execute any one of the foregoing methods.

[0045] The above at least one technical solution adopted in the embodiments of the present application can achieve the following beneficial effects: In the method for positioning an autonomous driving vehicle according to the embodiments of the present application, first, the visual lateral correction information and the heading angle of the autonomous driving vehicle at the previous moment are obtained; then, according to the visual lateral correction information and the heading angle at the previous moment, the turning state of the autonomous driving vehicle is determined; thereafter, according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous driving vehicle, the visual lateral correction information in the turning state is determined; then, the visual lateral correction information in the turning state is used to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information; finally, the corrected fusion positioning information is used as measurement information for fusion positioning to obtain the final fusion positioning result of the autonomous driving vehicle. The method for positioning an autonomous driving vehicle according to the embodiments of the present application takes into account the problem that accurate visual lateral correction information cannot be stably obtained in scenarios such as vehicle turning. By analyzing the turning state of the vehicle, the visual lateral correction information of the vehicle in different turning states is determined, and the fusion positioning result at the current moment is corrected based on this, improving the positioning accuracy and positioning stability of the autonomous driving vehicle in scenarios such as turning. Description of the Drawings

[0046] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0047] Figure 1 It is a schematic flowchart of a method for positioning an autonomous driving vehicle according to an embodiment of the present application;

[0048] Figure 2 It is a comparison histogram of the fusion positioning errors before and after the correction of the visual lateral correction information according to an embodiment of the present application;

[0049] Figure 3 It is a schematic structural diagram of a device for positioning an autonomous driving vehicle according to an embodiment of the present application;

[0050] Figure 4 It is a schematic structural diagram of an electronic device according to an embodiment of the present application. Detailed Embodiments

[0051] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0052] The following will describe in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.

[0053] For the convenience of understanding the embodiments of the present application, here, the fusion positioning process of an autonomous vehicle in a turning scenario is taken as an example for illustration. Of course, it should be noted that the application scenarios of the autonomous vehicle positioning method of the present application are not limited to this, and can be flexibly extended and adjusted according to actual needs.

[0054] Specifically, consider such a scenario. Suppose the vehicle is going straight on Road A, and during this driving process, visual lateral correction information can be stably obtained, and the obtained visual lateral correction information is used to correct the fusion positioning information in real time. Subsequently, the vehicle turns from Road A to Road B. During the turning process, due to the lack of lane line information, it is no longer possible to stably obtain visual lateral correction information. At this time, if the correction link of the visual lateral correction information is directly omitted, it will cause a large fluctuation in the fusion positioning result at the current moment compared with the previously corrected fusion positioning result, thereby affecting the stability and smoothness of the fusion positioning result. Moreover, due to the lack of correction of the visual lateral correction information, the accuracy of the fusion positioning is also affected.

[0055] Based on this, the embodiments of the present application provide an autonomous vehicle positioning method, as Figure 1 shown, which provides a schematic flowchart of an autonomous vehicle positioning method in the embodiments of the present application. The method at least includes the following steps S110 to step S150:

[0056] Step S110, obtain the visual lateral correction information and the heading angle of the autonomous vehicle at the previous moment.

[0057] The realization of various functions of an autonomous vehicle mainly depends on the autonomous driving system on the vehicle. The autonomous driving system includes two core subsystems, namely, the image processing subsystem and the fusion positioning subsystem. The image processing subsystem is used to collect road images in real time and perform processing and analysis, so as to provide a visual data source for other autonomous driving subsystems. The fusion positioning subsystem is an important basis for realizing the real-time performance and stability of vehicle positioning. It can perform fusion processing on the positioning data sent by other autonomous driving subsystems, so as to achieve high-precision positioning.

[0058] The autonomous driving vehicle positioning method according to the embodiments of the present application can be implemented by a fusion positioning subsystem in the autonomous driving system. When performing autonomous driving vehicle positioning, it is necessary to first obtain the visual lateral correction information sent by the image processing subsystem in the autonomous driving system. The image processing subsystem extracts and recognizes the lane line information in the image by collecting road images in real time and combining high-precision maps with visual recognition algorithms, so as to calculate the visual lateral correction information, which is used to characterize the lateral deviation of lane line recognition and is an important basis for improving fusion positioning accuracy and lane keeping. The heading angle can be obtained based on a positioning device in the inertial navigation system, such as an RTK positioning device.

[0059] In an actual application scenario, the frequency of the image processing subsystem sending the visual lateral correction information is generally different from the output frequency of the fusion positioning result, and there is a delay of about 70 ms to 100 ms in receiving the visual lateral correction information. Moreover, in scenarios such as vehicle turning, due to the lack of lane lines and other situations, the image processing subsystem cannot continuously and stably output the visual lateral correction information, resulting in that the received visual lateral correction information is not corresponding in time to the fusion positioning information of the current moment output by the fusion positioning subsystem. That is, the visual lateral correction information corresponding to the current moment of the fusion positioning information at the current moment has not been generated or has not been received yet. Therefore, the visual lateral correction information and the corresponding heading angle received at the previous moment, that is, the last time, are obtained here.

[0060] Step S120: Determine the turning state of the autonomous driving vehicle according to the visual lateral correction information and the heading angle at the previous moment.

[0061] Based on the foregoing steps, although the obtained visual lateral correction information is not the correction information corresponding to the fusion positioning information at the current moment, within a short period of time, the change in the visual lateral correction information is not too large. Therefore, the visual lateral correction information still has a certain degree of usability.

[0062] The turning state in the embodiments of the present application can include, for example, a left turn and a right turn. Since the visual lateral correction information generally needs to be first converted to a navigation coordinate system, such as the east-north-up coordinate system, before being used in subsequent correction links, different turning directions will affect the use of the visual lateral correction information in the navigation coordinate system. Therefore, here, the visual lateral correction information and the heading angle at the previous moment can be combined first to determine the current turning state of the autonomous driving vehicle, such as whether it is a left turn or a right turn.

[0063] Step S130: Determine the visual lateral correction information in the turning state according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous driving vehicle.

[0064] After determining the current turning state of the autonomous vehicle, it is necessary to use the heading angle to convert the visual lateral correction information at the previous moment to the corresponding turning state, that is, calculate the visual lateral correction information of the autonomous vehicle in the corresponding turning state. For example, if the current turning state of the autonomous vehicle is a left turn, then the visual lateral correction information at the previous moment needs to be converted to the visual lateral correction information in the left turn direction.

[0065] Step S140: Use the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information.

[0066] After obtaining the visual lateral correction information in the turning state, the visual lateral correction information in the turning state can be used to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information, so as to ensure the stability and fusion positioning accuracy of the fusion positioning result in scenarios such as turning.

[0067] Step S150: Use the corrected fusion positioning information as measurement information for fusion positioning to obtain the final fusion positioning result of the autonomous vehicle.

[0068] After obtaining the corrected fusion positioning information, in order to further improve the positioning accuracy, the corrected fusion positioning information can be used as an observation value for measurement update. Here, EKF (Extended Kalman Filter) can be used for fusion positioning to obtain the final fusion positioning result.

[0069] The autonomous vehicle positioning method in the embodiment of the present application takes into account the problem that accurate visual lateral correction information cannot be stably obtained in scenarios such as vehicle turning. By analyzing the turning state of the vehicle, the visual lateral correction information of the vehicle in different turning states is determined, and the fusion positioning result at the current moment is corrected based on this, improving the positioning accuracy and positioning stability of the autonomous vehicle in scenarios such as turning.

[0070] In an embodiment of the present application, the visual lateral correction information at the previous moment includes the position offset in the navigation coordinate system. The position offset includes the x-axis position offset and the y-axis position offset. Determining the turning state of the autonomous vehicle according to the visual lateral correction information and the heading angle at the previous moment includes: comparing the absolute value of the x-axis position offset with the absolute value of the y-axis position offset; if the absolute value of the x-axis position offset is greater than the absolute value of the y-axis position offset, determining the turning state of the autonomous vehicle according to the x-axis position offset and the heading angle; if the absolute value of the x-axis position offset is not greater than the absolute value of the y-axis position offset, determining the turning state of the autonomous vehicle according to the y-axis position offset and the heading angle.

[0071] As described above, the visual lateral correction information in the embodiment of the present application mainly refers to the position offset in the navigation coordinate system. The navigation coordinate system here can be the east-north-up coordinate system. The position offset specifically includes the x-axis position offset dx and the y-axis position offset dy. The position offsets dx and dy in the navigation coordinate system can be obtained by converting the visual lateral correction information including the position offset Δ and the heading angle yaw sent by the image processing subsystem. Specifically, it can be realized in the following way:

[0072] dx = -Δ * sin(yaw);

[0073] dy = Δ * sin(yaw).

[0074] After that, when determining the turning state of the autonomous vehicle, the absolute value of the x-axis position offset dx can be compared with the absolute value of the y-axis position offset dy first, so as to distinguish two situations. One situation is that the absolute value of the x-axis position offset dx is greater than the absolute value of the y-axis position offset dy, and the other situation is that the absolute value of the x-axis position offset dx is not greater than the absolute value of the y-axis position offset dy.

[0075] For the above two different comparison results, different processing logics can be further adopted in combination with the heading angle yaw to determine the turning state of the autonomous vehicle. Here, it can be defined that the left direction along the vehicle body is positive and pn_flag = 1, the right direction along the vehicle body is negative and pn_flag = 0, and the default initial pn_flag = 0. Then, the turning state of the autonomous vehicle can be determined in the following way:

[0076]

[0077] In an embodiment of the present application, the visual lateral correction information at the previous moment includes the position offset in the navigation coordinate system. Determining the visual lateral correction information in the turning state according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous vehicle includes: determining the absolute value of visual lateral correction according to the position offset in the navigation coordinate system; determining the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle, and the turning state of the autonomous vehicle.

[0078] After determining the turning state of the autonomous vehicle, the visual lateral correction information at the previous moment can be converted into the corresponding visual lateral correction information in the turning state by using the heading angle. Specifically, first calculate the absolute value of visual lateral correction _vio_dp according to the position offset in the navigation coordinate system, including the x-axis position offset dx and the y-axis position offset dy. The following method can be specifically adopted:

[0079] _vio_dp = sqrt(dx() * dx() + dy() * dy()),

[0080] After obtaining the absolute value of visual lateral correction _vio_dp, the visual lateral correction information in the turning state can be further determined by combining the heading angle yaw and the current turning state of the autonomous vehicle. The following method can be specifically adopted:

[0081]

[0082] It can be seen that different turning directions result in different specific calculation methods for the visual lateral correction information including the x-axis position offset dx and the y-axis position offset dy. After the above logical processing, the corresponding visual lateral correction information in different turning states can be obtained, thus solving the problem that the visual lateral correction information in the navigation coordinate system cannot directly correct the fusion positioning result in the turning scenario.

[0083] In an embodiment of the present application, after determining the absolute value of visual lateral correction, the method further includes: comparing the absolute value of visual lateral correction with a preset absolute value threshold; if the absolute value of visual lateral correction is not greater than the preset absolute value threshold, determining the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle, and the turning state of the autonomous vehicle; if the absolute value of visual lateral correction is greater than the preset absolute value threshold, discarding the absolute value of visual lateral correction.

[0084] Under normal driving conditions, the absolute value of visual lateral correction _vio_dp only fluctuates within a small range. Therefore, there is a certain upper limit to the magnitude of the absolute value of visual lateral correction _vio_dp, that is, there is a corresponding absolute value threshold. If the calculated absolute value of visual lateral correction _vio_dp exceeds the preset absolute value threshold, it indicates that the position offset obtained based on the visual recognition algorithm is inaccurate and has a large error, and cannot be used for subsequent correction. Therefore, it can be directly discarded; conversely, it indicates that the error of this position offset is acceptable and can be used for subsequent correction. Through the above process, the usability of the visual lateral correction information can be preliminarily verified, avoiding the influence of visual lateral correction information with large self-errors on the fusion positioning accuracy.

[0085] In an embodiment of the present application, the correcting the fusion positioning information at the current moment by using the visual lateral correction information in the turning state includes: determining the current cumulative time, which starts to be calculated from the time when the visual lateral correction information in the turning state is obtained; if the current cumulative time is less than the first cumulative time threshold, directly correct the fusion positioning information at the current moment by using the visual lateral correction information in the turning state; if the current cumulative time is not less than the first cumulative time threshold but less than the second cumulative time threshold, perform an attenuation process on the visual lateral correction information in the turning state, and correct the fusion positioning information at the current moment by using the attenuated visual lateral correction information; if the current cumulative time is not less than the second cumulative time threshold, set the attenuated visual lateral correction information to zero.

[0086] As mentioned above, the visual lateral correction information obtained at the previous moment in the embodiment of the present application is not the correction information corresponding to the fusion positioning information at the current moment. It still has a certain usability within a short period of time, but when new visual lateral correction information cannot be obtained for a long time, the usability of the visual lateral correction information at the previous moment will decrease or even become unavailable. Therefore, through a large number of experiments, the embodiment of the present application designs a logic for determining the usability of the visual lateral correction information at the previous moment under different cumulative times.

[0087] Here, it should be noted first that the "current cumulative time" defined in the embodiment of the present application can be used to represent the duration elapsed after obtaining the visual lateral correction information in the turning state. Of course, it can also be the duration elapsed after obtaining the initial visual lateral correction information at the previous moment, because the time deviation between the two is very small.

[0088] The first scenario is that when the above-mentioned current cumulative time has not reached the first cumulative time threshold, it indicates that the time between obtaining the visual lateral correction information in the turning state and the current moment is still relatively short. Therefore, the visual lateral correction information in the turning state still has high availability. At this time, the visual lateral correction information in the turning state can be directly used to correct the fusion positioning information at the current moment.

[0089] The second scenario is that when the above-mentioned current cumulative time reaches the first cumulative time threshold but has not reached the second cumulative time threshold. At this time, it means that a considerable amount of time has passed since the time when the visual lateral correction information in the turning state was obtained until the current moment. The availability of the visual lateral correction information in the turning state has decreased due to the accumulation of time. However, in order to avoid the situation of sudden changes in the positioning trajectory caused by directly discarding the correction link of the visual lateral correction information, a certain smoothing operation can be performed on the visual lateral correction information in the turning state. For example, the visual lateral correction information in the turning state can be gradually attenuated, so as to use the attenuated visual lateral correction information to correct the fusion positioning information at the current moment. This processing method takes into account both the stability and the fusion positioning accuracy of the corrected fusion positioning result.

[0090] The third scenario is that when the above-mentioned current cumulative time reaches the second cumulative time threshold, it indicates that the time between obtaining the visual lateral correction information in the turning state and the current moment is already relatively long. After this relatively long time, the visual lateral correction information may have changed relatively greatly. Therefore, the visual lateral correction information corresponding to the previous moment in the turning state will no longer be available. At this time, the visual lateral correction information can be directly set to zero.

[0091] Both the above-mentioned first cumulative time threshold and the second cumulative time threshold are empirical values and can be flexibly adjusted according to the actual scenario and actual requirements. For the convenience of understanding the embodiments of the present application, further examples are given here. Suppose the first cumulative time threshold is 3s and the second cumulative time threshold is 6s. When the current cumulative time is less than 3s, the visual lateral correction information in the turning state can be directly used to correct the fusion positioning information at the current moment. When the current cumulative time reaches 3s but has not reached 6s, the visual lateral correction information in the turning state can be attenuated and then used to correct the fusion positioning information at the current moment. When the current cumulative time reaches 6s, the visual lateral correction information is directly set to zero.

[0092] In an embodiment of the present application, the attenuation processing of the visual lateral correction information in the turning state includes: determining an attenuation factor according to the first cumulative time threshold and the second cumulative time threshold; performing attenuation processing on the visual lateral correction information in the turning state according to the attenuation factor to obtain the attenuated visual lateral correction information.

[0093] When the embodiment of the present application performs attenuation processing on the visual lateral correction information in the turning state, the attenuation factor can be calculated first, and the attenuation factor can characterize the degree of attenuation of the visual lateral correction information each time. For example, the first cumulative time threshold is 3s, the second cumulative time threshold is 6s, and the visual lateral correction information includes the x-axis position offset dx and the y-axis position offset dy. Then the attenuation factor vio_k can be calculated in the following way:

[0094] vio_k[0] = dx() / 300.0;

[0095] vio_k[1] = dy() / 300.0;

[0096] Among them, 300.0 in the above formula is mainly determined based on the output frequency of the fusion positioning information and the difference between the first cumulative time threshold and the second cumulative time threshold, that is, the attenuation duration. For example, the output frequency of the fusion positioning information is usually 100Hz. Then, within the attenuation duration of 3s, the attenuation ratio each time is 1 / 300.

[0097] Further, the attenuated visual lateral correction information can be calculated in the following way, including the attenuated x-axis position offset dx and the attenuated y-axis position offset dy:

[0098]

[0099] Based on the above logic, the visual lateral correction information obtained after each attenuation will be used as the basis for calculating the next attenuation factor, and the attenuation factor calculated each time will be used as the basis for calculating the attenuated visual lateral correction information next time. In this way, the cycle is repeated, and the magnitude of the visual lateral correction information used for correction gradually decreases, thereby realizing the smoothing processing of the fusion positioning result.

[0100] Based on the above embodiment, the visual lateral correction information in different situations can be obtained. Finally, the fusion positioning information at the current moment, such as the position coordinates (Posx, Posy) at the current moment, can be corrected based on the following way to obtain the corrected position coordinates (Posx', Posy'):

[0101] Posx' = Posx + dx;

[0102] Posy' = Posy + dy.

[0103] In one embodiment of the present application, after obtaining the visual lateral correction information and the heading angle of the autonomous vehicle at the previous moment, the method further includes: determining whether new visual lateral correction information is obtained; if so, using the new visual lateral correction information to correct the fusion positioning information at the current moment.

[0104] The above embodiment mainly differentiates the availability of the visual lateral correction information at the previous moment under different cumulative times, and is mainly set for the situation where no new visual lateral correction information is obtained in this process. Therefore, in the actual scenario, if new visual lateral correction information is obtained during the above process, subsequent processing can be based on the new visual lateral correction information.

[0105] Here, further in combination with the foregoing embodiment, it is assumed that the first cumulative time threshold is 3 s and the second cumulative time threshold is 6 s. It can be continuously detected within 3 s whether new visual lateral correction information can be obtained. If it can be obtained, the new visual lateral correction information is directly used for subsequent processing. If it cannot be obtained, within 3 seconds, the visual lateral correction information at the previous moment can be directly used for correction. When reaching the 3rd second, the visual lateral correction information at the previous moment is attenuated. During this process, it can still be continuously detected whether new visual lateral correction information is obtained. Once it is obtained, the new visual lateral correction information is directly used for subsequent processing. When reaching the 6th second, if new visual lateral correction information is still not obtained, the visual lateral correction information is directly set to zero, that is, no correction processing is performed in this case.

[0106] To verify the positioning effect of the autonomous vehicle positioning method of the present application, as Figure 2 shown, a comparison histogram of the fusion positioning errors before and after the correction of the visual lateral correction information in the embodiment of the present application is provided. It can be seen that after the correction of the visual lateral correction information of the present application, most of the fusion positioning errors are distributed within 0.15 m, which is greatly reduced compared with the fusion positioning errors before correction, thereby improving the fusion positioning accuracy.

[0107] The embodiment of the present application further provides an autonomous vehicle positioning device 300. As Figure 3 shown, a structural schematic diagram of an autonomous vehicle positioning device in the embodiment of the present application is provided. The device 300 at least includes: an acquisition unit 310, a first determination unit 320, a second determination unit 330, a first correction unit 340, and a fusion positioning unit 350, where:

[0108] The acquisition unit 310 is configured to acquire the visual lateral correction information and the heading angle of the autonomous vehicle at the previous moment;

[0109] A first determination unit 320, configured to determine a turning state of the autonomous vehicle according to the visual lateral correction information and the heading angle at the previous moment;

[0110] A second determination unit 330, configured to determine visual lateral correction information in a turning state according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous vehicle;

[0111] A first correction unit 340, configured to correct the fusion positioning information at the current moment by using the visual lateral correction information in the turning state to obtain corrected fusion positioning information;

[0112] A fusion positioning unit 350, configured to perform fusion positioning by using the corrected fusion positioning information as measurement information to obtain a final fusion positioning result of the autonomous vehicle.

[0113] In an embodiment of the present application, the visual lateral correction information at the previous moment includes a position offset in a navigation coordinate system, the position offset includes an x-axis position offset and a y-axis position offset, and the first determination unit 320 is specifically configured to: compare an absolute value of the x-axis position offset with an absolute value of the y-axis position offset; if the absolute value of the x-axis position offset is greater than the absolute value of the y-axis position offset, determine the turning state of the autonomous vehicle according to the x-axis position offset and the heading angle; if the absolute value of the x-axis position offset is not greater than the absolute value of the y-axis position offset, determine the turning state of the autonomous vehicle according to the y-axis position offset and the heading angle.

[0114] In an embodiment of the present application, the visual lateral correction information at the previous moment includes a position offset in a navigation coordinate system, and the second determination unit 330 is specifically configured to: determine an absolute value of visual lateral correction according to the position offset in the navigation coordinate system; determine the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle, and the turning state of the autonomous vehicle.

[0115] In an embodiment of the present application, the device further includes: a comparison unit, configured to compare the absolute value of visual lateral correction with a preset absolute value threshold; the second determination unit is specifically configured to: if the absolute value of visual lateral correction is not greater than the preset absolute value threshold, determine the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle, and the turning state of the autonomous vehicle; the device further includes: a discard unit, configured to discard the absolute value of visual lateral correction if the absolute value of visual lateral correction is greater than the preset absolute value threshold.

[0116] In an embodiment of the present application, the first correction unit 340 is specifically configured to: determine the current cumulative time, which starts to be calculated from the time when the visual lateral correction information in the turning state is obtained; if the current cumulative time is less than the first cumulative time threshold, directly correct the fusion positioning information at the current moment by using the visual lateral correction information in the turning state; if the current cumulative time is not less than the first cumulative time threshold but less than the second cumulative time threshold, perform attenuation processing on the visual lateral correction information in the turning state, and correct the fusion positioning information at the current moment by using the attenuated visual lateral correction information; if the current cumulative time is not less than the second cumulative time threshold, set the attenuated visual lateral correction information to zero.

[0117] In an embodiment of the present application, the first correction unit 340 is specifically configured to: determine an attenuation factor according to the first cumulative time threshold and the second cumulative time threshold; perform attenuation processing on the visual lateral correction information in the turning state according to the attenuation factor to obtain the attenuated visual lateral correction information.

[0118] In an embodiment of the present application, the device further includes: a third determination unit, configured to determine whether new visual lateral correction information is obtained; a second correction unit, configured to, if new visual lateral correction information is obtained, correct the fusion positioning information at the current moment by using the new visual lateral correction information.

[0119] It can be understood that the above-mentioned autonomous driving vehicle positioning device can implement each step of the autonomous driving vehicle positioning method provided in the foregoing embodiments. The relevant explanations regarding the autonomous driving vehicle positioning method are applicable to the autonomous driving vehicle positioning device and will not be elaborated herein.

[0120] Figure 4 It is a schematic structural diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 4 , at the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory, such as at least one disk memory, etc. Of course, the electronic device may also include other hardware required for other services.

[0121] The processor, network interface, and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, 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 4 only a bidirectional arrow is used in

[0122] Memory, used to store programs. Specifically, the program can include program code, and the program code includes computer operation instructions. The memory can include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0123] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming an autonomous driving vehicle positioning device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0124] Obtain the visual lateral correction information and heading angle of the autonomous driving vehicle at the previous moment;

[0125] Determine the turning state of the autonomous driving vehicle according to the visual lateral correction information and heading angle at the previous moment;

[0126] Determine the visual lateral correction information in the turning state according to the visual lateral correction information and heading angle at the previous moment, and the turning state of the autonomous driving vehicle;

[0127] Use the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information;

[0128] Use the corrected fusion positioning information as measurement information for fusion positioning to obtain the final fusion positioning result of the autonomous driving vehicle.

[0129] The above as in this application Figure 1The method executed by the autonomous vehicle positioning device disclosed in the illustrated embodiment can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in the processor or the instructions in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute each method, step, and logic block diagram disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method.

[0130] The electronic device can also execute Figure 1 the method executed by the autonomous vehicle positioning device in Figure 1 the illustrated embodiment and implement the functions of the autonomous vehicle positioning device in

[0131] Embodiments of the present application also propose a computer-readable storage medium that stores one or more programs. The one or more programs include instructions that, when executed by an electronic device including a plurality of application programs, can enable the electronic device to execute Figure 1 the method executed by the autonomous vehicle positioning device in the illustrated embodiment, and specifically used to execute:

[0132] Obtain the visual lateral correction information and the heading angle of the autonomous vehicle at the previous moment;

[0133] Determine the turning state of the autonomous vehicle according to the visual lateral correction information and the heading angle at the previous moment;

[0134] Determine the visual lateral correction information in the turning state according to the visual lateral correction information and the heading angle at the previous moment, and the turning state of the autonomous vehicle;

[0135] Use the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information;

[0136] Use the corrected fusion positioning information as measurement information for fusion positioning to obtain the final fusion positioning result of the autonomous vehicle.

[0137] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0138] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. 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 the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0139] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the specified functions in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide means for implementing the specified functions in one process Figure 1 one process or multiple processes and / or blocksFigure 1 Steps of functions specified in one or more boxes.

[0141] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0142] Memory may include non-permanent memory in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0143] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transitory media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0144] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0145] 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 a complete hardware embodiment, a complete 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-ROM, optical storage, etc.) containing computer-usable program code.

[0146] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for positioning an autonomous driving vehicle, wherein, The method includes: Obtaining the visual lateral correction information and heading angle of the autonomous vehicle at the previous moment; Determining the turning state of the autonomous vehicle according to the visual lateral correction information and heading angle at the previous moment; Determining the visual lateral correction information in the turning state according to the visual lateral correction information and heading angle at the previous moment, and the turning state of the autonomous vehicle; Using the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment to obtain the corrected fusion positioning information; Taking the corrected fusion positioning information as measurement information for fusion positioning to obtain the final fusion positioning result of the autonomous vehicle.

2. The method according to claim 1, wherein, The visual lateral correction information at the previous moment includes the position offset in the navigation coordinate system, the position offset includes the x-axis position offset and the y-axis position offset, and determining the turning state of the autonomous vehicle according to the visual lateral correction information and heading angle at the previous moment includes: Comparing the absolute value of the x-axis position offset with the absolute value of the y-axis position offset; If the absolute value of the x-axis position offset is greater than the absolute value of the y-axis position offset, determining the turning state of the autonomous vehicle according to the x-axis position offset and the heading angle; If the absolute value of the x-axis position offset is not greater than the absolute value of the y-axis position offset, determining the turning state of the autonomous vehicle according to the y-axis position offset and the heading angle.

3. The method according to claim 1, wherein, The visual lateral correction information at the previous moment includes the position offset in the navigation coordinate system, and determining the visual lateral correction information in the turning state according to the visual lateral correction information and heading angle at the previous moment, and the turning state of the autonomous vehicle includes: Determining the absolute value of visual lateral correction according to the position offset in the navigation coordinate system; Determining the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle and the turning state of the autonomous vehicle.

4. The method according to claim 3, wherein, After determining the absolute value of visual lateral correction, the method further includes: Comparing the absolute value of visual lateral correction with a preset absolute value threshold; If the absolute value of visual lateral correction is not greater than the preset absolute value threshold, determining the visual lateral correction information in the turning state according to the absolute value of visual lateral correction, the heading angle and the turning state of the autonomous vehicle; If the absolute value of visual lateral correction is greater than the preset absolute value threshold, discarding the absolute value of visual lateral correction.

5. The method according to claim 1, wherein, The using the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment includes: Determining the current cumulative time, which starts to be calculated from the time when the visual lateral correction information in the turning state is obtained; If the current cumulative time is less than the first cumulative time threshold, directly using the visual lateral correction information in the turning state to correct the fusion positioning information at the current moment; If the current cumulative time is not less than the first cumulative time threshold but less than the second cumulative time threshold, attenuate the visual lateral correction information in the turning state, and correct the fusion positioning information at the current moment using the attenuated visual lateral correction information; If the current cumulative time is not less than the second cumulative time threshold, set the attenuated visual lateral correction information to zero.

6. The method according to claim 5, wherein, The attenuating the visual lateral correction information in the turning state includes: Determine an attenuation factor according to the first cumulative time threshold and the second cumulative time threshold; Attenuate the visual lateral correction information in the turning state according to the attenuation factor to obtain the attenuated visual lateral correction information.

7. The method according to claim 5, wherein, After obtaining the visual lateral correction information and the heading angle of the previous moment of the autonomous vehicle, the method further includes: Determine whether new visual lateral correction information is obtained; If obtained, correct the fusion positioning information at the current moment using the new visual lateral correction information.

8. An apparatus for positioning an autonomous driving vehicle, wherein, The apparatus includes: An acquisition unit, configured to acquire the visual lateral correction information and the heading angle of the previous moment of the autonomous vehicle; A first determination unit, configured to determine the turning state of the autonomous vehicle according to the visual lateral correction information and the heading angle of the previous moment; A second determination unit, configured to determine the visual lateral correction information in the turning state according to the visual lateral correction information and the heading angle of the previous moment, and the turning state of the autonomous vehicle; A first correction unit, configured to correct the fusion positioning information at the current moment using the visual lateral correction information in the turning state to obtain the corrected fusion positioning information; A fusion positioning unit, configured to perform fusion positioning using the corrected fusion positioning information as measurement information to obtain the final fusion positioning result of the autonomous vehicle.

9. An electronic device, comprising: A processor; And A memory arranged to store computer-executable instructions, the executable instructions, when executed, cause the processor to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, cause the electronic device to execute the method according to any one of claims 1 to 7.

Images