Fusion positioning method, device, electronic device, and storage medium

Abstract

The present application discloses a fusion positioning method, device, electronic device, and storage medium, wherein the method includes obtaining a longitudinal positioning observation value of a visual dimension under a preset positioning scenario; determining the anteroposterior relationship of the distance between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value; when the longitudinal positioning value is positive, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain the positioning result of the vehicle. Through this application, a smooth and stable output result of the vehicle trajectory in the direction of travel of the vehicle is achieved, thereby ensuring that the planning and control of the autonomous driving is not affected after the fusion positioning process.

Description

Technical Field

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

[0002] With the popularization of high-precision maps and visual sensors, more and more high-precision positioning solutions for autonomous driving have transformed from the previous IMU / GNSS combined navigation solutions to multi-sensor fusion solutions. That is, high-precision maps, vehicle body information and visual sensor inputs are added to solve special situations such as positioning trajectory drift, interruption, and jump when the GNSS signal is poor.

[0003] Multi-sensor fusion solutions in related technologies use the Kalman filter as a foundational framework, matching the semantic information output by visual sensors with the corresponding semantic information in high-precision maps to provide additional observation information to better adapt to these challenging scenarios. After receiving the additional observations returned by the visual sensors, the Kalman filter ensures a smooth positioning trajectory, eliminating the effects of jumps, interruptions, and drift.

[0004] However, when using a multi-sensor fusion solution for visual longitudinal positioning, the positioning provided by the visual longitudinal position is at the rear of the current moment, and the positioning trajectory will smooth out to the rear of the vehicle, seriously affecting the vehicle's control planning. Summary of the Invention

[0005] The embodiments of the present application provide a fusion positioning method, device, electronic device, and storage medium to correct the visual longitudinal positioning and ensure the smoothness and stability of the vehicle trajectory in the driving direction.

[0006] The embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, an embodiment of the present application provides a fusion positioning method, which is used for an autonomous driving vehicle, and the method includes: obtaining a longitudinal positioning observation value in a visual dimension under a preset positioning scenario; determining a causal relationship of the distance between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes the longitudinal positioning being a positive value or the longitudinal positioning being a negative value; when the longitudinal positioning is a positive value, inputting the longitudinal positioning observation value in the visual dimension into a preset filter for fusion positioning to obtain a positioning result for the vehicle.

[0008] In a second aspect, an embodiment of the present application further provides a fusion positioning device, which is used for an autonomous driving vehicle, and the device includes: an acquisition module for acquiring the longitudinal positioning observation value of the visual dimension under a preset positioning scenario; a determination module for determining the anteroposterior relationship of the distance between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle based on the longitudinal positioning observation value of the visual dimension, wherein the relationship includes the longitudinal positioning being a positive value or the longitudinal positioning being a negative value; a fusion positioning module for inputting the longitudinal positioning observation value of the visual dimension into a preset filter for fusion positioning when the longitudinal positioning is a positive value, thereby obtaining the positioning result of the vehicle.

[0009] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor; and a memory arranged to store computer-executable instructions, wherein the executable instructions, when executed, enable the processor to perform the above method.

[0010] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes the above method.

[0011] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0012] The system obtains longitudinal positioning observations in the visual dimension under a preset positioning scenario. Based on the distance relationship between the ego vehicle's current position and the longitudinal positioning, it adjusts the longitudinal positioning observations input to the preset filter. This ensures a smooth trajectory during the fusion positioning process. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0014] Figure 1 This is a flow chart of the fusion positioning method in an embodiment of the present application;

[0015] Figure 2 This is a schematic diagram of the structure of the fusion positioning device in the embodiment of the present application;

[0016] Figure 3 This is a structural diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0017] To make the purpose, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0018] The inventors discovered that, although filtering after receiving additional observations from the visual sensor can ensure smooth positioning trajectories and eliminate the effects of jumps, interruptions, and drift, in some cases, due to the excessively long trajectory deduction time or the influence of slowly changing abnormal GNSS signals, the vision cannot provide longitudinal positioning during this period. For example, the relevant visual sensor only provides lateral corrections. Because lane markings are present in most scenarios, longitudinal positioning factors only appear in certain areas, such as road arrows or stop lines. As a result, when visual longitudinal positioning is performed, the position provided by the vision is behind the current moment, and the positioning trajectory will smooth out to the back of the vehicle, seriously affecting the vehicle's control planning.

[0019] To address the above shortcomings, this application constrains the vehicle's driving direction, determines the relationship between the current visual positioning information and the vehicle's current position (in front of or behind the vehicle), and adds a correction bias to the accelerometer based on the distance between the visual positioning information and the vehicle to ensure that the vehicle's trajectory is smooth and stable in the vehicle's driving direction.

[0020] The method in this application divides longitudinal positioning into positive and negative longitudinal positioning based on the relationship between longitudinal positioning and ego-vehicle positioning in the direction of travel. Considering that longitudinal observations are only valid in some scenarios, this observation is not fused using the traditional fusion process like lateral positioning observations or GNSS observations, but rather uses a special fusion method.

[0021] The following describes in detail the technical solutions provided by various embodiments of the present application in conjunction with the accompanying drawings.

[0022] The present application embodiment provides a fusion positioning method, such as Figure 1 As shown, a flow chart of the fusion positioning method in an embodiment of the present application is provided, and the method at least includes the following steps S110 to S130:

[0023] Step S110: Obtaining the longitudinal positioning observation value of the visual dimension under the preset positioning scenario.

[0024] Preset positioning scenarios, including fusion positioning scenarios, include visual sensors providing lateral and longitudinal positioning. However, longitudinal positioning only occurs in certain areas, such as when the visual sensor recognizes arrows or stop signs on the road.

[0025] It is understandable that when lateral positioning is successful, the field of view of the autonomous vehicle has been limited. For longitudinal positioning, it is the distance between the vehicle and the relevant target marker.

[0026] Therefore, the vertical positioning observation value of the visual dimension will appear at a certain moment, but it will not appear continuously. It is necessary to ensure that when the vertical positioning observation value of the visual dimension is obtained, it will not affect the result of positioning fusion.

[0027] Therefore, considering that the longitudinal observation value is only valid in some scenarios, this observation is not fused using the fusion process in related technologies like the lateral positioning observation or GNSS observation, but adopts the special fusion method in this application.

[0028] Step S120 , determining a causal relationship of the distance between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value.

[0029] The vehicle's current position can be obtained through relevant sensors. The longitudinal positioning correction value in the visual dimension can be positive or negative. If it is positive, appropriate acceleration is required; if it is negative, appropriate deceleration is required. Updates are not directly rejected (vehicles typically do not drive backwards) to ensure a smooth trajectory.

[0030] For better explanation, the position of the vehicle is represented as P car , the vertical positioning observation value is expressed as P vision .

[0031] According to the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, the front-to-back relationship between the distances can be determined. The front-to-back relationship can include P vision In P car The situation in front or behind. At the same time, in P vision When the confidence level is high, the preset filter can use the longitudinal positioning observation value.

[0032] This relationship includes whether longitudinal positioning is positive or negative. Regardless of whether longitudinal positioning is positive or negative, it can be used as an observation value (positive values are input directly, and negative values are input after being corrected to positive values) and input into a preset filter for multi-sensor fusion. In other words, the semantic information output by the visual sensor is used for matching, providing additional observation information.

[0033] Step S130: When the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain the vehicle positioning result.

[0034] If the longitudinal positioning value is positive, the longitudinal positioning observation value in the visual dimension can be input into the preset filter for fusion positioning, and finally the vehicle positioning result is obtained. In other words, only when the longitudinal positioning value is positive will it be used as the longitudinal positioning observation value and input into the preset filter for fusion positioning.

[0035] It is understood that the positive or negative longitudinal positioning values are relative values determined based on the fore-and-aft relationship between the longitudinal positioning observations in the visual dimension and the vehicle's current position. Furthermore, this application adds a correction bias to the accelerometer for observations that do not meet the conditions to ensure the stability of the fused positioning results. Furthermore, this ensures the stability of the autonomous vehicle's trajectory output.

[0036] In one embodiment of the present application, it also includes: when the longitudinal positioning is a negative value, determining the time parameter required for the longitudinal positioning in the relationship between the distance between the two to change from a negative value to a positive value; determining a target acceleration parameter value based on the time parameter and the distance between the two; and adding the target acceleration parameter value to the acceleration component in the direction of travel of the vehicle when predicting the current speed of the vehicle.

[0037] In specific implementation, the fusion positioning method of this application also requires determining the time parameter required for the longitudinal positioning in the relationship between the distance and the longitudinal positioning to change from a negative value to a positive value when the longitudinal positioning is negative, and determining the target acceleration parameter value based on the time parameter and the distance between the longitudinal positioning and the longitudinal positioning. Finally, based on the target acceleration parameter value, the acceleration component in the direction of travel of the vehicle is added to the target acceleration parameter value when predicting the current vehicle speed. For example, the acceleration value a required to reach the distance S within the time t is calculated according to the following formula.

[0038]

[0039] It's important to note that the time parameter can be flexibly set based on the vehicle's current speed and location, such as at an intersection or in the middle of the road. For example, at an intersection, t should be smaller to ensure that turning maneuvers are not affected. Similarly, in the middle of the road, t should be smaller to ensure that parking or waiting operations are not affected. For another example, on a straight road with good visibility, t should be larger.

[0040] In addition, based on the GNSS signal conditions at this time t, you can choose to increase the GNSS signal confidence for updates, or not use GNSS signals for updates, thereby reducing the influence of other sensor factors during fusion positioning.

[0041] Applying this processing method directly to the IMU's acceleration data ensures smooth positioning. By manually processing the accelerometer's output, the visual longitudinal positioning can be smoothly received in the direction of the vehicle's travel, preventing the positioning trajectory from merging in the opposite direction of the vehicle's travel.

[0042] In one embodiment of the present application, the longitudinal positioning observation value of the visual dimension and the current position of the vehicle are used to determine the causal relationship of the distance between the two, wherein the relationship includes the longitudinal positioning being a positive value or the longitudinal positioning being a negative value, and then further includes: calculating the distance S between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, wherein the direction of travel of the vehicle is taken as the positive direction; when the distance S is a negative value, shortening the distance between the two according to the calculation result of the acceleration component of the target acceleration parameter value in the direction of travel of the vehicle, which is also a negative value, wherein the target acceleration parameter value is used as a correction bias in the IMU.

[0043] In specific implementation, the distance S between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle is calculated, with the vehicle's driving direction as the positive direction. If S is a negative value, fusion is not performed until S becomes a positive value.

[0044] Furthermore, in the subsequent prediction phase, when calculating the ego vehicle's velocity, the accelerometer data is specially processed. By adding the acceleration component in the ego vehicle's direction of travel to the aforementioned acceleration value a, since S is a negative value, the resulting a is also negative. This is equivalent to decelerating the ego vehicle in the direction of travel, causing the distance between the ego vehicle and this negative longitudinal position to decrease, and vice versa.

[0045] It can be understood that the target acceleration parameter value obtained through calculation is used as the correction bias in the IMU.

[0046] In one embodiment of the present application, when the longitudinal positioning is a negative value, determining the time parameter required for the longitudinal positioning in the distance relationship between the two to change from a negative value to a positive value includes: when the longitudinal positioning is a negative value, determining the time parameter based on the current vehicle speed; and / or, when the longitudinal positioning is a negative value, determining the time parameter based on the current position of the vehicle; and / or, when the longitudinal positioning is a negative value, adjusting the GNSS signal confidence or not using the GNSS signal to update the vehicle position based on the GNSS signal condition at the corresponding moment in the time parameter.

[0047] During specific implementation, the methods for determining time parameters are different for different situations.

[0048] When the longitudinal positioning is a negative value, the time parameter is determined according to the current vehicle speed. That is, when the longitudinal positioning is a negative value, the current vehicle speed determines the time parameter.

[0049] Furthermore, when the longitudinal positioning is a negative value, the time parameter is determined according to the current position of the vehicle.

[0050] Furthermore, when the longitudinal positioning is a negative value, the GNSS signal confidence is adjusted or the GNSS signal is not used to update the vehicle position according to the GNSS signal condition at the corresponding moment in the time parameter.

[0051] Specifically, one of the methods may be selected for determination based on actual usage or the confidence level of the data obtained by the sensor.

[0052] In one embodiment of the present application, when the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a vehicle positioning result, including: when the longitudinal positioning is a positive value, based on the size relationship of the distance between the two, the longitudinal positioning observation value of the visual dimension is input into the preset filter for fusion positioning; when the first longitudinal positioning of the visual dimension is a positive value but does not match the preset prediction value, the longitudinal positioning observation value of the visual dimension is not input into the preset filter for fusion positioning; when the second longitudinal positioning of the visual dimension is a positive value and matches the preset prediction value, the second longitudinal positioning is used as the target value of the preset prediction value.

[0053] In specific implementation, the position of the vehicle is expressed as P car , the vertical positioning observation value is expressed as P vision Take this as an example to illustrate.

[0054] If P vision With P car The distance S between them is positive. According to the size relationship of the distance between the two, the longitudinal positioning observation value of the visual dimension is input into the preset filter for fusion positioning.

[0055] If P vision With P car If the distance S between them is positive, it can also be directly input into the filter for fusion positioning. That is, the current longitudinal positioning is not much different from the predicted position of the filter and can be used directly without causing the longitudinal positioning to be rejected or the trajectory to deviate to the rear of the vehicle.

[0056] According to the setting of the filter, if the first longitudinal positioning of the visual dimension is positive but does not match the preset predicted value, the longitudinal positioning observation value of the visual dimension will not be input into the preset filter for fusion positioning. For example, if the positive longitudinal positioning is too different from the predicted value, the filter will reject the positive longitudinal positioning.

[0057] When the second longitudinal positioning of the visual dimension is a positive value and is consistent with the preset prediction value, the second longitudinal positioning is used as the target value of the preset prediction value, that is, if it is consistent with the preset prediction value, it will approach this positive longitudinal positioning in a short time.

[0058] In one embodiment of the present application, the obtaining of the longitudinal positioning observation value in the visual dimension under the preset positioning scenario includes: obtaining the longitudinal positioning observation value in the visual dimension under the preset positioning scenario affected by track deduction timeout or GNSS signal anomaly through the vehicle's visual sensor; when the longitudinal positioning is a positive value, inputting the longitudinal positioning observation value in the visual dimension into a preset filter for fusion positioning to obtain the vehicle positioning result, including: when the longitudinal positioning is a positive value, inputting the longitudinal positioning observation value in the visual dimension into a Kalman filter for fusion positioning processing; until the visual sensor no longer provides the longitudinal positioning observation value, the Kalman filter returns to the normal processing state.

[0059] During specific implementation, under special circumstances, the vehicle's visual sensor is used to obtain the longitudinal positioning observation value of the visual dimension in the preset positioning scenario where the track deduction times out (which can be judged by the vehicle positioning node) or the GNSS signal is abnormal (which can be judged by the vehicle positioning node).

[0060] When the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into the Kalman filter for fusion positioning processing, and the longitudinal positioning observation value is continuously obtained until the visual sensor no longer provides the longitudinal positioning observation value, and the Kalman filter returns to the normal processing state.

[0061] In the above method, the Kalman filter will perform a chi-square test on the observation value when fusing the observation value. If the distance between the observation value and the predicted value is too large, the observation fusion will be rejected. Therefore, if the longitudinal positioning position fusion is performed, different processing needs to be done according to the size of S.

[0062] In one embodiment of the present application, the anteroposterior relationship of the distance between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle is determined based on the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, including: judging the relationship between the distance between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle based on the vehicle's driving direction.

[0063] During specific implementation, the relationship between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle is determined according to the vehicle's driving direction (the driving direction is defined as forward or reverse). If the driving direction is positive and the longitudinal positioning is negative, the relationship between the distances between the two is negative.

[0064] In a preferred embodiment of the present application, the longitudinal positioning observation value of the visual dimension is obtained in a preset positioning scenario; based on the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, the anteroposterior relationship of the distance between the two is determined, wherein the relationship includes the longitudinal positioning being a positive value or the longitudinal positioning being a negative value; when the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain the positioning result of the vehicle.

[0065] In some embodiments, the method further includes: when the longitudinal positioning is a negative value, determining the time parameter required for the longitudinal positioning in the relationship between the distance between the two to change from a negative value to a positive value; determining a target acceleration parameter value based on the time parameter and the distance between the two; and adding the target acceleration parameter value to the acceleration component in the direction of travel of the vehicle when predicting the current speed of the vehicle.

[0066] In some embodiments, the determining of a causal relationship between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value, further includes:

[0067] Calculate the distance S between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, where the direction of travel of the vehicle is considered as the positive direction;

[0068] When the distance S is a negative value, the distance between the two is shortened according to the acceleration component calculation result of the target acceleration parameter value in the direction of travel of the vehicle, which is also a negative value, wherein the target acceleration parameter value is used as a correction bias in the IMU.

[0069] In some embodiments, when the longitudinal position is a negative value, determining the time parameter required for the longitudinal position in the distance relationship between the two to change from a negative value to a positive value includes:

[0070] In the case where the longitudinal positioning is a negative value, determining the time parameter according to the current vehicle speed;

[0071] and / or, in the case where the longitudinal position is a negative value, determining the time parameter according to the current position of the vehicle;

[0072] And / or, when the longitudinal positioning is a negative value, the GNSS signal confidence is adjusted or the GNSS signal is not used to update the vehicle position according to the GNSS signal condition at the corresponding moment in the time parameter.

[0073] In some embodiments, when the longitudinal positioning is positive, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a positioning result of the vehicle, including:

[0074] In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning according to the size relationship between the two distances;

[0075] When the first longitudinal positioning value of the visual dimension is positive but does not conform to the preset prediction value, the longitudinal positioning observation value of the visual dimension is not input into the preset filter for fusion positioning;

[0076] When the second longitudinal position of the visual dimension is a positive value and is consistent with a preset prediction value, the second longitudinal position is used as a target value of the preset prediction value.

[0077] In some embodiments, obtaining the longitudinal positioning observation value of the visual dimension in a preset positioning scenario includes:

[0078] Through the vehicle's visual sensor, obtain longitudinal positioning observation values in the visual dimension in preset positioning scenarios affected by track deduction timeout or GNSS signal anomalies;

[0079] In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a positioning result of the vehicle, including:

[0080] In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into the Kalman filter for fusion positioning processing;

[0081] Until the visual sensor no longer provides the longitudinal positioning observation value, the Kalman filter returns to the normal processing state.

[0082] The present application embodiment also provides a fusion positioning device 200, such as Figure 2 , a schematic diagram of the structure of a fusion positioning device in an embodiment of the present application is provided. The fusion positioning device 200 includes at least: an acquisition module 210, a determination module 220, and a fusion positioning module 230, wherein:

[0083] An acquisition module 210 is used to obtain a longitudinal positioning observation value of a visual dimension in a preset positioning scenario;

[0084] A determination module 220 is configured to determine a causal relationship between the longitudinal positioning observation value in the visual dimension and the current position of the ego vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value;

[0085] The fusion positioning module 230 is used to input the longitudinal positioning observation value of the visual dimension into a preset filter for fusion positioning when the longitudinal positioning is a positive value, so as to obtain the positioning result of the vehicle.

[0086] In one embodiment of the present application, the acquisition module 210 is specifically configured to: In a preset positioning scenario, namely, a fusion positioning scenario, the visual sensor can provide lateral and longitudinal positioning. However, longitudinal positioning only occurs in certain specific areas, such as when the visual sensor recognizes an arrow or stop line on the road, in which case longitudinal positioning is generated.

[0087] It is understandable that when lateral positioning is successful, the field of view of the autonomous vehicle has been limited. For longitudinal positioning, it is the distance between the vehicle and the relevant target marker.

[0088] Therefore, the vertical positioning observation value of the visual dimension will appear at a certain moment, but it will not appear continuously. It is necessary to ensure that when the vertical positioning observation value of the visual dimension is obtained, it will not affect the result of positioning fusion.

[0089] Therefore, considering that the longitudinal observation value is only valid in some scenarios, this observation is not fused using the fusion process in related technologies like the lateral positioning observation or GNSS observation, but adopts the special fusion method in this application.

[0090] The current position of the vehicle can be obtained through relevant sensors.

[0091] For better explanation, the position of the vehicle is represented as P car , the vertical positioning observation value is expressed as P vision .

[0092] In one embodiment of the present application, the determining module 220 is specifically configured to determine the front-to-back relationship of the distance between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle. The front-to-back relationship may include P vision In P car The situation in front or behind. At the same time, in P vision The confidence level is high, and the preset filter can use the longitudinal positioning observation value.

[0093] This relationship includes whether longitudinal positioning is positive or negative. Regardless of whether longitudinal positioning is positive or negative, it can be used as an observation value (positive values are input directly, and negative values are input after being corrected to positive values) and input into a preset filter for multi-sensor fusion. In other words, the semantic information output by the visual sensor is used for matching, providing additional observation information.

[0094] In one embodiment of the present application, the fusion positioning module 230 is specifically configured to input the longitudinal positioning observation value in the visual dimension into a preset filter for fusion positioning when the longitudinal positioning value is positive, thereby obtaining a vehicle positioning result. In other words, only when the longitudinal positioning value is positive will it be input as a longitudinal positioning observation value into the preset filter for fusion positioning.

[0095] It is understood that the positive or negative longitudinal positioning values are relative values determined based on the fore-and-aft relationship between the longitudinal positioning observations in the visual dimension and the vehicle's current position. Furthermore, in this application, a correction bias is added to the accelerometer for observations that do not meet the conditions to ensure the stability of the fused positioning results.

[0096] It can be understood that the above-mentioned fusion positioning device can implement each step of the fusion positioning method provided in the aforementioned embodiment. The relevant explanations about the fusion positioning method are applicable to the fusion positioning device and will not be repeated here.

[0097] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 3 At the hardware level, the electronic device includes a processor and, optionally, an internal bus, a network interface, and memory. The memory may include internal memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for its services.

[0098] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0099] The memory is used to store programs. Specifically, the program may include program code, which includes computer operating instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.

[0100] The processor reads the corresponding computer program from the non-volatile memory into the internal memory and then runs it, forming a fusion positioning device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0101] Obtain the longitudinal positioning observation value of the visual dimension under the preset positioning scenario;

[0102] Determining a causal relationship between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value;

[0103] When the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain the vehicle positioning result.

[0104] The above application Figure 1 The methods performed by the fusion positioning device disclosed in the illustrated embodiments can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor or by software instructions. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can 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, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of this application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0105] The electronic device may also perform Figure 1 The method for executing the fusion positioning device in the Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0106] The embodiment of the present application also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, can enable the electronic device to execute Figure 1 The method performed by the fusion positioning device in the illustrated embodiment is specifically used to perform:

[0107] Obtain the longitudinal positioning observation value of the visual dimension under the preset positioning scenario;

[0108] Determining a causal relationship between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value;

[0109] When the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain the vehicle positioning result.

[0110] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0111] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] These computer program instructions may 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 produce an article of manufacture comprising an instruction device, which implements the process Figure 1a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0114] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0115] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0116] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. 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 technology, 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-transmission 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 (transitory media), such as modulated data signals and carrier waves.

[0117] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0118] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A fusion positioning method, wherein: For an autonomous driving vehicle, the method comprises: Obtain the longitudinal positioning observation value of the visual dimension under the preset positioning scenario; Determining a causal relationship between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value; When the longitudinal positioning is positive, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a positioning result of the vehicle; In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a positioning result of the vehicle, including: In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning according to the size relationship between the two distances; When the first longitudinal positioning value of the visual dimension is positive but does not conform to the preset prediction value, the longitudinal positioning observation value of the visual dimension is not input into the preset filter for fusion positioning; When the second longitudinal position of the visual dimension is a positive value and is consistent with a preset prediction value, the second longitudinal position is used as a target value of the preset prediction value.

2. The method according to claim 1, wherein: Also includes: In the case where the longitudinal positioning is a negative value, determining a time parameter required for the longitudinal positioning in the relationship between the distances to change from a negative value to a positive value; Determining a target acceleration parameter value according to the time parameter and the distance between the two; When predicting the current speed of the vehicle, the acceleration component in the direction of travel of the vehicle is added to the target acceleration parameter value.

3. The method according to claim 2, wherein: The method further comprises determining a causal relationship between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value, and further comprising: Calculate the distance S between the longitudinal positioning observation value of the visual dimension and the current position of the vehicle, where the direction of travel of the vehicle is considered as the positive direction; When the distance S is a negative value, the distance between the two is shortened according to the acceleration component calculation result of the target acceleration parameter value in the direction of travel of the vehicle, which is also a negative value, wherein the target acceleration parameter value is used as a correction bias in the IMU.

4. The method according to claim 2, wherein: When the longitudinal positioning value is a negative value, determining the time parameter required for the longitudinal positioning value in the distance relationship between the two to change from a negative value to a positive value includes: In the case where the longitudinal positioning is a negative value, determining the time parameter according to the current vehicle speed; and / or, in the case where the longitudinal position is a negative value, determining the time parameter according to the current position of the vehicle; And / or, when the longitudinal positioning is a negative value, the GNSS signal confidence is adjusted or the GNSS signal is not used to update the vehicle position according to the GNSS signal condition at the corresponding moment in the time parameter.

5. The method according to claim 1, wherein: The obtaining of the longitudinal positioning observation value of the visual dimension in the preset positioning scenario includes: Through the vehicle's visual sensor, obtain longitudinal positioning observation values in the visual dimension in preset positioning scenarios affected by track deduction timeout or GNSS signal anomalies; In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a positioning result of the vehicle, including: In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into the Kalman filter for fusion positioning processing; Until the visual sensor no longer provides the longitudinal positioning observation value, the Kalman filter returns to the normal processing state.

6. The method of claim 1, wherein: Determining the anteroposterior relationship of the distance between the longitudinal positioning observation value in the visual dimension and the current position of the vehicle includes: According to the vehicle's driving direction, the relationship between the longitudinal positioning observation value of the visual dimension and the distance between the current position of the vehicle is determined.

7. A fusion positioning device, wherein: For use in an autonomous driving vehicle, the device comprises: The acquisition module is used to obtain the longitudinal positioning observation value of the visual dimension under the preset positioning scenario; a determination module for determining a causal relationship of distances between the longitudinal positioning observation value in the visual dimension and the current position of the ego vehicle, wherein the relationship includes a positive longitudinal positioning value or a negative longitudinal positioning value; A fusion positioning module is used to input the longitudinal positioning observation value of the visual dimension into a preset filter for fusion positioning when the longitudinal positioning is a positive value, so as to obtain a positioning result of the vehicle; In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning to obtain a positioning result of the vehicle, including: In the case where the longitudinal positioning is a positive value, the longitudinal positioning observation value of the visual dimension is input into a preset filter for fusion positioning according to the size relationship between the two distances; When the first longitudinal positioning value of the visual dimension is positive but does not conform to the preset prediction value, the longitudinal positioning observation value of the visual dimension is not input into the preset filter for fusion positioning; When the second longitudinal position of the visual dimension is a positive value and is consistent with a preset prediction value, the second longitudinal position is used as a target value of the preset prediction value.

8. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method of any one of claims 1 to 6.

9. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute the method according to any one of claims 1 to 6.

Images