Auxiliary positioning method and auxiliary positioning system for vehicle
By setting markers on the roads in the mining area and using a combined auxiliary positioning system of perception modules and evaluation modules, the problem of low positioning accuracy of unmanned vehicles in the mining area is solved, precise positioning is achieved in a weak signal environment, and operating efficiency is improved.
Patent Information
- Application Number
- CN202310673907.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-07
- Publication Date
- 2024-12-10
AI Technical Summary
In mining areas, especially on roads on open-pit deep pit mines and on the side of the mountain, GPS and RTK signals are blocked, and the traditional positioning method has limited effect. There are cumulative errors in IMU combined navigation, which is difficult to achieve SLAM and VSLAM. High-precision map assisted positioning costs are high and The need for frequent updates has led to huge challenges in the positioning system of driverless vehicles.
By pre-setting multiple markers on the roads in the mining area, using the perception module to detect marker information, combining the predetermined actual shape and geographical information, correcting the relative position information, determining the vehicle position, and using a combined auxiliary positioning system of perception module and evaluation module , achieve accurate positioning.
In the case of weak GPS signal or loss of RTK signal, improve the positioning accuracy of unmanned vehicles, avoid fixed navigation solution failure, and improve operational efficiency.
Smart Images
Figure CN119105056A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an auxiliary positioning method and an auxiliary positioning system for vehicles, in particular unmanned vehicles, such as unmanned mining trucks, wherein the auxiliary positioning method and the auxiliary positioning system are used to determine the vehicle position information by using a plurality of markers pre-set on a section of a mining road, wherein the actual shape information and real geographic information of each of the markers are pre-determined and stored. Background Art
[0002] As we all know, the positioning system plays a key role in ensuring the safe operation of unmanned driving, and it is the basis for the stable operation of the entire system. However, in mining areas, especially in open-pit deep pits or on roads on the side of mountains, due to the small number of characteristic markers and the high similarity of topographic contours, and the fact that some roads are blocked by deep pits or the side of mountains, the GPS (Global Positioning System) signal and RTK (Real Time Kinematic) signal are very poor, so traditional positioning methods have their own limitations. For example, GPS signals are severely blocked on winding deep pit roads. The CORS (Continuously Operating Reference Stations) signal of RTK positioning technology has little coverage in mining areas, and the cost of self-built base stations is high, and the coverage rate of a single base station is low, so it is not easy to deploy. In the IMU (Inertial Measurement Unit) combined navigation solution, due to the bumpy roads in the mining area, the deployment effect of a simple IMU combined navigation is limited, and there is also a certain cumulative error. In addition, due to the lack of characteristic landforms in the mining area, SLAM (Simultaneous Localization and Mapping) and VSLAM (Visual Simultaneous Localization and Mapping) technologies are difficult to implement. As for high-precision map-assisted positioning, since the terrain of the mining area changes with the progress of mining, this technology requires frequent map updates, which is a complex process and high cost. Therefore, this series of problems has brought huge challenges to the positioning system of unmanned vehicles in mining areas, especially unmanned mining trucks. Summary of the invention
[0003] The object of the present invention is to provide an auxiliary positioning method and an auxiliary positioning system for a vehicle, especially an unmanned vehicle, such as an unmanned mining truck, by which one or more disadvantages of the above-mentioned prior art are at least partially overcome.
[0004] The object is achieved by an assisted positioning method and an assisted positioning system having the following characteristics.
[0005] According to one aspect of the present invention, an auxiliary positioning method for vehicles, in particular unmanned vehicles, such as unmanned mining trucks, is proposed, wherein the auxiliary positioning method is used to determine the vehicle position information of the vehicle by using a plurality of markers pre-set on a section of a mining road, wherein the actual shape information and real geographic information of each marker are predetermined and saved, wherein the auxiliary positioning method comprises the following steps: detecting a plurality of markers on the section to obtain marker information, wherein the marker information comprises shape measurement information of the corresponding marker and its relative position information relative to the vehicle; obtaining predetermined actual shape information and real geographic information of the detected marker; comparing the shape measurement information of the corresponding marker with the actual shape information to determine a corresponding perception deviation for the corresponding marker, and correcting the relative position information based on the corresponding perception deviation; and determining the vehicle position information based on the corrected relative position information and the real geographic information.
[0006] According to another aspect of the present invention, an auxiliary positioning system for a vehicle, in particular an unmanned vehicle, such as an unmanned mining truck, is proposed, wherein the auxiliary positioning system is used to determine the vehicle position information of the vehicle by using a plurality of markers pre-set on a section of a mining road, wherein the actual shape information and real geographic information of each marker are predetermined and saved, and the auxiliary positioning system comprises a perception module and an evaluation module, and the auxiliary positioning system is particularly configured to execute the auxiliary positioning method according to the present invention, wherein the perception module is configured to detect one or more of the markers on the section to obtain marker information, wherein the marker information comprises shape measurement information of the corresponding marker and relative position information of the marker relative to the vehicle, and the evaluation module is configured to obtain the predetermined actual shape information and real geographic information of the detected marker; compare the shape measurement information of the corresponding marker with the actual shape information to determine the corresponding perception deviation of the perception module for the corresponding marker, and correct the relative position information based on the corresponding perception deviation; and determine the vehicle position information based on the corrected relative position information and the real geographic information.
[0007] Through the auxiliary positioning method and auxiliary positioning system of the present invention, the vehicle can be accurately positioned on mining roads where the GPS signal is weak or the RTK signal age is too long or even the GPS or RTK signal is lost, thereby improving the positioning accuracy of unmanned vehicles such as unmanned mining trucks when the GPS signal or RTK signal is lost when operating in deep pit open-pit mines, avoiding the situation where the unmanned vehicle stops running after the traditional combined navigation fixed solution fails, thereby also improving the operating efficiency of unmanned vehicles in such road conditions.
[0008] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the scope of protection of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The present invention will be described in detail below with reference to embodiments in conjunction with the accompanying drawings. In the accompanying drawings:
[0010] Figure 1 An exemplary embodiment of the auxiliary positioning system according to the present invention is schematically shown.
[0011] Figure 2 A flow chart of an exemplary embodiment of the auxiliary positioning method according to the present invention is schematically shown.
[0012] Figure 3 The position conversion relationship between the plane coordinate system and the vehicle coordinate system is schematically shown. DETAILED DESCRIPTION
[0013] Figure 1 An exemplary embodiment of an auxiliary positioning system 100 for a vehicle 1 on a mining road, especially an unmanned vehicle, such as an unmanned mining truck, according to the present invention is schematically shown. The auxiliary positioning system 100 is used to determine the vehicle position information on sections 3 of the mining road, especially on those sections 3 where it is determined that the positioning accuracy of the main positioning system 200 currently deployed by the vehicle 1 cannot meet the requirements due to, for example, loss of GPS signals or too long RTK differential age. The main positioning system generally includes one or more of a GPS system, an RTK system, and an IMU system. The vehicle position information may, for example, include the position coordinates (x y ... v ,y v ) and the heading angle θ of vehicle 1, that is, the angle between the direction pointed by the front of the vehicle and the x-axis.
[0014] The auxiliary positioning system 100 includes a perception module 10 installed on the vehicle 1, such as a laser radar, a millimeter wave radar, and a camera. The perception module 10 is configured to detect the markers 2 on the road section 3 to obtain the marker information. The markers 2 have been deployed in advance and placed on one side or both sides of the road section 3. The markers 2 usually have a regular shape, and their material, color, and shape are easy to be detected and identified by the perception module 10. For example, the marker 2 can be configured as a cylinder with a regular cross section, such as a cylinder, a semi-cylinder, or a right prism with a regular cross section, especially a regular triangular prism, a cube, or a cuboid. The markers 2 are distributed on the road section 3 so that when the vehicle 1 is at any position on the road section 3, there are a sufficient number of markers 2 (for example, at least three) within the perception range of the perception module 10 to achieve vehicle positioning. That is, the markers 2 are arranged so that the layout of all the markers 2 can cover the entire road section 3. Each marker 2 can be provided with a mark that can be detected by the perception module 10 or other suitable sensors on the vehicle 1. The identification may be, for example, a number or code directly written on the surface of the marker, or other suitable marks associated with the number. The number may include, for example, numbers or letters or a combination thereof. The identification may also be a number plate containing a number that is fixed, for example, welded or bonded to the marker. The identification may be set on the top surface or side of the marker 2 where it is easily detected by the perception module 10. The real geographic information and actual shape information of each marker 2 may be predetermined and stored in the database 4 in association with the number of the marker 2. Therefore, the matching real geographic information and actual shape information may be obtained from the database 4 through the number of the marker 2. The database 4 may be a system library of a vehicle system on the vehicle 1, or it may be a database outside the vehicle, such as a database arranged on a background or cloud server. The real geographic information of the marker 2 may be obtained, for example, by surveying and mapping technology with the help of a surveying map. The real geographic information may include the geographic coordinates of the center point or central longitudinal axis M of the marker, such as the position coordinates (x M ,y M). The actual shape information of the marker 2 can be obtained through on-site field measurement after it is placed on the road section. Depending on the geometric shape of the marker, the actual shape information may include the characteristic dimensions of the marker. For example, when the marker is a cylinder, the actual shape information may include the height and diameter or radius of the marker. When the marker is a cuboid or a cube, the actual shape information may include the height, length and width of the marker. When the marker is a regular triangular prism, the marker may include the height and side length. Thus, when the geographic coordinates of the center point or the central longitudinal axis of the marker are known, the geographic coordinates of any point on the surface of the marker can be calculated through the shape and size information of the marker. In particular, the height of the marker is designed not to exceed the height of the detector of the perception module 10 installed on the vehicle, so that the perception module 10 can detect the entire top surface of the marker.
[0015] The marker information obtained by means of the perception module 10 may include shape measurement information of the marker 2 and relative position information of the marker 2 relative to the vehicle 1. The relative position information of the marker 2 relative to the vehicle 1 may include, for example, the horizontal distance D and direction angle α of the central longitudinal axis M of the marker 2 relative to any point on the vehicle 1, such as the center point V, or the position coordinates (x' -y') of the central longitudinal axis M of the marker 2 in the vehicle coordinate system x'-y' fixed relative to the vehicle 1. M ,y' M ). Here, the distance D and the direction angle α are related to the coordinate x' M ,y' M They can be simply converted to each other in the vehicle coordinate system through trigonometric function relationships. In this example, without loss of generality, it is assumed that the vehicle coordinate system x'-y' takes the center point V of the vehicle 1 as the origin, and the vehicle longitudinal axis direction (i.e., the vehicle driving direction or the direction pointed by the front of the vehicle) and the vehicle transverse axis direction perpendicular to the vehicle longitudinal axis direction are the x' axis and y' axis respectively. Since the vehicle's outer dimensions are known, the position and direction of the detector relative to the center point V of the vehicle 1 are known, that is, the coordinates of the detector of the perception module 10 in the vehicle coordinate system are known. Therefore, the measured position and direction of the central longitudinal axis M of the marker 2 relative to the detector of the perception module 10 on the vehicle 1 can be easily converted into the relative position information of the central longitudinal axis M of the marker 2 relative to the center point V of the vehicle 1 or relative to any specific point on the vehicle 1 in the vehicle coordinate system through trigonometric function relationships. Corresponding to the actual shape information, depending on the shape of the marker 2, the shape measurement information of the marker 2 may include the characteristic dimensions of the marker, such as height, diameter or radius, length, width, or side length.
[0016] The auxiliary positioning system 100 also includes an evaluation module 20, which is configured to determine the vehicle position information of the vehicle 1 based on the marker information of the marker, the predetermined real geographic information and the actual shape information, and, if necessary, the vehicle position information determined by the main positioning system 200. To this end, the evaluation module 20 is configured to: obtain the predetermined actual shape information and the real geographic information of the detected marker 2; compare the shape measurement information in the marker information of the corresponding marker with the stored actual shape information to determine the corresponding perception deviation of the perception module 10 for the corresponding marker, and correct the relative position information in the marker information based on the corresponding perception deviation; and determine the vehicle position information based on the corrected relative position information and the real geographic information of the marker 2. In particular, the evaluation module 20 is configured to use the above-mentioned vehicle position information determined based on the marker information as the first position information, and the vehicle position information of the vehicle 1 determined by the main positioning system 200 as the second position information, and based on the first position information and the second position information, calculate the third position information as the vehicle position information of the vehicle 1 by a weighted method. In this regard, the following will also be combined with Figure 2 An embodiment of the auxiliary positioning method is described in detail.
[0017] Figure 2 Schematically showing the Figure 1 Flow chart of an embodiment of the auxiliary positioning method of the present invention in which the auxiliary positioning system 100 determines the vehicle position information on the road section 3 of the mining area road. Before the method of the present invention starts, as a preparatory work, the markers 2 must first be deployed and placed on the road section 3 of the mining area road, especially on the road section 3 where the positioning accuracy of the main positioning system 200 of the vehicle 1 does not meet the requirements. To this end, the vehicle 1 runs on the mining area road based on the currently installed main positioning system 200 and collects a series of data to determine on which sections the GPS signal will be lost or the RTK differential age is too long, so as to identify the road sections 3 where markers need to be set. After identifying the road sections 3 where markers need to be set, a plurality of markers with regular shapes that can be detected by the perception module 10 of the auxiliary positioning system 100 of the vehicle 1 are placed on these road sections 3. Common markers can be in the form of cylinders or cubes, for example. Then, the real geographic information and actual shape information of each marker 2 are measured, and the real geographic information and actual shape information are stored in the database 4 together with the number of the marker 2.
[0018] When the vehicle 1 is driving on the road in the mining area, when it drives to the road section 3 where the GPS signal is weak and the RTK signal age is too long, the main positioning system 200 detects that the confidence of its own positioning is low. At this time, the auxiliary positioning system 100 is started to implement the auxiliary positioning method according to the present invention. In the auxiliary positioning method, in step S1, multiple markers 2 on the road section 3 are detected and identified to obtain marker information. The marker information may include shape measurement information of the marker and relative position information of the marker relative to the vehicle. The marker information can be detected by the perception module 10 of the auxiliary positioning system 100. The perception module 10 can be configured as a laser radar or include a laser radar. The laser radar can detect the distance, orientation and shape of the marker 2 from the vehicle 1. In particular, the laser radar can image a three-dimensional image of the marker 2. Taking a cylinder as an example, the shape measurement information may include, for example, the height and diameter or radius of the marker. The relative position information may include, for example, the horizontal distance D of the central longitudinal axis M of the marker 2 relative to the central point V of the vehicle 1 (i.e., the origin of the vehicle coordinate system) and the direction angle α, which is the angle between the line between the central longitudinal axis M and the central point V and the x' axis of the vehicle coordinate system. Figure 3 As shown, the distance D and the direction angle α can be easily converted into the position coordinates (x') of the central longitudinal axis M of the marker 2 in the vehicle coordinate system with the center point V of the vehicle 1 as the origin through the trigonometric function relationship. M ,y' M ):
[0019] x' M =D*cosα,y' M =D*sinα.
[0020] The marker information may also be determined by the evaluation module 20 instead of the perception module 10. For example, the perception module 10 transmits the detected three-dimensional image of the marker 2 to the evaluation module 20, and the evaluation module determines the position of the center point or the central longitudinal axis M of the marker 2 and the external dimension information based on the three-dimensional image using a known algorithm.
[0021] In step S2, the predetermined actual shape information and real geographic information of the detected marker 2 are obtained. Here, the number of the marker can be obtained by detecting and identifying the logo on the marker 2, and then the actual shape information and real geographic information associated with the number pre-stored there are obtained from the database 4 according to the number. The logo of the marker 2 can be detected by the perception module 10. Alternatively, the logo of the marker 2 can also be detected by other suitable sensors on the vehicle 1, such as a camera.
[0022] In another embodiment, if the number cannot be detected, the position information of the marker 2 can be estimated based on the relative position information of the marker 2 relative to the vehicle 1 detected in step S1 and the possibly inaccurate vehicle position information determined by the main positioning system 100, and the actual shape information and real geographic information of the matching marker 2 are traversed from the database 4 according to the estimated position information of the marker 2. To this end, the estimated position information is compared with the real geographic information, and the marker with the closest real geographic information is the matching marker.
[0023] In step S3, the shape measurement information of the corresponding marker is compared with the actual shape information to determine the corresponding perception deviation for the corresponding marker, and the relative position information is corrected based on the corresponding perception deviation. For example, taking a cylinder as an example, it is assumed that the quotient of the perceived distance deviation Δd and the sensed distance D is proportional to the quotient of the perceived radius deviation Δr and the sensed radius r, that is:
[0024] Δd / D=Δr / r,Δd=D c -D, Δr = r0 – r,
[0025] Among them, D c is the corrected distance, r0 is the actual radius predetermined and stored in the database, then the perceived distance deviation Δd can be obtained as
[0026] Δd=D / r*(r0-r)=D*(r0 / r-1).
[0027] So, the corrected distance is D c =D+Δd=D*r0 / r.
[0028] In another embodiment, in the case where the marker is a cuboid, instead of considering only the deviation of the radius or diameter as in the case of a cylinder, the deviations in both the length and width directions are considered. To this end, the measured length L and width W in the cross section perpendicular to the height are compared with the predetermined actual length L0 and actual width W0 of the marker, respectively, and the measured length L and width W are corrected according to the perceived length deviation ΔL (= L0-L) and the perceived width deviation ΔW (= W0-W). Therefore, the shape of the corrected cross section is different from the shape of the measured cross section, and therefore the position of the center point M of the cross section (i.e., the central longitudinal axis of the cuboid) is also changed. Then, the corrected distance D is calculated based on the position of the changed center point of the cross section. c and direction angle.
[0029] In summary, based on the comparison between the measured shape information and the actual shape information of the cross-section, the shape and size of the cross-section are corrected to obtain the corrected position of the central longitudinal axis of the marker, and the distance and direction angle of the vehicle relative to the marker are recalculated based on the corrected position of the central longitudinal axis.
[0030] In step S4, the vehicle position information is determined based on the corrected relative position information of the landmark and the real geographic information. Here, the pitch and roll of the vehicle are determined by the IMU positioning system, and the height information is not considered, so the vehicle position information may only include the plane position coordinates x of the center point V of the vehicle 1 v ,y v and heading angle θ. Based on the vehicle position information (x v ,y v , θ) and the corrected relative position information (D c , α), the position information of the marker 2 in the plane coordinate system can be obtained through the rotation and translation of the coordinate system, which should be equal to the real geographic information of the marker 2. Therefore, the x-axis is established. v ,y v , θ are the unknown quantities. For example, Figure 3 In a simple case shown in , it is assumed that the pitch and roll of the vehicle are both zero, and the elevation information is not considered, and the problem is simplified to a two-dimensional plane problem. Figure 3 It can be clearly seen that the vehicle coordinate system x'-y' rotates a heading angle θ relative to the plane coordinate system xy, and the origins of the two coordinate systems are translated by a distance, that is, the plane coordinates of the center point V of the vehicle 1 (x v ,y v ). Therefore, the vehicle position information (x v ,y v , θ) and the relative position information (D c , α), the position information of the marker 2 in the plane coordinate system can be calculated, which should be equal to the real geographic information of the marker 2 (x M ,y M ),Right now
[0031] x v +D c *cos(α+θ)=x M ,y v +D c *sin(α+θ)=y M .
[0032] Since there are three unknown quantities x v ,y vand heading angle θ, so the equations constructed using the information of the three markers 2 can be solved to obtain the values of these unknown quantities. From the equations, we can see that for x v and v , only two sets of parameters are needed to solve them separately, but the solutions may be two values. In order to obtain the exact solution, the third set of equations is introduced. An intuitive explanation for this is that with the central longitudinal axis M of each marker 2 as the center, the distance D between the center point V of the vehicle 1 and the corresponding central longitudinal axis M is c Draw a circle with radius , and the coordinates of the intersection of these three circles are the position coordinates of the center point V in the plane coordinate system (x v ,y v ). Here, two markers are not enough to uniquely determine the position of the vehicle in any case, because the two circles generally have two intersection points except for the tangent case, so a third marker must be introduced to determine the unique solution.
[0033] In another embodiment, the confidence of the detection results of the perception module 10 of the auxiliary positioning system 100 and the main positioning system 200 is further considered to more accurately determine the position information of the vehicle. To this end, in step S5, the vehicle position information (x v ,y v ) is used as the first position information (x1, y1), and the position information of the vehicle determined by the main positioning system 200 is used as the second position information (x2, y2), and the third position information (x3, y3) is calculated by a weighted method based on the first position information and the second position information as the vehicle position information of the vehicle.
[0034] In a simple implementation, the confidence of the detection results of the perception module 10 of the auxiliary positioning system 100 and the main positioning system 200 can be considered, and the first position information (x1, y1) and the second position information (x2, y2) can be weighted by weighting coefficients a1 and a2 respectively to calculate the third position information (x3, y3), that is,
[0035] x3=x1*a1+x2*a2, y3=y1*a1+y2*a2,
[0036] Wherein, a1 and a2 are real numbers greater than or equal to zero and less than or equal to 1, and a1+a2=1.
[0037] In another embodiment of the present invention, instead of directly weighting the vehicle position information, the perceived deviation and the horizontal position deviation and vertical position deviation of the vehicle are weighted. To this end, in the aforementioned step S3, the confidence of the sensed marker information is considered, the perceived deviation is weighted using the first weighting coefficient q1, and the weighted perceived deviation is used to correct the relative position information. In the example given above, that is, the perceived distance deviation Δd is weighted using the first weighting coefficient q1 to obtain the corrected distance D' between the marker 2 and the vehicle 1 c :
[0038] D' c =D+q1*Δd=D+q1*D*(r0 / r-1)=D*(1-q1+q1*r0 / r).
[0039] Then, in the aforementioned step S4, based on the corrected relative position information (D') of the corresponding markers c , α) and the real geographic information of the corresponding landmark to calculate the first position information (x1, y1). Then, in step S5, the position deviation between the first position information (x1, y1) and the second position information (x2, y2) determined by the main positioning system 200, that is, the horizontal position deviation Δx = x1-x2 and the vertical position deviation Δy = y1-y2 are calculated, and considering the confidence of the second position information, the horizontal and vertical position deviations Δx and Δy are weighted using the second weighting coefficient q2, and the weighted position deviations q2*Δx and q2*Δy are used to correct the second position information (x2, y2), that is,
[0040] x3'=x2+q2*Δx, y3'=y2+q2*Δy.
[0041] The first and second weighting coefficients q1 and q2 are real numbers greater than or equal to zero and less than or equal to 1, respectively, and they can be configured through a map diagram (or error weight allocation table) obtained in advance from experimental data. For example, the duration of the weak GPS signal can be used to estimate the size of the deviation through experience, thereby giving the corresponding values of the first and second weighting coefficients q1 and q2. That is, it is necessary to consider the confidence of the detection results of the perception module 10 and the confidence of the detection results of the main positioning system 200, such as the GPS system, and assign certain weights q1 and q2 respectively. That is, consider which of the perception module detection and the GPS detection is more credible, and what proportion each accounts for in the final result.
[0042] In general, the present invention describes how to accurately determine the position of a vehicle on a road section where the positioning accuracy of conventional positioning systems is not sufficient.
Claims
1. An auxiliary positioning method for a vehicle, in particular an unmanned vehicle, such as an unmanned mining truck, the auxiliary positioning method is used to determine the vehicle position information of the vehicle by using a plurality of markers pre-set on a section of a mining road, wherein: The actual shape information and real geographic information of each of the landmarks are predetermined and stored, and the auxiliary positioning method comprises the following steps: detecting one or more of the markers to obtain marker information, the marker information including shape measurement information of the corresponding marker and relative position information thereof with respect to the vehicle; Obtaining predetermined actual shape information and real geographic information of the detected landmark; comparing the shape measurement information of the corresponding marker with the actual shape information to determine a corresponding perceptual deviation for the corresponding marker, and correcting the relative position information based on the corresponding perceptual deviation; and The vehicle position information is determined based on the corrected relative position information and the real geographic information.
2. The auxiliary positioning method according to claim 1, characterized in that: Correcting the relative position information based on the corresponding perceptual deviation includes: considering the confidence of the landmark information, weighting the perceptual deviation using a first weighting coefficient, and correcting the relative position information using the weighted perceptual deviation.
3. The auxiliary positioning method according to claim 2, characterized in that: The vehicle has a main positioning system, and the auxiliary positioning method also includes: using the vehicle position information determined based on the corrected relative position information and the real geographic information as the first position information, determining the position deviation between the first position information and the second position information of the vehicle determined by the main positioning system of the vehicle, considering the confidence of the second position information, using a second weighting coefficient to weight the position deviation, and using the weighted position deviation to correct the second position information, and outputting the corrected second position information as the vehicle position information.
4. The auxiliary positioning method according to claim 3, characterized in that: The first weighting factor and the second weighting factor are given by a map that is predetermined, in particular by experimental data.
5. The auxiliary positioning method according to any one of claims 1 to 4, characterized in that: The predetermined actual shape information and the real geographic information of the landmark are stored in a database in association with the landmark number. In particular, the database is included in the vehicle system or is provided on a server outside the vehicle.
6. The auxiliary positioning method according to claim 5, characterized in that: An identification on the marker associated with the number of the marker is detected, and based on the identification, predetermined actual shape information and real geographic information of the marker that matches the number is searched from the database.
7. The auxiliary positioning method according to any one of claims 1 to 6, characterized in that: The multiple markers are arranged on one side or both sides of the road section, and / or the multiple markers are distributed on the road section so that when the vehicle is at any position on the road section, there are at least three markers within the perception range of the perception module.
8. The auxiliary positioning method according to any one of claims 1 to 7, characterized in that: The marker has a regular shape, in particular, the marker is a column with a regular cross section, such as a cylinder, a regular triangular prism, a cuboid or a cube.
9. The auxiliary positioning method according to any one of claims 1 to 8, characterized in that: Depending on the shape of the marker, the actual shape information and the shape measurement information respectively include characteristic dimensions of the marker, such as height, diameter, radius, length, width or side length.
10. The auxiliary positioning method according to any one of claims 1 to 9, characterized in that: The relative position information includes the distance of the central longitudinal axis of the marker relative to the center point of the vehicle and the direction angle of the central longitudinal axis of the marker relative to the longitudinal axis of the vehicle, or includes the relative position coordinates of the central longitudinal axis of the marker relative to the center point of the vehicle.
11. The auxiliary positioning method according to any one of claims 1 to 10, characterized in that: The vehicle position information includes the position coordinates of the center point of the vehicle and the yaw angle of the vehicle.
12. The auxiliary positioning method according to any one of claims 1 to 11, characterized in that: Three landmarks are used to determine vehicle position information based on the corrected relative position information and the real geographic information.
13. The auxiliary positioning method according to any one of claims 1 to 12, characterized in that: A perception module is used to detect the multiple markers to obtain marker information, and the perception module includes a laser radar, and / or a millimeter wave radar, and / or a camera.
14. The auxiliary positioning method according to any one of claims 1 to 13, characterized in that: The vehicle is provided with a main positioning system, which includes a GPS system, and / or an RTK system, and / or an IMU system.
15. An auxiliary positioning system for a vehicle, in particular an unmanned vehicle, such as an unmanned mining truck, the auxiliary positioning system being used to determine the vehicle position information of the vehicle by using a plurality of markers pre-set on a section of a mining road, wherein: The actual shape information and real geographic information of each of the landmarks are predetermined and stored, the auxiliary positioning system comprises a perception module and an evaluation module, and the auxiliary positioning system is particularly configured to execute the auxiliary positioning method according to any one of claims 1 to 14, It is characterized in that The perception module is configured to detect one or more of the markers to obtain marker information, wherein the marker information includes shape measurement information of the corresponding marker and relative position information of the marker relative to the vehicle. The evaluation module is configured to obtain predetermined actual shape information and real geographic information of the detected marker; compare the shape measurement information of the corresponding marker with the actual shape information to determine the corresponding perception deviation of the perception module for the corresponding marker, and correct the relative position information based on the corresponding perception deviation; and determine the vehicle position information based on the corrected relative position information and the real geographic information.