Vehicle positioning method and system based on weight adjustment

Abstract

The invention discloses a vehicle positioning method and system based on weight adjustment, and relates to the field of vehicle positioning, a positioning system arranged on a vehicle is used for executing the following steps: the positioning system generates an image quality value and a lane line confidence degree, the positioning system generates an image confidence degree according to the image quality value and the lane line confidence degree, and the image confidence degree corresponds to the image quality value and the lane line confidence degree. The positioning system generates a positioning confidence degree according to the received plurality of positioning data, and adjusts a driving image weight value, a positioning data weight value and an inertial data weight value according to the image confidence degree and the positioning confidence degree. And the positioning system adjusts an original positioning coordinate value obtained by the positioning system according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value. According to the invention, the use of a high-unit-price optical radar is avoided, and meanwhile, the problem of positioning judgment failure or abnormality possibly caused by excessive dependence on a driving image is avoided.

Description

Technical Field

[0001] The present application relates to the field of vehicle positioning, and is related to a vehicle positioning method and system based on weight adjustment, and in particular to a vehicle positioning method and system that generates positioning coordinate values ​​by adjusting driving image weight values, positioning data weight values, and inertial data weight values. Background Art

[0002] In recent years, autonomous driving has been a technology field that major car manufacturers are competing to develop. With the vigorous development of artificial intelligence and hardware technology, it is an inevitable trend that major road transportation tools will be gradually replaced by autonomous vehicles in the future. Positioning methods are crucial technologies in the field of autonomous driving. Among the many known technologies, one uses multiple cameras as the main perception, combined with artificial intelligence deep learning, to detect the surrounding environment and road markings, to achieve traffic perception cruise positioning, lane change and automatic parking and other driving assistance functions. However, excessive reliance on cameras may cause the driving image to be unable to simultaneously show the details of the dark and bright parts when encountering a situation where the light contrast in and out of the tunnel is too large, or when encountering a road section where the road markings are too worn, which can easily lead to the failure or abnormality of the driving image positioning judgment. In addition, in winter or rainy days, due to the large temperature difference between the inside and outside of the objective lens of the camera lens, water vapor condenses on the outer surface of the objective lens, resulting in the phenomenon of lens fogging, or rain splashes on the outer surface of the objective lens of the camera lens on rainy days, making the captured driving image partially blurred. These situations will also cause the driving image positioning judgment to fail or be abnormal. Another well-known technology is to combine multiple cameras and multiple optical radars (LiDAR: Light Detection And Ranging) with different ranging ranges. The driving images taken by the camera can determine the characteristic points of the lane lines. Combined with the optical radar with an accuracy of up to centimeters, and then matched with the vector map, the vehicle's position can indeed be accurately located on the vector map. However, although the optical radar has high accuracy, the cost is too high. It is difficult for car manufacturers to reduce the manufacturing cost, and the selling price naturally cannot be reduced, which affects the popularity of autonomous driving vehicles. Summary of the invention

[0003] The technical problem to be solved by the present invention is how to adjust the driving image weight value and other sensing and measurement data weight values, and use other sensing and measurement data for temporary reinforcement to generate the final positioning coordinate value when the driving image is abnormal; thereby eliminating the use of high-priced optical radars and avoiding the problem of positioning judgment failure or abnormality that may be caused by excessive reliance on driving images.

[0004] In view of this, the present invention develops a simple assembly design that can avoid the above-mentioned disadvantages, is easy to install, and has the advantage of low cost, so as to take into account considerations such as flexibility of use and economy.

[0005] In order to solve the above-mentioned problems and achieve the expected effect, the present invention provides a vehicle positioning system based on weight adjustment, which is arranged on a vehicle. The vehicle positioning system includes an image quality module, a lane line confidence module, an image confidence module, a positioning confidence module, a weight adjustment module and a positioning adjustment module. The image quality module is used to generate an image quality value according to an exposure concentration of each of a plurality of driving images taken by a plurality of image capture modules provided for the vehicle. The lane line confidence module is used to generate a lane line confidence based on a plurality of driving images. The image confidence module is used to generate an image confidence based on the image quality value and the lane line confidence. The positioning confidence module is used to generate a positioning confidence based on a plurality of positioning data received by a positioning sensing module provided for the vehicle. The weight adjustment module is used to adjust a vehicle image weight value, a positioning data weight value and an inertial data weight value according to the image confidence and the positioning confidence. The positioning adjustment module is used to adjust an original positioning coordinate value obtained by the positioning system according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value.

[0006] In addition, the present invention also provides a vehicle positioning method based on weight adjustment, which is performed using a vehicle positioning system set in a vehicle, and the vehicle positioning method includes the following steps: using an image quality module of the vehicle positioning system to generate an image quality value according to an exposure concentration of each of a plurality of driving images captured by a plurality of image capture modules set in the vehicle; using a lane line confidence module of the vehicle positioning system to generate a lane line confidence according to a plurality of driving images; using an image confidence module of the vehicle positioning system to generate an image confidence according to the image quality value and the lane line confidence; using a positioning confidence module of the vehicle positioning system to generate a positioning confidence according to a plurality of positioning data received by a positioning sensing module set in the vehicle; using a weight adjustment module of the vehicle positioning system to adjust a driving image weight value, a positioning data weight value and an inertial data weight value according to the image confidence and the positioning confidence; and using a positioning adjustment module of the vehicle positioning system to adjust an original positioning coordinate value obtained by the positioning system according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value.

[0007] During implementation, the image quality module further generates an image quality value according to an exposure concentration and an image clarity of each of a plurality of driving images captured by a plurality of image capture modules.

[0008] During implementation, the image quality module lowers the image quality value when the exposure concentration is greater than a first threshold or less than a second threshold, and increases the image quality value when the exposure concentration is less than or equal to the first threshold and greater than or equal to the second threshold, wherein the first threshold is greater than the second threshold.

[0009] In the implementation, the plurality of positioning data includes a plurality of satellite positioning data, wherein the plurality of satellite positioning data respectively correspond to the plurality of positioning data, the positioning confidence module generates a precision reduction factor value according to the plurality of satellite positioning data, and then generates the positioning confidence according to the precision reduction factor value. The positioning sensing module receives the plurality of satellite positioning data transmitted by the plurality of positioning satellites, wherein the plurality of satellite positioning data respectively correspond to the plurality of positioning satellites, wherein the magnitude of the precision reduction factor value is at least related to a number of the plurality of positioning satellites, a distribution orientation of the plurality of positioning satellites relative to the vehicle, and a distance of the plurality of positioning satellites relative to the vehicle.

[0010] During implementation, the weight adjustment module (1) adjusts the driving image weight value to 100% when the image confidence is greater than or equal to an image confidence threshold, and adjusts the positioning data weight value and the inertial data weight value to 0 respectively; (2) adjusts the driving image weight value to less than 100% when the image confidence is less than an image confidence threshold and the positioning confidence is greater than or equal to a positioning confidence threshold, and adjusts the positioning data weight value and the inertial data weight value to greater than 0 and less than the driving image weight value respectively; or (3) adjusts the driving image weight value to less than 100% when the image confidence is less than an image confidence threshold and the positioning confidence is less than a positioning confidence threshold, and adjusts the positioning data weight value to 0, and adjusts the inertial data weight value to greater than 0 and less than the driving image weight value.

[0011] During implementation, the lane confidence module detects a set of lane feature points of a plurality of driving images, generates a lane feature point distribution, and analyzes lane continuity and lane features based on the lane feature point distribution to generate a lane confidence.

[0012] During implementation, the positioning adjustment module generates a final positioning coordinate value through the following steps: using the positioning adjustment module to match the lane line feature with a vector map to generate a vehicle image coordinate value, wherein the vector map is stored in a storage module of the positioning system or obtained from a mapping module set in a cloud through a wireless receiving module set by the vehicle; when the positioning data weight value is not equal to 0, using the positioning adjustment module to generate a positioning data coordinate value according to a plurality of positioning data; when the inertial data weight value is not equal to 0, using the positioning adjustment module to generate an inertial data coordinate value according to the original positioning coordinate value and an inertial measurement data measured by an inertial measurement module set by the vehicle; and using the positioning adjustment module to generate a final positioning coordinate value, wherein the final positioning coordinate value=((driving image coordinate value×driving image weight value)+(positioning data coordinate value×positioning data weight value)+(inertial data coordinate value×inertial data weight value)), wherein the inertial data weight value=(100%-driving image weight value-positioning data weight value).

[0013] During implementation, when the positioning system has never obtained the original positioning coordinate value, the positioning system obtains the original positioning coordinate value for the first time by executing the following steps: using the positioning adjustment module to match the lane line features with a vector map to generate a driving image coordinate value, wherein the vector map is stored in a storage module of the positioning system or obtained from a mapping module set in a cloud via a wireless receiving module set in the vehicle; and using the positioning adjustment module to copy the generated driving image coordinate value to the original positioning coordinate value, so that the positioning system obtains the original positioning coordinate value with the driving image coordinate value.

[0014] During implementation, when the positioning system has obtained the original positioning coordinate value when the vehicle positioning method was executed last time, the positioning adjustment module copies the final positioning coordinate value generated when the vehicle positioning method is executed this time to the original positioning coordinate value.

[0015] During implementation, the positioning adjustment module continuously generates a positioning failure signal and continuously generates the final positioning coordinate value when the image confidence is continuously less than an image confidence threshold and the positioning confidence is continuously less than a positioning confidence threshold within a time interval; the positioning adjustment module stops continuously generating the positioning failure signal when the image confidence is greater than or equal to the image confidence threshold or the positioning confidence is greater than or equal to the positioning confidence threshold.

[0016] During implementation, the positioning adjustment module generates a warning signal when the image confidence is continuously less than an image confidence threshold and the positioning confidence is continuously greater than or equal to a positioning confidence threshold within a time interval; the positioning adjustment module stops generating a warning signal when the image confidence is greater than or equal to the image confidence threshold or the positioning confidence is less than the positioning confidence threshold.

[0017] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0018] When the driving image is abnormal, the driving image weight value, positioning data weight value and inertial data weight value are adjusted, and the positioning data and inertial data are used to provide temporary reinforcement, thereby eliminating the need to use high-priced optical radars and avoiding the problem of positioning judgment failure or abnormality caused by excessive reliance on driving images. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a functional block diagram of a specific embodiment of the present invention.

[0021] Figure 2 The figure is a flowchart of a specific embodiment of the present invention.

[0022] Figure 3 This is the exposure histogram generated by the present invention.

[0023] Figure 4 It is a schematic diagram of the variation of the DOP value generated by the present invention over time.

[0024] Figure 5 FIG. 4 is a schematic diagram showing how the image confidence level generated by the present invention when a vehicle is traveling on a flat road changes with time.

[0025] Figure 6 It is a schematic diagram showing the change of the positioning confidence level generated by the present invention when the vehicle is traveling on a flat road over time.

[0026] Figure 7 It is a schematic diagram showing the change of image confidence over time when a vehicle is traveling at a multi-road intersection and under an elevated bridge according to the present invention.

[0027] Figure 8 It is a schematic diagram showing the change of positioning confidence over time when a vehicle is traveling at a multi-road intersection and under an elevated bridge according to the present invention.

[0028] Fig. 9 The figure is a schematic diagram showing the change of the image confidence level over time when a vehicle is driving out of a tunnel according to the present invention.

[0029] Fig.10The figure is a schematic diagram showing the change of the positioning confidence level with time when a vehicle is driving out of a tunnel according to the present invention.

[0030] Fig.11 It is a schematic diagram showing the change of image confidence level over time when a vehicle is driving out of another tunnel according to the present invention.

[0031] Fig.12 It is a schematic diagram showing the change of the positioning confidence level with time when a vehicle is driving out of another tunnel according to the present invention.

[0032] Description of reference numerals:

[0033] 1: vehicle positioning system, 10: image quality module, 11: lane confidence module, 12: image confidence module, 13: positioning confidence module, 14: weight adjustment module, 15: positioning adjustment module, 16: storage module, 2: vehicle, 3: global navigation satellite system, 4: cloud, 40: mapping module, 5: inertial measurement module, 6: image capture module, 7: positioning sensing module, 8: wireless receiving module;

[0034] A1,A2,A3,B1,B2,B3,C1,C2,C3,C4,D1,D2,D3,D4,D5,D6: time interval;

[0035] E1,E2,E3,F1,F2,G1,G2,H: time interval;

[0036] T1,T2,T3,T4,T5,T6,T7: time. DETAILED DESCRIPTION

[0037] The embodiments of the present invention will be further explained below with the help of the relevant drawings. As much as possible, the same reference numerals represent the same or similar components in the drawings and the specification. In the drawings, the shapes and thicknesses may be exaggerated for simplicity and convenience. It is understood that the components not specifically shown in the drawings or described in the specification are in the form known to those skilled in the art. Those skilled in the art may make various changes and modifications based on the contents of the present invention.

[0038] See also Figure 1, which is a functional block diagram of a specific embodiment of the present invention. The vehicle positioning system 1 based on weight adjustment of the present invention is arranged on a vehicle 2. The vehicle positioning system 1 includes an image quality module 10, a lane line confidence module 11, an image confidence module 12, a positioning confidence module 13, a weight adjustment module 14, a positioning adjustment module 15 and a storage module 16. The image quality module 10 is connected to a plurality of image capture modules 6 arranged on the vehicle 2, and the plurality of image capture modules 6 can be cameras. The plurality of image capture modules 6 are used to shoot driving images. The image quality module 10 is used to receive a plurality of driving images shot by the plurality of image capture modules 6, and generate an image quality value according to an exposure concentration and an image clarity of each of the plurality of driving images. The lane confidence module 11 is connected to the plurality of image capture modules 6 , and is used for receiving the plurality of driving images captured by the plurality of image capture modules 6 , and generating a lane confidence according to the plurality of driving images.

[0039] Furthermore, the image confidence module 12 is connected to the image quality module 10 and the lane confidence module 11 respectively, and is used to receive the image quality value generated by the image quality module 10 and the lane confidence generated by the lane confidence module 11 respectively, and generate an image confidence according to the image quality value and the lane confidence. The positioning confidence module 13 is connected to a positioning sensing module 7 provided on the vehicle 2, and is used to receive a plurality of positioning data received by the positioning sensing module 7, and generate a positioning confidence according to the plurality of positioning data. The weight adjustment module 14 is connected to the image confidence module 12 and the positioning confidence module 13 respectively, and is used to receive the image confidence generated by the image confidence module 12 and the positioning confidence generated by the positioning confidence module 13 respectively, and adjust the vehicle image weight value, a positioning data weight value and an inertial data weight value according to the image confidence and the positioning confidence.

[0040] Furthermore, the positioning adjustment module 15 is respectively connected to the lane confidence module 11, the weight adjustment module 14, the positioning confidence module 13, an inertial measurement module 5 disposed in the vehicle 2, and the storage module 16. The storage module 16 is connected to a wireless receiving module 8 disposed in the vehicle 2. The storage module 16 can be used to pre-store a plurality of vector maps, or the storage module 16 can obtain vector maps from a mapping module 40 disposed in a cloud 4 via a wireless receiving module 8 at irregular intervals, and then store the vector maps in the storage module 16 in advance, or the storage module 16 can obtain vector maps from a mapping module 40 disposed in the cloud 4 in real time via the wireless receiving module 8, and then transmit the vector maps to the positioning adjustment module 15 (and store them in the storage module 16 at the same time). The positioning adjustment module 15 is used to receive the driving image weight value, positioning data weight value and inertial data weight value adjusted by the weight adjustment module 14, and adjust an original positioning coordinate value obtained by the vehicle positioning system 1 according to the driving image weight value, positioning data weight value and inertial data weight value to generate a final positioning coordinate value; when the driving image is abnormal, the driving image weight value, positioning data weight value and inertial data weight value are adjusted, and the positioning data and inertial data are used to perform temporary reinforcement, thereby eliminating the use of high-priced optical radars and avoiding the problem of positioning judgment failure or abnormality that may be caused by excessive reliance on driving images.

[0041] In addition, the present invention also provides a vehicle positioning method based on weight adjustment (please also refer to Figure 2 , which is a flowchart of a specific embodiment of the present invention), which is performed using a vehicle positioning system 1 disposed on a vehicle 2, and the vehicle positioning method includes the following steps:

[0042] Step 100: Using an image quality module 10 of the vehicle positioning system 1, generating an image quality value according to an exposure concentration and an image clarity of each of a plurality of driving images taken by a plurality of image capture modules 6 disposed on the vehicle 2;

[0043] Step 200: Using the lane confidence module 11 of the vehicle positioning system 1 to generate a lane confidence according to a plurality of driving images captured by a plurality of image capture modules 6;

[0044] Step 300: Using an image confidence module 12 of the vehicle positioning system 1 to generate an image confidence according to the image quality value generated by the image quality module 10 and the lane confidence generated by the lane confidence module 11;

[0045] Step 400: Using a positioning confidence module 13 of the vehicle positioning system 1 to generate a positioning confidence according to a plurality of positioning data received by a positioning sensor module 7 provided in the vehicle 2;

[0046] Step 500: Using a weight adjustment module 14 of the vehicle positioning system 1, according to the image confidence generated by the image confidence module 12 and the positioning confidence generated by the positioning confidence module 13, adjust a vehicle image weight value, a positioning data weight value and an inertial data weight value; and

[0047] Step 600: Using a positioning adjustment module 15 of the vehicle positioning system 1, according to the driving image weight value, positioning data weight value and inertial data weight value adjusted by the weight adjustment module 14, an original positioning coordinate value obtained by the vehicle positioning system 1 is adjusted to generate a final positioning coordinate value.

[0048] In some embodiments, in step 100, the image quality module 10 generates an image quality value based on at least the exposure concentration of each of the plurality of driving images. When the exposure concentration of the driving image is greater than a first threshold or less than a second threshold, the image quality module 10 adjusts the image quality value down; and when the exposure concentration of the driving image is less than or equal to the first threshold and greater than or equal to the second threshold, the image quality module 10 adjusts the image quality value up, wherein the first threshold is greater than the second threshold. That is, the image quality value generated by the image quality module 10 is affected by the exposure concentration of each driving image.

[0049] Generally speaking, a photosensitive element of the image capture module 6 has a dynamic range. In a single driving image captured by the image capture module 6, the darkest and brightest ranges that can be distinguished are limited by the dynamic range of the photosensitive element of the image capture module 6. When the driving image is captured under the condition that the ambient light contrast is too large, the driving image captured by the image capture module 6 cannot show the details of the dark part and the details of the bright part at the same time. That is, under the condition that the ambient light contrast is too large, when the captured driving image can clearly show the details of the dark part, the bright part will be overexposed (or called dead white), making it impossible to distinguish the details of the bright part; conversely, when the captured driving image can clearly show the details of the bright part, the dark part will be underexposed (or called dead black), making it impossible to distinguish the details of the dark part.

[0050] For example, when a vehicle is about to enter a tunnel from a tunnel entrance, if the sunlight outside the tunnel is sufficient, the light inside the tunnel is very dark in comparison (i.e., the ambient light contrast is too large), so it is possible that the captured driving image is underexposed for a short period of time. When a vehicle is about to exit a tunnel from a tunnel, if the sunlight outside the tunnel is sufficient, the light inside the tunnel is very dark in comparison (i.e., the ambient light contrast is too large), so it is possible that the captured driving image is overexposed for a short period of time. For example, when driving on a mountain road at night, the oncoming vehicle turns on the high beam, causing the image capture module 6 to be directly illuminated by the high beam, causing the driving image to be overexposed. Another example of overexposure is when driving during the dawn and dusk period, when the sunlight is at a low angle, and the vehicle is driving in the direction of the sunlight, it is easy to cause the driving image to be overexposed. An example of underexposure is when driving on a mountain road, it is easy to cause the driving image to be underexposed when driving from a road illuminated by strong sunlight to a road where the strong sunlight is blocked by the shadow of the mountain and becomes very dark.

[0051] In the above examples, when the exposure is over-exposed, that is, when the exposure concentration of the driving image is greater than the first threshold, the image quality module 10 adjusts the image quality value down; and when the exposure is under-exposed, that is, when the exposure concentration of the driving image is less than the second threshold, the image quality module 10 adjusts the image quality value down. In examples other than the above, if the exposure is neither over-exposed nor under-exposed, the image quality module 10 adjusts the image quality value up.

[0052] In a preferred embodiment, the image quality module 10 of the present invention can generate an exposure histogram for each of the plurality of driving images, and determine the exposure concentration through the exposure histogram. When the exposure concentration of the exposure histogram is excessively concentrated in a dark area, it indicates that the driving image is underexposed; and when the exposure concentration of the exposure histogram is excessively concentrated in a bright area, it indicates that the driving image is overexposed. Please also refer to Figure 3 , which is the exposure histogram generated by the present invention. Figure 3 In the exposure histogram of the driving image generated by the image quality module 10, it can be seen that the exposure concentration is obviously concentrated in the bright area, that is, the driving image is overexposed. When the exposure concentration of the exposure histogram of the driving image is greater than the first threshold, the image quality module 10 lowers the image quality value because the driving image is overexposed; when the exposure concentration of the exposure histogram of the driving image is less than the second threshold, the image quality module 10 lowers the image quality value because the driving image is underexposed; when the exposure concentration of the exposure histogram is less than or equal to the first threshold and greater than or equal to the second threshold, the image quality module 10 increases the image quality value.

[0053] In some embodiments, the image quality module 10 of the present invention generates an image quality value based on an exposure concentration and an image clarity of a plurality of driving images. The image clarity of the driving image may be affected by some factors. For example, on rainy days, rain may splash onto the outer surface of the objective lens of the lens of the image capture module 6, causing a portion of the captured driving image to be blurred, thereby affecting the image clarity. Another situation often encountered in winter or rainy days is that when the temperature difference between the inside and outside of the objective lens of the lens of the image capture module 6 is large, water vapor condenses on the outer surface of the objective lens, causing the lens to fog, which will make the image capture module 6 unable to capture a clear image. In the above situation, when the exposure concentration of the driving image is determined alone, the exposure concentration may not fall into the overexposed or underexposed area, but the quality of the driving image will indeed be affected. Therefore, in the case of poor image clarity, the image quality module 10 will lower the image quality value. During implementation, the image quality module 10 will refer to both the exposure concentration and the image clarity to generate an image quality value.

[0054] In step 200, the lane line confidence module 11 generates a lane line feature point distribution by detecting a lane line feature point set of a plurality of driving images captured by a plurality of image capture modules 6, and analyzes a lane continuity and a lane line feature according to the lane line feature point distribution to generate a lane line confidence. In a preferred embodiment, the lane line confidence module 11 detects a plurality of driving images, such as performing a second-order image processing (e.g., edge detection image processing such as Sobel or Canny) to detect a lane line feature point set. Then, based on the lane line feature point set, a lane line feature point distribution is generated, and analysis is performed based on the lane line feature point distribution to analyze a lane continuity and a lane line feature.

[0055] Lane continuity does not simply refer to whether the lane line is continuous (some lane lines are drawn as dotted lines, so they are not always continuous), but refers to whether the lane line confidence module 11 can analyze and determine, based on the distribution of lane line feature points, whether the lane along the direction in which vehicle 2 intends to travel is a lane that vehicle 2 is highly dependent on, and whether vehicle 2 can continue to travel in the direction in which vehicle 2 intends to travel along the lane determined by the lane line confidence module 11.

[0056] When the lane confidence module 11 analyzes that the lane continuity is high, it means that the lane confidence module 11 analyzes the lane along the direction in which the vehicle 2 intends to travel according to the distribution of lane feature points, and determines that the lane along the direction in which the vehicle 2 intends to travel is a lane that the vehicle 2 is highly dependent on, so that the vehicle 2 can continue to travel in the direction in which the vehicle 2 intends to travel along the lane determined by the lane confidence module 11.

[0057] When the lane confidence module 11 analyzes that the lane continuity is low, it means that the lane confidence module 11 determines that the lane along the direction in which the vehicle 2 intends to travel is not a lane that the vehicle 2 can rely on highly. Low lane continuity may be that the lane confidence module 11 is completely unable to analyze the lane along the direction in which the vehicle 2 intends to travel, such as when encountering a wide intersection without any markings; or the lane along the direction in which the vehicle 2 intends to travel analyzed by the lane confidence module 11 is not clear or doubtful, such as a large amount of lane wear, so that the lane confidence module 11 cannot accurately determine the lane.

[0058] In addition, in step 200, the lane line confidence module 11 can also analyze the lane line features, wherein the lane line features include the lane line features of all lanes around the vehicle 2. After the lane line confidence module 11 analyzes the lane line features, it transmits them to the positioning adjustment module 15 for use.

[0059] In step 500, the weight adjustment module 14 adjusts the driving image weight value, the positioning data weight value and the inertial data weight value according to the image confidence and the positioning confidence. When the weight adjustment module 14 adjusts the weight value, it is limited to (driving image weight value + positioning data weight value + inertial data weight value) = 100%.

[0060] When the image confidence is greater than or equal to an image confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to 100%, and adjusts the positioning data weight value and the inertial data weight value to 0 respectively (at this time, the coordinates of the vehicle 2 can be located entirely by relying on the information of the multiple driving images captured by the multiple image capture modules 6).

[0061] When the image confidence is less than the image confidence threshold and the positioning confidence is greater than or equal to a positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to be less than 100%, and respectively adjusts the positioning data weight value and the inertial data weight value to be greater than 0 and less than the driving image weight value (at this time, the coordinates of the vehicle 2 cannot be completely located by relying on the information of the plurality of driving images captured by the plurality of image capture modules 6, but need to rely partially on the information of the plurality of driving images, partially on the plurality of positioning data received by the positioning sensing module 7, and partially on an inertial measurement data measured by an inertial measurement module 5 provided on the vehicle 2 combined with an original positioning coordinate value obtained by the vehicle positioning system 1 to locate the coordinates of the vehicle 2).

[0062] When the image confidence is less than the image confidence threshold and the positioning confidence is less than the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to be less than 100%, and adjusts the positioning data weight value to be equal to 0, and adjusts the inertial data weight value to be greater than 0 and less than the driving image weight value (at this time, it is impossible to completely rely on the information of the multiple driving images captured by the multiple image capture modules 6 to locate the coordinates of the vehicle 2, and it is completely impossible to rely on the multiple positioning data received by the positioning sensing module 7, so it is necessary to partially rely on the information of the multiple driving images, and partially rely on the inertial measurement data measured by the inertial measurement module 5 set in the vehicle 2 combined with the original positioning coordinate value obtained by the vehicle positioning system 1 to locate the coordinates of the vehicle 2).

[0063] In step 600, the positioning adjustment module 15 adjusts the original positioning coordinate value obtained by the vehicle positioning system 1 according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate the final positioning coordinate value, including the following steps:

[0064] (1) Using the positioning adjustment module 15, the lane features are matched with a vector map to generate a vehicle image coordinate value, wherein the lane features are analyzed by the lane confidence module 11, wherein the vector map can be pre-stored in the storage module 16, or obtained in real time by the wireless receiving module 8 provided in the vehicle 2 from the mapping module 40 provided in the cloud 4, stored in the storage module 16, and transmitted to the positioning adjustment module 15 via the storage module 16;

[0065] (2) When the positioning data weight value is not equal to 0, the positioning adjustment module 15 is used to generate a positioning data coordinate value according to the plurality of positioning data received by the positioning sensing module 7;

[0066] (3) when the inertial data weight value is not equal to 0, using the positioning adjustment module 15, an inertial data coordinate value is generated according to the original positioning coordinate value obtained by the vehicle positioning system 1 and an inertial measurement data measured by an inertial measurement module 5 provided in the vehicle 2; and

[0067] (4) Use the positioning adjustment module 15 to generate a final positioning coordinate value, wherein the final positioning coordinate value = ((driving image coordinate value × driving image weight value) + (positioning data coordinate value × positioning data weight value) + (inertial data coordinate value × inertial data weight value)), wherein the inertial data weight value = (100% - driving image weight value - positioning data weight value).

[0068] The driving image coordinate value is the coordinate of the vehicle 2 obtained when the vehicle 2 is positioned completely by relying on the information of the plurality of driving images captured by the plurality of image capture modules 6 (since the lane line features analyzed by the lane line confidence module 11 include the lane line features of all lanes around the vehicle 2, the positioning adjustment module 15 matches the lane line features of all these lanes with the vector map to compare the position of the vehicle 2 in the vector map to generate the driving image coordinate value; that is, the coordinate of the vehicle 2 is obtained when the vehicle 2 is positioned completely by relying on the information of the plurality of driving images captured by the plurality of image capture modules 6); the positioning data coordinate value is the coordinate of the vehicle 2 obtained when the vehicle 2 is positioned completely by relying on the plurality of positioning data received by the positioning sensing module 7; and the inertial data coordinate value is the coordinate of the vehicle 2 obtained when the vehicle 2 is positioned completely by relying on the original positioning coordinate value combined with the inertial measurement data measured by the inertial measurement module 5. When the driving image is abnormal, the driving image weight value, positioning data weight value and inertial data weight value are adjusted, and the positioning data and inertial data are used to provide temporary reinforcement, thereby eliminating the need to use high-priced optical radars and avoiding the problem of positioning judgment failure or abnormality caused by excessive reliance on driving images.

[0069] In step 600, when the vehicle positioning system 1 has never obtained the original positioning coordinate value (that is, when the vehicle positioning system 1 of the present invention is used to perform the vehicle positioning method of the present invention for the first time), the vehicle positioning system 1 performs the following steps to obtain the original positioning coordinate value for the first time:

[0070] Using the positioning adjustment module 15, the lane features are matched with a vector map to generate a vehicle image coordinate value, wherein the lane features are analyzed by the lane confidence module 11, wherein the vector map can be pre-stored in the storage module 16, or obtained in real time from the mapping module 40 disposed in the cloud 4 via the wireless receiving module 8 disposed in the vehicle 2, stored in the storage module 16, and transmitted to the positioning adjustment module 15 via the storage module 16; and

[0071] The positioning adjustment module 15 is used to copy the generated driving image coordinate value to the original positioning coordinate value, so that the vehicle positioning system 1 obtains the original positioning coordinate value for the first time. The above-mentioned step of the vehicle positioning system 1 obtaining the original positioning coordinate value for the first time is a step performed when the coordinates of the vehicle 2 can be completely located by the information of the plurality of driving images captured by the plurality of image capture modules 6.

[0072] When executing step 600, the vehicle positioning system 1 has obtained the original positioning coordinate value (that is, when it is not the first time to use the vehicle positioning system 1 of the present invention to perform the vehicle positioning method of the present invention), the vehicle positioning method of the present invention further includes the following step after executing step 600: using the positioning adjustment module 15 to copy the final positioning coordinate value generated when the vehicle positioning method of the present invention is executed this time to the original positioning coordinate value. Therefore, the original positioning coordinate value obtained by the vehicle positioning system 1 when the vehicle positioning method of the present invention is executed next time is the final positioning coordinate value generated by the positioning adjustment module 15 when the vehicle positioning method of the present invention is executed this time.

[0073] In a preferred embodiment, the inertial measurement module 8 of the present invention includes three-directional accelerometers and three-axis gyroscopes; for example, the three-directional accelerometers are accelerometers facing the right direction, the front direction, and the top direction respectively; the three-axis gyroscopes are gyroscopes along an axis along the right direction, an axis along the front direction, and an axis along the top direction respectively. The inertial measurement module 8 is used to measure an inertial measurement data, wherein the inertial measurement data includes accelerations in three directions and a three-axis attitude angle.

[0074] When the vehicle positioning method of the present invention is executed for the last time, the vehicle 2 is located at the final positioning coordinate value generated by the positioning adjustment module 15 at that time. After a time interval (the time interval is a time difference between the last execution of the vehicle positioning method of the present invention and the current execution of the vehicle positioning method of the present invention), when the vehicle positioning method of the present invention is executed this time, the vehicle 2 moves from the position of the original positioning coordinate value (i.e., the final positioning coordinate value generated by the positioning adjustment module 15 when the vehicle positioning method of the present invention was executed last time) to the inertial data coordinate value (this is the coordinate of the vehicle 2 obtained when the vehicle 2 is positioned completely by relying on the original positioning coordinate value obtained by the vehicle positioning system 1 in combination with the inertial measurement data measured by the inertial measurement module 5), wherein the inertial data coordinate value is calculated by the positioning adjustment module 15 based on the acceleration in three directions and the three-axis attitude angle of the inertial measurement data and the position of the original positioning coordinate value obtained by the vehicle positioning system 1 (i.e., the final positioning coordinate value generated by the positioning adjustment module 15 when the vehicle positioning method of the present invention was executed last time) After the time interval (the time interval is a time difference between the last execution of the vehicle positioning method of the present invention and the current execution of the vehicle positioning method of the present invention), the vehicle 2 will move from the position of the original positioning coordinate value to the position of the inertial data coordinate value.

[0075] In other embodiments, the inertial measurement module 8 of the present invention includes a three-directional accelerometer, a three-axis gyroscope and three magnetometers, wherein the inertial measurement data measured by the inertial measurement module 8 includes acceleration in three directions, a three-axis attitude angle, a magnetic field strength of the three axes, a magnetic inclination and a magnetic declination.

[0076] In step 600, when the positioning adjustment module 15 has an image confidence level that is continuously less than an image confidence threshold and a positioning confidence level that is continuously greater than or equal to a positioning confidence threshold within a first time interval, the positioning adjustment module 15 generates a warning signal, thereby warning a driver of the vehicle 2 that within the aforementioned first time interval, the vehicle positioning system 1 partially relies on the driving image coordinate value, partially relies on the positioning data coordinate value, and partially relies on the inertial data coordinate value to locate the coordinates of the vehicle 2, and at this time, it is necessary to pay careful attention to the status of the vehicle 2. When the image confidence level of the positioning adjustment module 15 is greater than or equal to the image confidence threshold or the positioning confidence level is less than the positioning confidence threshold, the positioning adjustment module 15 stops generating the warning signal.

[0077] In step 600, when the positioning adjustment module 15 continuously generates a positioning failure signal when the image confidence is continuously less than an image confidence threshold and the positioning confidence is continuously less than a positioning confidence threshold within a second time interval, the positioning adjustment module 15 continuously generates a positioning failure signal to warn a driver of the vehicle 2, and the final positioning coordinate value is continuously generated before the driver manually takes over to drive the vehicle 2. Within the aforementioned second time interval, the vehicle positioning system 1 partially relies on the driving image coordinate value and partially relies on the inertial data coordinate value to locate the coordinates of the vehicle 2, and the driver needs to manually take over to drive the vehicle 2. When the image confidence of the positioning adjustment module 15 is greater than or equal to the image confidence threshold or the positioning confidence is greater than or equal to the positioning confidence threshold, the positioning adjustment module 15 stops continuously generating the positioning failure signal.

[0078] In a preferred embodiment, the positioning sensing module 7 is a satellite positioning sensing module. The satellite positioning sensing module is used to receive a plurality of satellite positioning data transmitted by a plurality of satellites of a global navigation satellite system (GNSS: Global Navigation Satellite System) 3. In step 400, the plurality of positioning data received by the positioning sensing module 7 includes a plurality of satellite positioning data, wherein the plurality of satellite positioning data respectively correspond to the plurality of positioning data. The positioning confidence module 13 generates a precision reduction factor value according to the plurality of satellite positioning data received by the positioning sensing module 7, and then generates a positioning confidence according to the precision reduction factor value.

[0079] Please also see Figure 4, which is a schematic diagram of the change of the precision reduction factor value generated by the present invention over time. Figure 4 It is shown that the value of the precision reduction factor changes dynamically over time. When the precision reduction factor value is larger, the relative positions between the satellites are closer and the positioning accuracy is lower; when the precision reduction factor value is smaller, the relative positions between the satellites are more dispersed and the positioning accuracy is higher. For example, when the precision reduction factor value is greater than 6, the positioning accuracy is poor; and when the precision reduction factor value is less than 1, the positioning accuracy is excellent. The size of the precision reduction factor value is at least related to the number of multiple positioning satellites, the distribution orientation of the multiple positioning satellites relative to the vehicle 2, the distance of the multiple positioning satellites relative to the vehicle 2, whether there are obstacles between the multiple positioning satellites and the vehicle 2, and weather factors.

[0080] In some embodiments, the image confidence module 12 may adjust an image quality value weight and a lane line confidence weight, wherein the image confidence generated by the image confidence module 12 = image quality value × image quality value weight) + (lane line confidence × lane line confidence weight). Wherein (image quality value weight + lane line confidence weight) = 100%. When the image quality value is high and the lane line confidence is high, the image confidence generated by the image confidence module 12 is high; when the image quality value is low and the lane line confidence is low, the image confidence generated by the image confidence module 12 is low; and when the image quality value is low and the lane line confidence is high, or when the image quality value is high and the lane line confidence is low, the image confidence generated by the image confidence module 12 is calculated based on the image quality value weight and the lane line confidence weight.

[0081] See also Figure 5 and Figure 6 , which are schematic diagrams showing the image confidence and positioning confidence generated by the present invention when the vehicle is traveling on a flat road over time. In this embodiment, the positioning sensing module 7 is a satellite positioning sensing module. Figure 5 and Figure 6 The time axis in the two is exactly the same and corresponding. Figure 5 In the embodiment, the vehicle positioning system 1 of the present invention is used to respectively execute the steps 100, 200 and 300 of the vehicle positioning method of the present invention to obtain a schematic diagram of image confidence varying with time, wherein the range of the image confidence is set to be greater than or equal to 0 and less than or equal to 1, and the image confidence threshold is set to 0.5; in Figure 6 In the embodiment, step 400 of the vehicle positioning method of the present invention is executed to obtain a schematic diagram of positioning confidence varying with time, wherein the range of positioning confidence is set to be greater than or equal to 0 and less than or equal to 1, and the positioning confidence threshold is set to 0.5.

[0082] exist Figure 5 and Figure 6 In the embodiment of A1, A2 and A3, the vehicle 2 is traveling on a road with clear and good markings; while in the B1, B2 and B3 time intervals, the vehicle 2 is traveling to a large intersection. When the vehicle 2 is traveling in the A1 time interval, the road markings are clear and good, and the image confidence is 1 (greater than the image confidence threshold of 0.5). Therefore, in step 500, the weight adjustment module 14 adjusts the driving image weight value to 100%, and adjusts the positioning data weight value and the inertial data weight value to 0 respectively. At this time, the coordinates of the vehicle 2 can be completely located by relying entirely on the information of the plurality of driving images captured by the plurality of image capture modules 6. In step 600, the positioning adjustment module 15 generates a final positioning coordinate value = (driving image coordinate value × 100%) based on the driving image weight value, the positioning data weight value and the inertial data weight value.

[0083] When the vehicle 2 travels to the B1 time interval, it encounters a large crossroad, and the lane line confidence module 11 lowers the lane line confidence, resulting in an image confidence of 0 (less than the image confidence threshold of 0.5); but since the positioning sensing module 7 receives a good signal and the precision reduction factor value is high, the positioning confidence is 1 (greater than the positioning confidence threshold of 0.5); therefore, in step 500, the weight adjustment module 14 adjusts the driving image weight value to less than 100% (for example, the driving image weight value is adjusted to 80%), and respectively adjusts the positioning data weight value and the inertial data weight value to 100%. The weight value is adjusted to be greater than 0 and less than the driving image weight value (for example, the positioning data weight value and the inertial data weight value are adjusted to 10% respectively). At this time, the coordinates of the vehicle 2 cannot be completely located by relying on the information of the plurality of driving images captured by the plurality of image capture modules 6, but need to rely partially on the information of the plurality of driving images, partially on the plurality of positioning data received by the positioning sensing module 7, and partially on the inertial measurement data measured by the inertial measurement module 5 combined with the original positioning coordinate value obtained by the vehicle positioning system 1 to locate the coordinates of the vehicle 2. In step 600, the positioning adjustment module 15 adjusts the original positioning coordinate value according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value = ((driving image coordinate value × 80%) + (positioning data coordinate value × 10%) + (inertial data coordinate value × 10%)).

[0084] When vehicle 2 travels to time intervals A2 and A3, the image confidence is 1 (image confidence is not considered at this time), so the situation is the same as when traveling in time interval A1. When vehicle 2 travels to time intervals B2 and B3, the image confidence is 0, but the positioning confidence is 1, so the situation is the same as when traveling to time interval B1.

[0085] See also Figure 7 and Figure 8 , which are schematic diagrams showing the image confidence and positioning confidence generated by the present invention when the vehicle is traveling at a multi-fork intersection and under an elevated bridge. In this embodiment, the positioning sensing module 7 is a satellite positioning sensing module. Figure 7 and Figure 8 The time axis in the two is exactly the same and corresponding. Figure 7 In the example, the steps for obtaining a schematic diagram of image confidence changing over time, the range of image confidence setting, and the size of image confidence threshold setting are all the same as Figure 5 same; in Figure 8 In the above, the steps of obtaining the schematic diagram of the change of positioning confidence over time, the range of positioning confidence setting and the size of positioning confidence threshold setting are all the same as Figure 6 same.

[0086] exist Figure 7 and Figure 8 In the embodiment, during the period between 0 and T1 on the time axis and after T2 on the time axis, the vehicle 2 is traveling on a flat road, and the satellite signal is good at this time; during the period between T1 and T2 on the time axis, the vehicle 2 is traveling under an elevated bridge, and the satellite signal is often interfered by the elevated bridge; and in the C1 time interval, the vehicle 2 encounters a multi-fork intersection when traveling on a flat road; in the C2, C3, and C4 time intervals, the vehicle 2 is traveling under an elevated bridge and some shadows cover the lane lines or some lane lines are partially worn; in the D1 and D6 time intervals, the vehicle 2 is traveling on a flat road and the markings are clear and good; in the D2, D3, D4, and D5 time intervals, the vehicle 2 is traveling under an elevated bridge and the markings are clear and good.

[0087] When the vehicle 2 is traveling in the C1 time interval, it encounters a multi-fork intersection, and the lane line confidence module 11 lowers the lane line confidence, resulting in an image confidence of 0 (less than the image confidence threshold of 0.5); but since the positioning sensing module 7 receives a good signal and the precision reduction factor value is high, the positioning confidence is 1 (greater than the positioning confidence threshold of 0.5); therefore, in step 500, the weight adjustment module 14 adjusts the driving image weight value to less than 100% (for example, the driving image weight value is adjusted to 70%), and respectively adjusts the positioning data weight value and the inertial data weight value When the weight value is adjusted to be greater than 0 and less than the driving image weight value (for example, the positioning data weight value is adjusted to 20%, and the inertial data weight value is adjusted to 10%), the coordinates of the vehicle 2 cannot be completely located by relying on the information of the plurality of driving images captured by the plurality of image capture modules 6, but it is necessary to partially rely on the information of the plurality of driving images, partially rely on the plurality of positioning data received by the positioning sensing module 7, and partially rely on the inertial measurement data measured by the inertial measurement module 5 combined with the original positioning coordinate value obtained by the vehicle positioning system 1 to locate the coordinates of the vehicle 2.

[0088] In step 600, the positioning adjustment module 15 adjusts the original positioning coordinate value according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value = ((driving image coordinate value × 70%) + (positioning data coordinate value × 20%) + (inertial data coordinate value × 10%)); however, since the section of the multi-fork intersection encountered in the C1 time interval is long, the time (T4-0) that the vehicle 2 travels in the C1 time interval exceeds a pre-set first time interval 9T3-0); therefore, when the vehicle 2 travels to the T3 time, the positioning adjustment module 15 immediately generates a warning signal to warn the driver that the vehicle positioning system 1 partially relies on the driving image coordinate value, partially relies on the positioning data coordinate value, and partially relies on the inertial data coordinate value to locate the coordinates of the vehicle 2, and at this time, it is necessary to pay attention to the condition of the vehicle 2.

[0089] When the vehicle 2 travels to the D1 time interval, the road markings are clear and good, and the image confidence is 1 (greater than the image confidence threshold 0.5). Therefore, in step 500, the weight adjustment module 14 adjusts the driving image weight value to 100%, and adjusts the positioning data weight value and the inertial data weight value to 0 respectively. At this time, the coordinates of the vehicle 2 can be located completely by relying on the information of the multiple driving images captured by the multiple image capture modules 6; in step 600, since the coordinates of the vehicle 2 are located completely by relying on the information of the multiple driving images captured by the multiple image capture modules 6, the positioning adjustment module 15 immediately stops generating a warning signal, and then generates a final positioning coordinate value = (driving image coordinate value × 100%) based on the driving image weight value, the positioning data weight value and the inertial data weight value.

[0090] When the vehicle 2 travels to the C2, C3, and C4 time intervals, since the image confidence is 0, but the positioning confidence is greater than or equal to the positioning confidence threshold of 0.5, the situation is roughly the same as when the vehicle 2 travels to the C1 time interval, except that the time when the vehicle 2 travels to the C2, C3, and C4 time intervals does not exceed the first time interval (T3-0), so the positioning adjustment module 15 does not generate a warning signal. When the vehicle 2 travels to the D2, D3, D4, D5, and D6 time intervals, the image confidence is 1 (image confidence is not required at this time), so the situation is roughly the same as when the vehicle 2 travels to the D1 time interval, except that the warning signal is not generated before the vehicle 2 travels to the D2, D3, D4, D5, and D6 time intervals, so the positioning adjustment module 15 will not stop generating the warning signal.

[0091] See also Fig. 9 and Fig.10 , are schematic diagrams showing the image confidence and positioning confidence generated by the present invention when a vehicle is driving out of a tunnel. In this embodiment, the positioning sensing module 7 is a satellite positioning sensing module. Fig. 9 and Fig.10 The time axis in the two is exactly the same and corresponding. Fig. 9 In the example, the steps for obtaining a schematic diagram of image confidence changing over time, the range of image confidence setting, and the size of image confidence threshold setting are all the same as Figure 5 same; in Fig.10 In the above, the steps of obtaining the schematic diagram of the change of positioning confidence over time, the range of positioning confidence setting and the size of positioning confidence threshold setting are all the same as Figure 6 same.

[0092] exist Fig. 9 and Fig.10In the embodiment of the invention, the E1 time interval is when the vehicle 2 is traveling in the tunnel; the F1, E2 and F2 time intervals are when the vehicle 2 is traveling near the tunnel entrance; and the E3 time interval is when the vehicle 2 has just exited the tunnel entrance. When the vehicle 2 is traveling in the E1 time interval, the lighting in the tunnel is good, the road markings are clear and good, and the image confidence is 1 (greater than the image confidence threshold of 0.5). Therefore, in step 500, the weight adjustment module 14 adjusts the driving image weight value to 100%, and adjusts the positioning data weight value and the inertial data weight value to 0 respectively. At this time, the coordinates of the vehicle 2 can be completely located by relying on the information of the plurality of driving images captured by the plurality of image capture modules 6; in step 600, the positioning adjustment module 15 generates a final positioning coordinate value = (driving image coordinate value × 100%) according to the driving image weight value, the positioning data weight value and the inertial data weight value.

[0093] When the vehicle 2 reaches the F1 time interval (just reaching the tunnel entrance), the sunlight outside the tunnel is sufficient. In contrast, the light inside the tunnel is very dark, causing the captured driving image to be overexposed in a short period of time (the exposure concentration is obviously concentrated in the bright area). Therefore, the lane line confidence module 11 lowers the lane line confidence to make the image confidence 0 (less than the image confidence threshold 0.5). At this time, because the vehicle has not completely exited the tunnel, the positioning sensing module 7 has poor signal reception, resulting in a low precision reduction factor value, so the positioning confidence is 0 (less than the positioning confidence threshold 0.5). Therefore, in step 500, the weight adjustment module 14 adjusts the driving image weight value to less than 100% (for example, the driving image weight value is adjusted to 80%), and adjusts the positioning data weight value to equal to 0, and adjusts the inertial data weight value to equal to 0. When the value is greater than 0 and less than the driving image weight value (for example, the inertial data weight value is adjusted to 20%), the coordinates of the vehicle 2 cannot be completely located by the information of the plurality of driving images captured by the plurality of image capture modules 6, and the plurality of positioning data received by the positioning sensing module 7 cannot be completely relied upon. Therefore, the coordinates of the vehicle 2 need to be located partially by the information of the plurality of driving images, and partially by the inertial measurement data measured by the inertial measurement module 5 disposed on the vehicle 2 in combination with the original positioning coordinate value obtained by the vehicle positioning system 1. In step 600, the positioning adjustment module 15 adjusts the original positioning coordinate value according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value = ((driving image coordinate value × 80%) + (inertial data coordinate value × 20%)).

[0094] When vehicle 2 travels to time intervals E2 and E3, the image confidence is 1 (image confidence is not considered at this time), so the situation is the same as when traveling in time interval E1. When vehicle 2 travels to time interval F2, the image confidence is 0 and the positioning confidence is 0, so the situation is the same as when traveling to time interval F1.

[0095] See also Fig.11 and Fig.12 , which are schematic diagrams showing the image confidence and positioning confidence generated by the present invention when a vehicle is driving out of another tunnel. In this embodiment, the positioning sensing module 7 is a satellite positioning sensing module. Fig.11 and Fig.12 The time axis in the two is exactly the same and corresponding. Fig.11 In the example, the steps for obtaining a schematic diagram of image confidence changing over time, the range of image confidence setting, and the size of image confidence threshold setting are all the same as Figure 5 same; in Fig.12 In the above, the steps of obtaining the schematic diagram of the change of positioning confidence over time, the range of positioning confidence setting and the size of positioning confidence threshold setting are all the same as Figure 6 same.

[0096] exist Fig.11 and Fig.12 In the embodiment of the invention, the G1 time interval is when the vehicle 2 is driving in the tunnel; the H time interval is when the vehicle 2 is driving near the tunnel entrance; and the G2 time interval is when the vehicle 2 has just driven out of the tunnel entrance. When the vehicle 2 is driving in the G1 time interval, the lighting in the tunnel is good, the road markings are clear and good, and the image confidence is 1 (greater than the image confidence threshold of 0.5). This situation is similar to the situation when the vehicle 2 is driving in the G1 time interval. Fig. 9 and Fig.10 Therefore, for details of steps 500 and 600, please refer to the case where the vehicle 2 is traveling in Fig. 9 and Fig.10 The description during the E1 time interval in the embodiment.

[0097] When vehicle 2 reaches the H time interval (just reaching the tunnel entrance), the exposure concentration is obviously concentrated in the bright area, that is, the driving image is overexposed. After the lane line confidence module 11 lowers the lane line confidence, the image confidence is 0 (less than the image confidence threshold 0.5). At this time, because vehicle 2 has not completely exited the tunnel, the positioning sensing module 7 has poor signal reception, resulting in a low precision reduction factor value, so the positioning confidence is 0 (less than the positioning confidence threshold 0.5). This situation is similar to the situation when vehicle 2 is driving in Fig. 9 and Fig.10Therefore, for details of steps 500 and 600, please refer to the case where the vehicle 2 is traveling in Fig. 9 and Fig.10 The description of the F1 time interval in the embodiment.

[0098] However, since the time (T7-T5) during which the vehicle 2 travels in the H time interval (the image confidence is 0 and the positioning confidence is 0) exceeds a pre-set second time interval (T6-T5), when the vehicle 2 travels to the T6 time, the positioning adjustment module 15 immediately continues to generate a positioning failure signal and continues to generate the final positioning coordinate value, thereby warning a driver of the vehicle 2 that the vehicle positioning system 1 partially relies on the driving image coordinate value and partially relies on the inertial data coordinate value to locate the coordinates of the vehicle 2, and the driver needs to manually take over to drive the vehicle 2. When the vehicle 2 travels to the G2 time interval, the image confidence is all 1 (the image confidence does not need to be considered at this time), and the positioning adjustment module 15 stops generating the positioning failure signal, so the situation is the same as when the vehicle 2 travels in the G1 time interval.

[0099] In some embodiments, when the image confidence is less than the image confidence threshold and the positioning confidence is greater than or equal to the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to be greater than or equal to 70% and less than or equal to 90%, and adjusts the positioning data weight value and the inertial data weight value to be greater than or equal to 5% and less than or equal to 15%, respectively. In other embodiments, when the image confidence is less than the image confidence threshold and the positioning confidence is greater than or equal to the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to be greater than or equal to 60% and less than or equal to 90%, and adjusts the positioning data weight value and the inertial data weight value to be greater than or equal to 5% and less than or equal to 20%, respectively.

[0100] In some embodiments, when the image confidence is less than the image confidence threshold and the positioning confidence is less than the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to be greater than or equal to 70% and less than or equal to 90%, and adjusts the inertial data weight value to be greater than or equal to 10% and less than or equal to 30%. In other embodiments, when the image confidence is less than the image confidence threshold and the positioning confidence is less than the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to be greater than or equal to 65% and less than or equal to 95%, and adjusts the inertial data weight value to be greater than or equal to 5% and less than or equal to 35%.

[0101] In some embodiments, the vehicle positioning system 1 adopts an inertial measurement module 5 with higher measurement accuracy, so the weight value of the inertial data can be appropriately increased; for example, when the image confidence is less than the image confidence threshold and the positioning confidence is greater than or equal to the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to greater than or equal to 60% and less than or equal to 85%, adjusts the positioning data weight value to greater than or equal to 5% and less than or equal to 10%, and adjusts the inertial data weight value to greater than or equal to 10% and less than or equal to 30%; for another example, when the image confidence is less than the image confidence threshold and the positioning confidence is less than the positioning confidence threshold, the weight adjustment module 14 adjusts the driving image weight value to greater than or equal to 60% and less than or equal to 85%, and adjusts the inertial data weight value to greater than or equal to 15% and less than or equal to 40%. In some embodiments, the vehicle positioning system 1 adopts an inertial measurement module 5 with higher measurement accuracy, and at this time, the aforementioned first time interval and second time interval can be appropriately extended.

[0102] The above are specific embodiments of the present invention and the technical means used. According to the disclosure or teaching of this article, many changes and modifications can be derived and deduced, which can still be regarded as equivalent changes made to the concept of the present invention. The effects produced still do not exceed the essential spirit covered by the description and drawings, and should be regarded as within the technical scope of the present invention. It is stated in advance.

[0103] To summarize, according to the content disclosed above, when the driving image is abnormal, the vehicle positioning method and system based on weight adjustment of the present invention adjusts the driving image weight value, the positioning data weight value and the inertial data weight value, and uses the positioning data and inertial data for temporary reinforcement, thereby eliminating the use of high-priced optical radars and avoiding the problem of positioning judgment failure or abnormality that may be caused by excessive reliance on driving images.

[0104] The above description is only used to illustrate the preferred implementation mode of the present invention, and is not intended to limit the scope of implementation. All simple replacements and equivalent changes made according to the scope of the patent application of the present invention and the content of the patent specification are within the scope of the patent application of the present invention.

Claims

1. A vehicle positioning method based on weight adjustment, which is performed using a positioning system arranged on a vehicle, and the vehicle positioning method based on weight adjustment comprises the following steps: Using an image quality module of the positioning system, generating an image quality value according to an exposure concentration of each of a plurality of driving images taken by a plurality of image capture modules disposed on the vehicle; Using a lane confidence module of the positioning system to generate a lane confidence according to the plurality of driving images; Using an image confidence module of the positioning system to generate an image confidence according to the image quality value and the lane line confidence; Using a positioning confidence module of the positioning system to generate a positioning confidence according to a plurality of positioning data received by a positioning sensing module provided in the vehicle; Using a weight adjustment module of the positioning system to adjust a vehicle image weight value, a positioning data weight value, and an inertial data weight value according to the image confidence level and the positioning confidence level; and A positioning adjustment module of the positioning system is used to adjust an original positioning coordinate value obtained by the positioning system according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value.

2. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The image quality module further generates the image quality value according to an exposure concentration and an image clarity of each of the plurality of driving images captured by the plurality of image capture modules.

3. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The image quality module lowers the image quality value when the exposure concentration is greater than a first threshold or less than a second threshold, and increases the image quality value when the exposure concentration is less than or equal to the first threshold and greater than or equal to the second threshold, wherein the first threshold is greater than the second threshold.

4. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The plurality of positioning data includes a plurality of satellite positioning data, and the plurality of satellite positioning data respectively correspond to the plurality of positioning data. The positioning confidence module generates a precision reduction factor value according to the plurality of satellite positioning data, and then generates the positioning confidence according to the precision reduction factor value.

5. The vehicle positioning method based on weight adjustment according to claim 4, characterized in that: The positioning sensing module receives the plurality of satellite positioning data transmitted by a plurality of positioning satellites, the plurality of satellite positioning data respectively corresponding to the plurality of positioning satellites, and the magnitude of the precision reduction factor value is at least related to a number of the plurality of positioning satellites, a distribution position of the plurality of positioning satellites relative to the vehicle, and a distance of the plurality of positioning satellites relative to the vehicle.

6. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The weight adjustment module adjusts the driving image weight value to 100% when the image confidence is greater than or equal to an image confidence threshold, and adjusts the positioning data weight value and the inertial data weight value to 0 respectively; when the image confidence is less than an image confidence threshold and the positioning confidence is greater than or equal to a positioning confidence threshold, the driving image weight value is adjusted to less than 100%, and the positioning data weight value and the inertial data weight value are adjusted to be greater than 0 and less than the driving image weight value respectively; or when the image confidence is less than an image confidence threshold and the positioning confidence is less than a positioning confidence threshold, the driving image weight value is adjusted to less than 100%, the positioning data weight value is adjusted to 0, and the inertial data weight value is adjusted to be greater than 0 and less than the driving image weight value.

7. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The lane line confidence module detects a lane line feature point set of the plurality of driving images to generate a lane line feature point distribution, and analyzes a lane continuity and a lane line feature according to the lane line feature point distribution to generate the lane line confidence.

8. The vehicle positioning method based on weight adjustment according to claim 7, characterized in that: Using the positioning adjustment module to generate the final positioning coordinate value includes the following steps: Using the positioning adjustment module, matching the lane line features with a vector map to generate a vehicle image coordinate value, wherein the vector map is stored in a storage module of the positioning system or obtained from a mapping module disposed in a cloud via a wireless receiving module disposed in the vehicle; When the positioning data weight value is not equal to 0, using the positioning adjustment module to generate a positioning data coordinate value according to the plurality of positioning data; When the inertial data weight value is not equal to 0, using the positioning adjustment module to generate an inertial data coordinate value according to the original positioning coordinate value and an inertial measurement data measured by an inertial measurement module provided on the vehicle; and The positioning adjustment module is used to generate the final positioning coordinate value, wherein the final positioning coordinate value = ((driving image coordinate value × driving image weight value) + (positioning data coordinate value × positioning data weight value) + (inertial data coordinate value × inertial data weight value)), wherein the inertial data weight value = (100% - driving image weight value - positioning data weight value).

9. The vehicle positioning method based on weight adjustment according to claim 7, characterized in that: When the positioning system has never obtained the original positioning coordinate value, the positioning system obtains the original positioning coordinate value for the first time by performing the following steps: Using the positioning adjustment module, matching the lane line features with a vector map to generate a vehicle image coordinate value, wherein the vector map is stored in a storage module of the positioning system or obtained from a mapping module disposed in a cloud via a wireless receiving module disposed in the vehicle; and The positioning adjustment module is used to copy the generated driving image coordinate value to the original positioning coordinate value, so that the positioning system obtains the original positioning coordinate value having the driving image coordinate value.

10. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: When the positioning system has obtained the original positioning coordinate value when the vehicle positioning method was executed last time, the positioning adjustment module copies the final positioning coordinate value generated when the vehicle positioning method is executed this time to the original positioning coordinate value.

11. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The positioning adjustment module continuously generates a positioning failure signal and continuously generates the final positioning coordinate value when the image confidence is continuously less than an image confidence threshold and the positioning confidence is continuously less than a positioning confidence threshold within a time interval.

12. The vehicle positioning method based on weight adjustment according to claim 11, characterized in that: The positioning adjustment module stops continuously generating the positioning failure signal when the image confidence is greater than or equal to the image confidence threshold or when the positioning confidence is greater than or equal to the positioning confidence threshold.

13. The vehicle positioning method based on weight adjustment according to claim 1, characterized in that: The positioning adjustment module generates a warning signal when the image confidence level is continuously less than an image confidence threshold and the positioning confidence level is continuously greater than or equal to a positioning confidence threshold within a time interval.

14. The vehicle positioning method based on weight adjustment according to claim 13, characterized in that: The positioning adjustment module stops generating the warning signal when the image confidence is greater than or equal to the image confidence threshold or when the positioning confidence is less than the positioning confidence threshold.

15. A vehicle positioning system based on weight adjustment, arranged on a vehicle, the vehicle positioning system comprising: an image quality module, used for generating an image quality value according to an exposure concentration of each of a plurality of driving images taken by a plurality of image capture modules disposed on the vehicle; A lane line confidence module, used to generate a lane line confidence according to the plurality of driving images; an image confidence module, configured to generate an image confidence according to the image quality value and the lane line confidence; a positioning confidence module, used to generate a positioning confidence according to a plurality of positioning data received by a positioning sensing module provided in the vehicle; a weight adjustment module, used for adjusting a vehicle image weight value, a positioning data weight value and an inertial data weight value according to the image confidence level and the positioning confidence level; and A positioning adjustment module is used to adjust an original positioning coordinate value obtained by the positioning system according to the driving image weight value, the positioning data weight value and the inertial data weight value to generate a final positioning coordinate value.

16. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The image quality module further generates the image quality value according to an exposure concentration and an image clarity of each of the plurality of driving images captured by the plurality of image capture modules.

17. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The image quality module is used to lower the image quality value when the exposure concentration is greater than a first threshold or less than a second threshold, and to increase the image quality value when the exposure concentration is less than or equal to the first threshold and greater than or equal to the second threshold, wherein the first threshold is greater than the second threshold.

18. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The plurality of positioning data includes a plurality of satellite positioning data, and the plurality of satellite positioning data respectively corresponds to the plurality of positioning data. The positioning confidence module is used to generate a precision reduction factor value according to the plurality of satellite positioning data, and then generate the positioning confidence according to the precision reduction factor value.

19. The vehicle positioning system based on weight adjustment according to claim 18, characterized in that: The positioning sensing module receives the plurality of satellite positioning data transmitted by a plurality of positioning satellites, the plurality of satellite positioning data respectively corresponding to the plurality of positioning satellites, and the magnitude of the precision reduction factor value is at least related to a number of the plurality of positioning satellites, a distribution position of the plurality of positioning satellites relative to the vehicle, and a distance of the plurality of positioning satellites relative to the vehicle.

20. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The weight adjustment module is used to: when the image confidence is greater than or equal to an image confidence threshold, adjust the driving image weight value to 100%, and adjust the positioning data weight value and the inertial data weight value to 0 respectively; when the image confidence is less than an image confidence threshold and the positioning confidence is greater than or equal to a positioning confidence threshold, adjust the driving image weight value to less than 100%, and adjust the positioning data weight value and the inertial data weight value to greater than 0 and less than the driving image weight value respectively; when the image confidence is less than an image confidence threshold and the positioning confidence is less than a positioning confidence threshold, adjust the driving image weight value to less than 100%, adjust the positioning data weight value to 0, and adjust the inertial data weight value to greater than 0 and less than the driving image weight value.

21. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The lane line confidence module is used to detect a lane line feature point set of the plurality of driving images, generate a lane line feature point distribution, and analyze a lane continuity and a lane line feature according to the lane line feature point distribution to generate the lane line confidence.

22. The vehicle positioning system based on weight adjustment according to claim 21, characterized in that: The positioning system further includes a storage module, the positioning adjustment module is used to match the lane line feature with a vector map to generate a vehicle image coordinate value, wherein the vector map is stored in the storage module or obtained from a mapping module set in a cloud via a wireless receiving module set in the vehicle; when the positioning data weight value is not equal to 0, the positioning adjustment module generates a positioning data coordinate value according to the plurality of positioning data; when the inertial data weight value is not equal to 0, the positioning adjustment module generates an inertial data coordinate value according to the original positioning coordinate value and an inertial measurement data measured by an inertial measurement module set in the vehicle; the positioning adjustment module then generates the final positioning coordinate value, wherein the final positioning coordinate value=((driving image coordinate value×driving image weight value)+(positioning data coordinate value×positioning data weight value)+(inertial data coordinate value×inertial data weight value)), wherein the inertial data weight value=(100%-the driving image weight value-the positioning data weight value).

23. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The positioning adjustment module continuously generates a positioning failure signal and continuously generates the final positioning coordinate value when the image confidence is continuously less than an image confidence threshold and the positioning confidence is continuously less than a positioning confidence threshold within a time interval.

24. The vehicle positioning system based on weight adjustment according to claim 15, characterized in that: The positioning adjustment module generates a warning signal when the image confidence level is continuously less than an image confidence threshold and the positioning confidence level is continuously greater than or equal to a positioning confidence threshold within a time interval.

Images