Long-distance moving path planning method and device for lunar rover

Abstract

The invention provides a long-distance moving path planning method and device for a lunar rover, and the method comprises the steps: carrying out the classification of topographic features based on the whole-moon typical landform features through the wide-angle camera image of a lunar investigation orbiter; dividing the to-be-planned moving path into a plurality of local passable areas based on the classified passable areas; for each local passable area, generating a digital elevation model and a digital orthoimage by using a narrow-angle camera image of a moon investigation orbiter; performing global path planning to generate a global reference path; constructing a local digital elevation model and a local digital orthoimage by using a three-dimensional lunar surface image sensed by the lunar rover in real time; and according to the local digital elevation model, the local digital orthoimage, the measured data of the lunar rover and the prior lunar soil mechanical parameters, verifying the global reference path based on the kinetic model, and outputting a global planning strategy. The method can be used for long-distance moving path planning of the lunar rover, and is high in accuracy.

Description

Technical Field

[0001] The present invention relates to the fields of image processing, path planning, simulation prediction and software development, and in particular to a method and device for long-distance movement path planning of a lunar rover. Background Art

[0002] As human beings continue to explore the moon, with the goal of developing lunar resources and establishing a lunar scientific research station, unmanned or manned lunar rovers will conduct large-scale, long-distance patrol exploration, material transportation, and scientific experiments on the lunar surface. Mobile exploration will be the main mode of transportation on the lunar surface. The accessibility and safety of the lunar rover's mobile path are the key to ensuring the reliable and efficient driving of the lunar rover, and are the basis of all deep space exploration missions. At present, the single planned distance of the lunar rover is about ten meters. In the future, the single driving range of the lunar rover will inevitably span hundreds of meters, kilometers, or even tens of kilometers. The patrol exploration range may span the entire lunar surface. However, the terrain and other environments on the lunar surface are complex. Under the premise of meeting the obstacle-crossing ability of the lunar rover, it has become an urgent need for current lunar exploration to carry out long-distance mobile path planning for the lunar rover. The accessibility and safety of the lunar rover's mobile path refer to the requirement that the lunar rover's path from the starting point to the target point is the shortest, and the lunar rover can safely avoid obstacles during driving, taking into account various environmental factors and the mechanical properties of the lunar rover itself.

[0003] Currently, there is a lack of a solution for long-distance movement path planning for the lunar rover to ensure that the path from the starting point to the target point is the shortest and that obstacles can be safely avoided during driving. Summary of the invention

[0004] The embodiment of the present invention provides a method for long-distance movement path planning of a lunar rover, which is used for long-distance movement path planning of a lunar rover with high accuracy. The method includes:

[0005] Based on the typical landform features of the whole moon, the Lunar Reconnaissance Orbiter wide-angle camera images are used to classify the terrain features, and the inaccessible and accessible areas are marked based on the classification results;

[0006] Based on the passable area, the moving path to be planned is divided into multiple local passable areas;

[0007] For each local traversable area, digital elevation models and digital orthophotos were generated using Lunar Reconnaissance Orbiter narrow-angle camera images;

[0008] Use digital elevation models and digital orthophotos to perform global path planning, determine the optimal path based on the cost function, and generate a global reference path;

[0009] Use the three-dimensional lunar surface images perceived in real time by the lunar rover to build local digital elevation models and local digital orthophotos;

[0010] According to the local digital elevation model and local digital orthophoto, the measured data of the lunar rover and the prior mechanical parameters of the lunar soil, the global reference path is verified based on the dynamic model and the global planning strategy is output.

[0011] The embodiment of the present invention provides a long-distance moving path planning device for a lunar rover, which is used for long-distance moving path planning of a lunar rover with high accuracy. The device includes:

[0012] The terrain classification module is used to classify the terrain features based on the typical landform features of the whole moon and the images of the Lunar Reconnaissance Orbiter Wide-angle Camera, and mark the inaccessible and accessible areas based on the classification results;

[0013] A path division module is used to divide the moving path to be planned into multiple local passable areas based on the passable area;

[0014] The model generation module is used to generate digital elevation models and digital orthophotos for each local traversable area using the Lunar Reconnaissance Orbiter narrow-angle camera images;

[0015] A global reference path generation module is used to perform global path planning using a digital elevation model and digital orthophotos, determine the optimal path based on a cost function, and generate a global reference path;

[0016] The local model building module is used to build a local digital elevation model and local digital orthophoto using the stereoscopic lunar surface images perceived by the lunar rover in real time;

[0017] The global planning strategy output module is used to verify the global reference path based on the dynamic model according to the local digital elevation model and local digital orthophoto, the lunar rover's measured data and the prior lunar soil mechanical parameters, and output the global planning strategy.

[0018] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned lunar rover long-distance movement path planning method when executing the computer program.

[0019] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned lunar rover long-distance movement path planning method is implemented.

[0020] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned lunar rover long-distance movement path planning method is implemented.

[0021] In an embodiment of the present invention, based on typical landform features of the whole moon, the wide-angle camera image of the lunar reconnaissance orbiter is used to classify the terrain features, and the inaccessible area and the traversable area are marked based on the classification results; based on the traversable area, the moving path to be planned is divided into multiple local traversable areas; for each local traversable area, the narrow-angle camera image of the lunar reconnaissance orbiter is used to generate a digital elevation model and a digital orthophoto; the digital elevation model and the digital orthophoto are used to perform global path planning, and the optimal path is determined according to the cost function to generate a global reference path; the stereoscopic lunar surface image perceived in real time by the lunar rover is used to construct a local digital elevation model and a local digital orthophoto; according to the local digital elevation model and the local digital orthophoto, the measured data of the lunar rover and the prior mechanical parameters of the lunar soil, the global reference path is verified based on the dynamic model, and the global planning strategy is output. In the above process, the path planning method is combined with global reference path planning based on prior information (including image information and lunar soil mechanical parameters, where the image information includes lunar reconnaissance orbiter wide-angle camera images and lunar reconnaissance orbiter narrow-angle camera images) and local planning based on the three-dimensional lunar surface images perceived by the lunar rover in real time. This can effectively solve the current situation that obstacles cannot be accurately identified when using low-resolution historical images for path planning, and the path planning cannot support long-distance movement using the lunar rover's real-time perception images. The lunar soil mechanical parameters are identified on-orbit using measured data, and the global reference path is verified based on the dynamic model. The output global planning strategy is more accurate, which will further improve the reliability of the lunar rover's long-distance movement planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0023] Figure 1 This is a flow chart of a method for planning a long-distance movement path of a lunar rover in an embodiment of the present invention;

[0024] Figure 2 Schematic diagram of four types of terrain defined in the embodiment of the present invention;

[0025] Figure 3 It is a schematic diagram of marking inaccessible areas on the full-month topographic map in an embodiment of the present invention;

[0026] Figure 4 It is a schematic diagram of marking the moving direction in sections in the passable area in an embodiment of the present invention;

[0027] Figure 5 The method flow of using LRO NAC images to produce digital elevation models and digital orthophotos in a local area in an embodiment of the present invention;

[0028] Figure 6 This is a method flow for performing LRO NAC image optimization in an embodiment of the present invention;

[0029] Figure 7 A method for calculating an incident angle factor, a stereo intersection angle factor and a shadow top distance factor provided in an embodiment of the present invention;

[0030] Figure 8 A method for calculating the conversion relationship between an image space coordinate system and a star-fixed coordinate system provided in an embodiment of the present invention;

[0031] Fig. 9 The present invention provides a method flow for global path planning using a digital elevation model and a digital orthophoto;

[0032] Fig.10 A method for calculating slope provided by the present invention;

[0033] Fig.11 This is a method flow for verifying a global reference path using a three-dimensional lunar surface image perceived in real time by a lunar rover in an embodiment of the present invention;

[0034] Fig.12 The process of calculating the internal friction angle of lunar soil in an embodiment of the present invention;

[0035] Fig.13 The schematic diagram of calculating the internal friction angle of lunar soil in an embodiment of the present invention is as follows;

[0036] Fig.14 A schematic diagram of displaying the positions of the lunar rover and wheels on a local digital orthophoto in an embodiment of the present invention;

[0037] Fig.15 The process of calculating the wheel-ground interaction force and torque of each wheel in an embodiment of the present invention;

[0038] Fig.16 The process of calculating the position and speed of the lunar rover body and each wheel in an embodiment of the present invention;

[0039] Fig.17 Schematic diagram of a long-distance moving path planning device for a lunar rover in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0041] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0042] The inventors found that due to the delay in transmission between the Earth and the Moon and the resource allocation on the lunar rover, the lunar rover's ability to plan autonomously is very limited. The lunar surface topography, lighting conditions, etc. are complex, changeable and unknown. In order to ensure the safety of detection, lunar rovers mostly use a semi-autonomous detection method of "ground remote operation + limited autonomy". Most of the time, ground personnel analyze and judge the surrounding terrain based on limited telemetry and image information, conduct a single path planning and inject control instructions into the lunar rover. Generally, a single moving distance does not exceed 10 meters, the distance is relatively close, and most of the time is spent on information transmission and ground processing. Therefore, the current detection method is inefficient.

[0043] At present, although full-moon mapping products can provide a wide-area perception range, their resolution is generally low. Low-resolution mapping products cannot accurately identify obstacles and cannot directly perform precise path planning. The lunar rover's real-time perception data (stereo camera images) and its derived terrain products have high resolution, but their perception range is limited and cannot support long-distance mobile planning.

[0044] In addition, when the lunar rover moves over long distances and over a large area on the lunar surface, it will experience different geological types. Some soils are soft, some are rough, and some are compacted. The mechanical parameters of soils of different ground types will be inconsistent, and the movement effect of the lunar rover will inevitably be biased. The current simulation verification is mainly based on kinematics, and does not take into account the mechanical properties of the soil, so it is difficult to truly reflect the movement of the lunar rover.

[0045] Therefore, an embodiment of the present invention proposes a method for long-distance movement path planning of a lunar rover, which is used to solve the current situation that obstacles cannot be accurately identified when path planning is performed based on a low-resolution full-moon topographic map, and path planning using the lunar rover's real-time perception of images cannot support long-distance movement.

[0046] The principle of the long-distance movement path planning method for the lunar rover proposed in the present invention is: a long-distance movement path planning method that combines global planning based on prior information with local planning based on real-time perception information, and verifies the global reference path using a dynamic model in combination with the mechanical properties of the lunar soil, in order to further improve the reliability of the long-distance movement planning of the lunar rover.

[0047] Figure 1 The process of the long-distance moving path planning method of the lunar rover in the embodiment of the present invention is as follows Figure 1 ,like Figure 1 As shown, the method includes:

[0048] Step 101, based on typical landform features of the whole moon, using the Lunar Reconnaissance Orbiter wide-angle camera image, classify the landform features, and mark the inaccessible areas and the accessible areas based on the classification results;

[0049] Step 102, based on the traversable area, dividing the moving path to be planned into a plurality of local traversable areas;

[0050] Step 103, for each local traversable area, using the Lunar Reconnaissance Orbiter narrow angle camera image, generate a digital elevation model and a digital orthophoto;

[0051] Step 104, performing global path planning using the digital elevation model and the digital orthophoto, determining the optimal path according to the cost function, and generating a global reference path;

[0052] Step 105, constructing a local digital elevation model and a local digital orthophoto using the three-dimensional lunar surface image perceived in real time by the lunar rover;

[0053] Step 106, verifying the global reference path based on the dynamics model according to the local digital elevation model and the local digital orthophoto, the measured data of the lunar rover and the prior lunar soil mechanical parameters, and outputting the global planning strategy;

[0054] In the embodiment of the present invention, according to the local digital elevation model and local digital orthophoto, the measured data of the lunar rover and the prior mechanical parameters of the lunar soil, a path planning method combining global reference path planning based on prior information (including image information and lunar soil mechanical parameters, wherein the image information includes the lunar reconnaissance orbiter wide-angle camera image and the lunar reconnaissance orbiter narrow-angle camera image) and local planning based on the three-dimensional lunar surface image perceived by the lunar rover in real time is adopted, which can effectively solve the current situation that obstacles cannot be accurately identified when using low-resolution historical images for path planning, and the path planning status that long-distance movement cannot be supported by using the lunar rover's real-time perception image. The lunar soil mechanical parameters are identified on-orbit using the measured data, and the global reference path is verified based on the dynamic model. The output global planning strategy is more accurate, which will further improve the reliability of the lunar rover's long-distance movement planning.

[0055] In step 101, based on typical landform features of the whole moon, the terrain features are classified using the Lunar Reconnaissance Orbiter wide-angle camera image, and the inaccessible areas and the accessible areas are marked based on the classification results;

[0056] The present invention divides the full moon topography into four categories according to the topographic features, namely grabens, wrinkle ridges, impact craters and domes, as shown in the figure below: Figure 2 These four types of terrain features are marked as inaccessible areas during path planning, which can be marked by manual annotation or deep learning; the remaining areas are accessible areas, as shown in the figure below. Figure 3 As shown. Among them, Figure 3 (a) is the original full moon (partial) topographic map. Figure 3 Middle (b) is a full-month (partial) topographic map with inaccessible areas marked.

[0057] In step 102, based on the traversable area, the moving path to be planned is divided into a plurality of local traversable areas;

[0058] In one embodiment, based on the traversable area, the moving path to be planned is divided into a plurality of local traversable areas, including:

[0059] According to the starting position and the end position of the moving path to be planned, the moving direction is marked in sections in the passable area;

[0060] According to the moving direction at the marked position, the moving path is divided into a plurality of local traversable areas.

[0061] In the above embodiment, within the passable area, the impassable area is avoided, and the moving direction is marked in sections according to the starting position and the end position. Figure 4 As shown, the moving path is divided into five local traversable areas.

[0062] In step 103, for each local traversable area, a digital elevation model and a digital orthophoto are generated using the Lunar Reconnaissance Orbiter narrow angle camera image;

[0063] Since the Lunar Reconnaissance Orbiter (LRO) has high orbit and attitude determination accuracy, the lunar surface images (LRO NAC) taken by its narrow angle camera (NAC) almost achieve full lunar surface coverage, with the highest spatial resolution reaching 0.5 m / pixel. Therefore, the embodiment of the present invention selects the Lunar Reconnaissance Orbiter narrow angle camera images as the data source of the digital elevation model and digital orthophoto.

[0064] See also Figure 5 For each local traversable area, the Lunar Reconnaissance Orbiter narrow-angle camera images are used to generate digital elevation models and digital orthophotos, including:

[0065] Step 501, filtering the lunar reconnaissance orbiter narrow-angle camera image according to preset indicators and corresponding thresholds to obtain filtered lunar surface images;

[0066] Step 502, constructing a strict imaging geometry model of the lunar surface image;

[0067] Step 503, constructing a lunar image rational function model according to the lunar image strict imaging geometry model;

[0068] Step 504, performing block adjustment according to the lunar image rational function model to obtain refined imaging geometry model parameters of the page image;

[0069] Step 505, using the lunar image to perform dense matching to generate a digital elevation model DEM.

[0070] Step 506: orthorectify the lunar image using the refined imaging geometry model parameters and the digital elevation model (DEM) to generate a digital orthorectified image.

[0071] Step 507, performing light and color uniformity processing on the digital orthophoto.

[0072] In step 501, the preset indicators include an incident angle indicator, a stereo intersection indicator, and a shadow top distance indicator. Figure 6 This is a flow chart of screening the images of the lunar reconnaissance orbiter narrow-angle camera in an embodiment of the present invention, see Figure 6 , each indicator is introduced below. Among them, the Lunar Reconnaissance Orbiter Narrow Angle Camera image is the LRO NAC image. Through screening, the LRO NAC image for mapping is obtained, that is, the screened lunar surface image.

[0073] 1) Incident angle index i: It indicates the angle between the incident direction of the sun and the tangent direction of the moon.

[0074] In order to avoid the appearance of large-scale black shadows or bright spots, the threshold of the incident angle index is 40<i<80, that is, the embodiment of the present invention preferably selects lunar images with an incident angle range of 40<i<80.

[0075] 2) Stereo intersection angle index r: It indicates the angle of light rays when two images of the Lunar Reconnaissance Orbiter narrow angle camera form a stereo pair. The schematic diagram is as follows: Figure 7 As shown in the figure. Since the imaging mode of the narrow-angle camera of the lunar reconnaissance orbiter is a linear array push-scan and the photographing height is relatively high, when calculating the stereo intersection angle index, the present invention takes any point P in the overlapping area of ​​the narrow-angle camera image of the lunar reconnaissance orbiter, and calculates the intersection angle index through two photographing positions (P l , P r ) and P to solve the triangular relationship. First, the photographic positions of the two lunar reconnaissance orbiter narrow-angle camera images can be obtained according to the orbital parameters of the lunar image (P l and P r ); then they intersect in front and calculate the position of point P; finally, according to the law of cosines, calculate the angle r.

[0076] The threshold value of the stereo intersection angle is r>10, and the present invention preferably uses the lunar reconnaissance orbiter narrow-angle camera image with a stereo intersection angle range of r>10.

[0077] 3) Shadow top distance index d: It represents the vector difference of the shadows of two LRO NAC images in the XY plane. The calculation method of d is as follows: let the solar azimuth angle of the LRO NAC image be sa, then the vector coordinates of the shadow endpoints of the two LRO NAC images in the XY plane can be calculated using formula (1):

[0078]

[0079] Then, the shadow top distance index d can be calculated according to formula (2):

[0080]

[0081] Among them, (sh_x1, sh_y1) and (sh_x2, sh_y2) are the shadow endpoint vector coordinates of the two lunar reconnaissance orbiter narrow angle camera images in the XY plane.

[0082] The threshold value of the shadow top distance index is d<3, and the present invention preferably selects images with a shadow top distance index range of d<3.

[0083] In step 502, a strict imaging geometry model of the lunar surface image may be constructed based on the internal parameters of the lunar reconnaissance orbiter, the satellite ephemeris, and the orbital attitude;

[0084] The internal parameters of the lunar reconnaissance orbiter include focal length, center coordinates in row and column directions, pixel size and distortion parameters, which can be obtained from the IK file of the SPICE (Spacecraft Planet Instrument Camera-Matrix and Events) kernel. The strict imaging geometry model of the lunar surface image includes the internal orientation model and the external orientation model.

[0085] The construction of the strict imaging geometry model of lunar images includes the following steps:

[0086] (1) According to the internal parameters of the lunar reconnaissance orbiter, the internal orientation model expressed by formula (3) is constructed.

[0087]

[0088] In formula (3), sample is the column coordinate of the image point of the Lunar Reconnaissance Orbiter Narrow Angle Camera image in the original data pixel coordinate system; since LRO NAC is a linear array push-broom camera, BORESIGHT_SAMPLE and BORESIGHT_LINE are the coordinates of the image principal point; PIXEL_PITCH is the pixel size; xd and yd are the coordinates of the image point including distortion; k1 is the radial distortion parameter; r is the distance from the image point to the image principal point; xc and yc are the coordinates of the corrected image point in the focal plane coordinate system, in millimeters, and xc and yc constitute the image space coordinate system.

[0089] (2) Based on the internal orientation model, an external orientation model represented by formula (4) is constructed, wherein the external orientation model is used to represent the transformation relationship between the image space coordinate system and the star-fixed coordinate system.

[0090]

[0091] In formula (4), (xc, yc) is the image coordinate of the image point in the image space, f is the focal length, (X, Y, Z) is the object coordinate of the object point corresponding to the image point in the satellite-fixed coordinate system, (X s ,Y s ,Z s ) is the coordinate of the photographic center in the star-fixed coordinate system during imaging, i.e., the external azimuth line element, λ is the scale factor; R is the rotation matrix from the image space coordinate system to the star-fixed coordinate system. R consists of three external azimuth angle elements (camera posture during imaging, i.e., Schematic diagram as Figure 8 ) is calculated as shown in formula (5). Where ω is the roll angle, that is, the rotation angle around the X-axis of the satellite body coordinate system; is the pitch angle, i.e., the rotation angle around the Y axis of the satellite body coordinate system (perpendicular to the satellite's flight direction); γ is the yaw angle, i.e., the rotation angle around the Z axis of the satellite body coordinate system (forming a right-handed system with the XY plane).

[0092]

[0093] in,

[0094] The LRO narrow-angle camera is a push-broom imaging system, and each scan line corresponds to an exterior orientation element. However, the satellite orbit measurement time interval is much longer than the time interval of the Lunar Reconnaissance Orbiter narrow-angle camera image scanning imaging, so interpolation is used to obtain the exterior orientation elements at any time. The present invention uses a third-order polynomial to establish the exterior orientation element function at imaging time t, as shown in formula (6).

[0095]

[0096] Where, X s (t),Y s (t),Z s (t) represents the line element of the exterior orientation at time t, represents the external azimuth element at time t; a i ,b i ,c i ,d i ,e i ,f i i=0,1,2,3 are polynomial coefficients, which can be solved using the least squares method.

[0097] In step 503, it is necessary to construct a lunar image rational function model according to the strict imaging geometry model of the lunar image, that is, to construct a lunar image rational function model according to the internal orientation model and the external orientation model. The specific process is as follows:

[0098] (1) Formula (7) is constructed to represent the rational function model of the lunar image. The rational function model of the lunar image is used to represent the relationship between the normalized coordinates of the image coordinates and the object coordinates.

[0099]

[0100] Among them, (xc n ,yc n ) and (X n ,Y n ,Z n ) represent the normalized coordinates of the image point’s image space coordinates (xc, yc) and the object space coordinates (X, Y, Z) corresponding to the image point, ranging from [-1, 1]. i(i=1,2,3,4) is a third-order polynomial, and its expression is shown in equations (8)-(11):

[0101]

[0102]

[0103]

[0104]

[0105] In the formula, a i ,b i ,c i ,d i (i=1,2,…,20) are polynomial coefficients, b1,d1 are usually 1. The normalization method is shown in formula (12).

[0106]

[0107] Where LINE_SCALE, SAMP_SCALE, LINE_OFF and SAMP_OFF ​​are the normalized parameters of the image space coordinate system; LAT_OFF, LONG_OFF, HEIGHT_OFF, LAT_SCALE, LONG_SCALE and HEIGHT_SCALE are the normalized parameters of the object space coordinate system; lat, lon and hei are longitude, latitude and altitude respectively.

[0108] (2) Divide the elevation of the lunar image coverage area into several elevation surfaces, and generate grid points of the lunar image at a certain interval in the image space coordinate system as virtual control points in the image space. Then, according to the strict imaging geometry model of the lunar image (Equation (6)), project the image coordinates of the grid points to each elevation surface, and calculate the object coordinates of these virtual control points in the image space. Finally, substitute into equations (8)-(11) to solve the polynomial coefficients (i.e., a i ,b i ,c i ,d i (i=1,2,…,20)) to obtain a complete rational function model of the lunar surface image.

[0109] In step 504, a block adjustment is performed based on the rational function model of the lunar image to obtain the refined imaging geometry model parameters of the lunar image expressed by formula (13):

[0110]

[0111] Wherein, e0, e1, e2 and f0, f1, f2 are the affine transformation parameters in the row and column directions respectively, sample and line are the row and column coordinates in the normalized coordinates of the object space coordinates obtained by equation (7), and x and y are the image space coordinates (xc, yc) of the measured image point, i.e., the row and column coordinates of the image point.

[0112] Then, the Taylor series is used to expand equation (13) to obtain equation (14). Where Δe0, Δe1, Δe2, Δf0, Δf1 and Δf2 are the corrections of the affine transformation parameters, Δlat and Δlon are the corrections of sample and line respectively, and the elevation coordinate z can be obtained by interpolating the longitude and latitude coordinates from the DEM.

[0113]

[0114] According to formula (14), the error equation is established for the connection points of the lunar reconnaissance orbiter narrow angle camera image (pre-set points evenly distributed in the overlapping area of ​​the lunar image) and the control points (points selected by the preset manual control point method, a number of control points are evenly distributed in the lunar reconnaissance orbiter narrow angle camera image, and the control points on the planetary image are generally selected from the center of a small impact crater) and solved by least squares iteration until the iteration converges.

[0115] Afterwards, the Lunar Reconnaissance Orbiter narrow-angle camera images are used for dense matching to generate a digital elevation model (DEM). The Lunar Reconnaissance Orbiter narrow-angle camera images are then orthorectified using the refined imaging geometry model parameters and the digital elevation model (DEM) to generate a digital orthophoto. Finally, the digital orthophoto is homogenized and color-uniformed to obtain a digital orthophoto of the target.

[0116] In step 104, according to the cost function, global path planning is performed using the digital elevation model and the digital orthophoto to generate a global reference path;

[0117] See also Fig. 9 , according to the cost function, the global path planning is carried out using the digital elevation model and digital orthophoto to generate a global reference path, including:

[0118] Step 901, using digital orthophotos, determining a safe area in a passable area;

[0119] In the traversable area, the ArcGIS plug-in Crater Survey can be used to extract the depth and diameter of the impact crater from the digital orthophoto and calculate the depth-to-diameter ratio. The depth-to-diameter ratio is compared with the set threshold. The area greater than the set threshold is defined as the danger zone, and the remaining area is the safe zone. The calculation method of the depth-to-diameter ratio is shown in formula (15).

[0120]

[0121] Where d is the depth of the crater and l is the diameter of the crater.

[0122] Step 902, calculating the slope value of each location point in the safety area according to the digital elevation model;

[0123] Specifically, let the elevation change rate of any point in the digital elevation model in the x direction be f x , the rate of change of elevation in the y direction is f y , then the slope β at this location can be expressed by formula (16):

[0124]

[0125] Among them, f x and f y It can be obtained in a local range (3×3 window) by numerical differentiation method or local surface fitting method, such as Fig.10 The present invention adopts the third-order unweighted difference method for calculation, as shown in formula (17). i It represents the elevation value of any point in DEM, and g represents the grid interval of DEM.

[0126]

[0127] Step 903, determining the slope cost of each location point in the safety area;

[0128] In specific implementation, the slope threshold is defined in combination with the mechanical properties of the lunar rover itself, and the slope cost of each position point in the safe area is determined, which is recorded as λ(n). For example, a lunar rover with a slope value greater than the first slope threshold (e.g., 20°) is absolutely not allowed to pass, and the slope cost of the corresponding position point is λ(n)=1; a lunar rover with a slope value less than the second slope threshold (e.g., 5°) can pass safely, and the slope cost of the corresponding position point is λ(n)=0.1; a lunar rover with a slope value between the second slope threshold and the first slope threshold (5°~20°) has a higher cost when passing, and the slope cost of the corresponding position point is λ(n)=0.5.

[0129] Step 904, determining a cost function according to the slope cost, length cost, comfort cost and safety cost;

[0130] 1) Length cost

[0131] The present invention considers the planned path to be composed of straight lines and arcs, and the length cost g(n) of the path is evaluated by the sum of the Euclidean distance of the vectors between nodes and the length of the arc. Assume that the global path includes N nodes, n i =(x i ,y i),i=0,1,…,N, represents the i-th node in the path, n0, n N Represent the starting point and the end point respectively. min is the minimum turning radius, which is related to the mobility of the lunar rover itself, θ i Represents the steering angle of the i-th node. The calculation formula of the length cost g(n) is as follows:

[0132]

[0133] 2) Comfort cost

[0134] The comfort cost h(n) of a path is evaluated by smoothness and consistency. sm (r i ) represents the smoothness function of the i-th path, which is defined as the integral of the square of the curvature over the candidate path, i.e., formula (19). i (s) is the curvature of the i-th path with arc length s, and the upper and lower limits of the integral are the starting point and end point corresponding to the arc length s. co (r i ) represents the consistency function of i paths, which is defined as the integral of the difference between the lateral offset distances of the previous and next two plans along the overlapping area, that is, formula (20). Among them, Δρ(s) is the difference in lateral offset distances at the same arc length. The comfort cost h(n) is calculated using formula (21):

[0135] h(n)=a·h sm (r i )+b·h co (r i ) (19)

[0136] h sm (r i ) = ∫k i 2 (s)ds (20)

[0137]

[0138] 3) Security cost

[0139] The safety cost s(n) of the path is evaluated by the ratio of the distance between the candidate path and the nearest obstacle to the safety distance, see formula (22). In the embodiment of the present invention, the center of the circumscribed circle of the obstacle is used to represent the position of the obstacle. In formula (22), d i Represents the straight-line distance between the candidate path and the nearest obstacle, d max Represents the safe distance between the lunar rover and obstacles, which is a fixed constant.

[0140]

[0141] The final cost function is shown in formula (23):

[0142] f(n)=α·g(n)+β·h(n)+γ·s(n)+δ·λ(n) (23)

[0143] Among them, g(n) represents the length cost, h(n) represents the comfort cost, s(n) represents the safety cost, and λ(n) represents the slope cost; α, β, γ, and δ are weights that can be set according to actual needs.

[0144] Step 905, in the safe area, a path search is performed according to the starting position, the end position and the cost function to obtain a global reference path;

[0145] In the specific implementation, a global path planning algorithm is used to search for paths in the safe area, and the optimal path is determined according to the starting position, the end position and the cost function. The global path planning algorithm in the present invention can be selected from Field D* or A* algorithms, where the optimal path is the global reference path, denoted by L={l i ,i=0,1,…,N-1},l i Represents the path between nodes i and i+1.

[0146] In step 105, a local digital elevation model and a local digital orthophoto are constructed using the stereoscopic lunar surface image perceived in real time by the lunar rover;

[0147] Specifically, stereo matching is performed using the three-dimensional lunar surface images perceived in real time by the lunar rover to generate a local digital elevation model and a local digital orthophoto.

[0148] In step 106, according to the local digital elevation model and local digital orthophoto, the measured data of the lunar rover and the prior lunar soil mechanical parameters, the global reference path is verified based on the dynamic model and the global planning strategy is output.

[0149] See also Fig.11 , step 106 further comprises:

[0150] Step 1101, using the prior lunar soil mechanical parameters and the lunar rover measured data, the lunar soil internal friction angle is identified on-orbit; wherein the prior lunar soil mechanical parameters and the lunar soil internal friction angle identified on-orbit constitute all the lunar soil mechanical parameters;

[0151] In one embodiment, the lunar rover measured data includes payload telemetry information, motor current telemetry information, and lunar rover image information;

[0152] Internal friction angle of lunar soil Will play a dominant role in the wheel-ground action model. Fig.12 , which shows that the internal friction angle of the lunar soil can be calculated by using the measured data of the lunar rover. The steps for on-orbit identification are as follows. Fig.13 .

[0153] Step 1201, calculate the maximum normal stress σ of the wheel according to the lunar rover wheel parameters, the wheel normal support force in the load telemetry information and the lunar rover image information. m ;

[0154]

[0155] θ1=arccos[(r s -z) / r s ] (25)

[0156] θ2=c3θ1 (26)

[0157] θ m =(c1+c2s)θ1 (27)

[0158] r s =r+λh (28)

[0159] z=z0+dz′ (29)

[0160]

[0161] In the formula, F N It can be obtained from the payload telemetry information transmitted by the lunar rover. r is the wheel radius, b is the wheel width, both of which are lunar rover wheel parameters. θ1, θ2 are the wheel-ground interaction angles, θ m is the maximum stress angle, c1, c2 and c3 are coefficients. For the convenience of calculation, θ2 and c3 are both set to 0 in the embodiment of the present invention. s is the equivalent shear radius, λ is the wheel spur coefficient, h is the wheel spur height, and in the embodiment of the present invention, λ=0.5. z is the wheel sinking amount, z0 is the priori value of the wheel sinking amount in the priori lunar soil mechanical parameters, d is the sinking coefficient, and z′ is the sinking amount estimated based on the lunar rover image information. s is the wheel slip rate, and in the embodiment of the present invention, the average slip rate is taken. That is formula (30). is the average wheel linear speed, is the average angular velocity, and All of these can be obtained from the payload telemetry information transmitted by the lunar rover.

[0162] From formula (24), we can see that σ m Yes (F N ,s,r,b,h). Given the normal support force F of the wheel N, wheel slip s, wheel sinking z and wheel parameters (r, b, h), the maximum normal stress σ of the wheel can be calculated m .

[0163] Step 1202: according to the wheel driving resistance torque M in the motor current telemetry information R , calculate the maximum tangential stress τ of the wheel m ;

[0164]

[0165] Among them, τ m is the maximum tangential stress of the wheel, M R is the wheel driving resistance torque, r s is the equivalent shear radius of the wheel, b is the wheel width, θ1, θ2 are the wheel-ground angles.

[0166] Step 1203, according to the maximum normal stress σ of the wheel m and the maximum tangential stress τ m , calculate the internal friction angle of lunar soil

[0167] Maximum tangential stress of the wheel τ m It can also be expressed by formula (32):

[0168]

[0169] θ1′=arccos[(rz) / (r+h)] (33)

[0170] in, is the internal friction angle of lunar soil, σ m is the maximum normal stress of the wheel, c is the cohesion of the lunar soil, K is the shear deformation modulus, θ1′ is the equivalent entry angle considering the effect of the wheel spur, θ m is the maximum stress angle, s is the wheel slip rate, r is the wheel radius, z is the wheel sinking amount, and h is the wheel spur height.

[0171] At the known maximum tangential stress τ m , wheel slip ratio s and wheel sinking amount z, in equation (32), the only unknown parameters are c, If any two typical values ​​are substituted into formula (32), the remaining value can be obtained. In the present invention, c and K are set to a priori values, and the obtained

[0172] Step 1102, calculating the initial position of each wheel of the lunar rover according to the initial position of the lunar rover and in combination with the mechanical structure of the lunar rover;

[0173] Step 1103, displaying the global reference path on the local digital orthophoto based on the dynamics model according to the lunar soil mechanical parameters, the local digital elevation model and the initial positions of the wheels of the lunar rover;

[0174] See also Fig.14 , step 1103 specifically includes:

[0175] Step 1401, read the local digital elevation model, and calculate the wheel-ground interaction force and wheel-ground interaction torque of each wheel according to the lunar soil mechanical parameters and the initial position of each wheel of the lunar rover;

[0176] See also Fig.15 , step 1401 specifically includes:

[0177] Step 1501, read the local digital elevation model, and calculate the state parameters of each wheel of the lunar rover according to the lunar soil mechanical parameters and the initial position of each wheel of the lunar rover. The state parameters include the actual wheel sinking amount z s , slip rate and sideslip angle β;

[0178] The mechanical parameters of lunar soil mainly include the following six: cohesive deformation modulus k c , internal friction deformation modulus Subsidence index n, internal friction angle Cohesion c, shear deformation modulus K, except for the internal friction angle Except for the above, the a priori values, i.e. the a priori lunar soil mechanical parameters, are used. Among them, the slip rate s is calculated as shown in formula (30);

[0179] Actual wheel sinking amount z s The formula is as follows:

[0180]

[0181] The formula for the sideslip angle β is as follows:

[0182] β=σ1-σ2 (35)

[0183] Among them, [A t B t C t ] is the normal vector of the wheel-ground contact plane, (x1, y1, z1) is the intersection point of the wheel with the ground in the opposite direction of the wheel entry angle, and [A t B t C t ] and (x1, y1, z1) can be calculated by the mechanical parameters of the lunar soil. c ,y c ,z c ) is the wheel center position in the initial position of each wheel, σ1 is the direction of the lunar rover head, and σ2 is the tangent direction of the current motion trajectory.

[0184] Step 1502, calculate the wheel-ground interaction force and wheel-ground interaction torque of each wheel according to the wheel-ground interaction force model and the state parameters of each wheel of the lunar rover.

[0185] The wheel-ground interaction forces mainly include the following four: the normal stress F of the wheel N , hook traction force F DP , driving resistance torque M R and the lateral force F L The wheel-ground interaction torque includes: driving resistance torque M R , Steering resistance torque M S and the overturning moment M O When the wheel moves forward, the main stress F N , hook traction force F DP and driving resistance torque M R ; When turning or sliding sideways, the slip angle β will cause the lunar soil to generate a lateral force F on the wheel L , Steering resistance torque M S and the overturning moment M O .

[0186] The calculation formula is as follows:

[0187] F N =rbσ m A+rbτ m B (36)

[0188] F DP =rbτ m A-rbσ m B (37)

[0189] M R =r 2 bτ m C (38)

[0190] in,

[0191]

[0192]

[0193]

[0194] Using the linearization method, the analytical expression of the lateral force can be obtained:

[0195] F L =r s bτ m C{1-exp[-r(1-s)(θ1-θ)tanβ / K y ]} (42)

[0196] The overturning moment M is controlled by concentrated force. O and steering resistance torque M S Perform an approximate calculation:

[0197] M O ≈F L r (43)

[0198] M S ≈F L rsinθ m +0.25F DP bθ s / (2π) (44)

[0199]

[0200] Among them, σ m , τ m are the maximum normal stress and maximum shear stress of wheel-ground action, K y is the ground lateral shear deformation modulus, both of which are empirical values, that is, known. θ is a point on the wheel, θ s is the shear stress angle, F xy is the shear stress in the xy plane. These three quantities can be obtained from the telemetry parameters.

[0201] Step 1402, converting the global reference path into a control strategy;

[0202] Step 1403, calculating the position and speed of the lunar rover body and the position and speed of each wheel according to the control strategy, the wheel-ground interaction force and the wheel-ground interaction torque of each wheel;

[0203] See also Fig.16 , step 1403 specifically includes:

[0204] Step 1601, according to the control strategy control i , the wheel-ground interaction force and wheel-ground interaction torque of each wheel, and the contact angle, calculate the force on the active joints of the lunar rover; among them, the contact angle includes the external friction angle and the internal friction angle of lunar soil

[0205] Step 1602, calculating the expected acceleration of the active joints of the lunar rover according to the forces on the active joints of the lunar rover;

[0206] Step 1603, linearize and integrate the expected acceleration of the active joint of the lunar rover to obtain the position and velocity of the lunar rover body and the position and velocity of each wheel. The formula is as follows:

[0207]

[0208] Among them, for each wheel, is the expected acceleration of the active joint of the wheel at the i-th moment, are the velocity vectors of the wheel at the i-th and i+1-th moments, Φ i , Φ i+1 are the position vectors of the lunar rover at the i-th and i+1-th moments respectively, and Δt is the time step.

[0209] Step 1404, displaying the position of the lunar rover body and each wheel on the local digital orthophoto.

[0210] Step 1104, performing a movement effect evaluation on the global reference path on the local digital orthophoto, and outputting a global planning strategy when the movement effect evaluation result is not greater than a preset threshold.

[0211] In one embodiment, the method further comprises:

[0212] When the mobile effect evaluation result is greater than the preset threshold, the following steps are performed again until the mobile effect evaluation result is not greater than the preset threshold:

[0213] Convert the global reference path into a control strategy;

[0214] Calculate the position vector of the lunar rover body according to the control strategy, the wheel-ground interaction force and contact angle of each wheel of the current lunar rover;

[0215] According to the position vector of the lunar rover body, the positions of the lunar rover body and each wheel are displayed on the local digital orthophoto, and the moving effect of the global reference path is evaluated.

[0216] In one embodiment, the preset threshold is that the path does not crush stones that do not meet the driving requirements or the path does not fall into a deep pit that does not meet the driving requirements.

[0217] In summary, in the method proposed in the embodiment of the present invention, based on the typical landform features of the whole moon, the wide-angle camera images of the lunar reconnaissance orbiter are used to classify the terrain features, and the inaccessible areas and the passable areas are marked based on the classification results; based on the passable areas, the moving path to be planned is divided into multiple local passable areas; for each local passable area, the narrow-angle camera images of the lunar reconnaissance orbiter are used to generate a digital elevation model and a digital orthophoto; the digital elevation model and the digital orthophoto are used to perform global path planning, and the optimal path is determined according to the cost function to generate a global reference path; the stereoscopic lunar surface images perceived in real time by the lunar rover are used to construct a local digital elevation model and a local digital orthophoto; based on the local digital elevation model and the local digital orthophoto, the measured data of the lunar rover and the prior mechanical parameters of the lunar soil, the global reference path is verified based on the dynamic model, and the global planning strategy is output. In the above process, the path planning method is combined with global reference path planning based on prior information (including image information and lunar soil mechanical parameters, where the image information includes lunar reconnaissance orbiter wide-angle camera images and lunar reconnaissance orbiter narrow-angle camera images) and local planning based on the three-dimensional lunar surface images perceived by the lunar rover in real time. This can effectively solve the current situation that obstacles cannot be accurately identified when using low-resolution historical images for path planning, and the path planning cannot support long-distance movement using the lunar rover's real-time perception images. The lunar soil mechanical parameters are identified on-orbit using measured data, and the global reference path is verified based on the dynamic model. The output global planning strategy is more accurate, which will further improve the reliability of the lunar rover's long-distance movement planning.

[0218] The embodiment of the present invention further proposes a lunar rover long-distance movement path planning device, the principle of which is similar to the lunar rover long-distance movement path planning method, which will not be repeated here.

[0219] Fig.17 This is a schematic diagram of a long-distance moving path planning device for a lunar rover in an embodiment of the present invention, comprising:

[0220] A terrain classification module 1701 is used to classify terrain features based on typical landform features of the whole moon using the wide-angle camera image of the lunar reconnaissance orbiter, and mark inaccessible areas and accessible areas based on the classification results;

[0221] A path division module 1702 is used to divide the moving path to be planned into a plurality of local traversable areas based on the traversable area;

[0222] The model generation module 1703 is used to generate a digital elevation model and a digital orthophoto for each local traversable area using the Lunar Reconnaissance Orbiter narrow angle camera image;

[0223] The global reference path generation module 1704 is used to perform global path planning using the digital elevation model and the digital orthophoto, determine the optimal path according to the cost function, and generate a global reference path;

[0224] The local model building module 1705 is used to build a local digital elevation model and a local digital orthophoto using the stereoscopic lunar surface image perceived by the lunar rover in real time;

[0225] The global planning strategy output module 1706 is used to verify the global reference path based on the dynamic model according to the local digital elevation model and local digital orthophoto, the lunar rover's measured data and the prior lunar soil mechanical parameters, and output the global planning strategy.

[0226] In one embodiment, the path division module is specifically used for:

[0227] According to the starting position and the end position of the moving path to be planned, the moving direction is marked in sections in the passable area;

[0228] According to the moving direction at the marked position, the moving path is divided into a plurality of local traversable areas.

[0229] In one embodiment, the model generation module is specifically used to:

[0230] According to preset indicators and corresponding thresholds, the lunar reconnaissance orbiter narrow-angle camera images are screened to obtain screened lunar surface images;

[0231] Construct a strict imaging geometry model for lunar images;

[0232] According to the strict imaging geometry model of lunar images, a rational function model of lunar images is constructed;

[0233] According to the rational function model of lunar images, regional block adjustment is performed to obtain the refined imaging geometry model parameters of the page image.

[0234] Use lunar images for dense matching to generate digital elevation models;

[0235] The refined imaging geometry model parameters and digital elevation model are used to orthorectify the lunar image to generate digital orthophotos.

[0236] Perform light and color uniformity processing on digital orthophotos.

[0237] In one embodiment, the global reference path generation module is specifically used to:

[0238] Using digital orthophotos, identify safe areas within traversable areas;

[0239] According to the digital elevation model, the slope value of each location point in the safety area is calculated;

[0240] Determine the slope cost of each location point in the safety area;

[0241] Determining a cost function according to the slope cost, length cost, comfort cost and safety cost;

[0242] In the safe area, a path search is performed based on the starting position, the end position and the cost function to obtain a global reference path.

[0243] In one embodiment, the global planning strategy output module is specifically used to:

[0244] By using the prior lunar soil mechanical parameters and the lunar rover measured data, the lunar soil internal friction angle is identified on-orbit; the prior lunar soil mechanical parameters and the lunar soil internal friction angle identified on-orbit constitute all the lunar soil mechanical parameters;

[0245] According to the initial position of the lunar rover and the mechanical structure of the lunar rover, the initial position of each wheel of the lunar rover is calculated;

[0246] According to the mechanical parameters of lunar soil, the local digital elevation model and the initial positions of the wheels of the lunar rover, the global reference path is displayed on the local digital orthophoto based on the dynamic model;

[0247] The movement effect of the global reference path is evaluated on the local digital orthophoto, and when the movement effect evaluation result is not greater than the preset threshold, the global planning strategy is output.

[0248] In one embodiment, the lunar rover measured data includes payload telemetry information, motor current telemetry information, and lunar rover image information;

[0249] The global planning strategy output module is specifically used for:

[0250] The maximum normal stress of the wheel is calculated based on the wheel parameters of the lunar rover, the normal support force of the wheel in the load telemetry information and the image information of the lunar rover;

[0251] Calculate the maximum tangential stress of the wheel according to the wheel driving resistance torque in the motor current telemetry information;

[0252] The internal friction angle of the lunar soil is calculated based on the maximum normal stress and maximum tangential stress of the wheel.

[0253] In one embodiment, the global planning strategy output module is specifically used to:

[0254] Read the local digital elevation model and calculate the wheel-ground interaction force and wheel-ground interaction torque of each wheel based on the lunar soil mechanical parameters and the initial position of each wheel of the lunar rover;

[0255] Convert the global reference path into a control strategy;

[0256] Calculate the position and speed of the lunar rover body and the position and speed of each wheel according to the control strategy, the wheel-ground interaction force and wheel-ground interaction torque of each wheel;

[0257] The positions of the lunar rover body and wheels are displayed on the local digital orthophoto.

[0258] In one embodiment, the global planning strategy output module is specifically used to:

[0259] Read the local digital elevation model, and calculate the state parameters of each wheel of the lunar rover according to the mechanical parameters of the lunar soil and the initial position of each wheel of the lunar rover, wherein the state parameters include the actual wheel sinking amount, slip rate and sideslip angle;

[0260] According to the wheel-ground interaction force model and the state parameters of each wheel of the lunar rover, the wheel-ground interaction force and wheel-ground interaction torque of each wheel are calculated.

[0261] In one embodiment, the global planning strategy output module is specifically used to:

[0262] Calculate the forces on the active joints of the lunar rover based on the control strategy, the wheel-ground interaction force and wheel-ground interaction torque of each wheel, and the contact angle; the contact angle includes the external friction angle and the lunar soil internal friction angle;

[0263] According to the forces on the active joints of the lunar rover, the expected acceleration of the active joints of the lunar rover is calculated;

[0264] The expected acceleration of the lunar rover's active joints is linearly integrated to obtain the position and velocity of the lunar rover body and the position and velocity of each wheel.

[0265] In one embodiment, the global planning strategy output module is specifically used to:

[0266] When the mobile effect evaluation result is greater than the preset threshold, the following steps are performed again until the mobile effect evaluation result is not greater than the preset threshold:

[0267] Convert the global reference path into a control strategy;

[0268] According to the control strategy, the wheel-ground interaction force and wheel-ground interaction torque of the current lunar rover wheels, the position and speed of the lunar rover body and the position and speed of each wheel are calculated;

[0269] The positions of the lunar rover body and wheels are displayed on the local digital orthophoto, and the movement effect of the global reference path is evaluated.

[0270] In one embodiment, the preset threshold is that the path does not crush stones that do not meet the driving requirements or the path does not fall into a deep pit that does not meet the driving requirements.

[0271] In summary, in the device proposed in the embodiment of the present invention, based on the typical landform features of the whole moon, the wide-angle camera images of the lunar reconnaissance orbiter are used to classify the terrain features, and the inaccessible areas and the passable areas are marked based on the classification results; based on the passable areas, the moving path to be planned is divided into multiple local passable areas; for each local passable area, the narrow-angle camera images of the lunar reconnaissance orbiter are used to generate a digital elevation model and a digital orthophoto; the digital elevation model and the digital orthophoto are used to perform global path planning, and the optimal path is determined according to the cost function to generate a global reference path; the stereoscopic lunar surface images perceived in real time by the lunar rover are used to construct a local digital elevation model and a local digital orthophoto; based on the local digital elevation model and the local digital orthophoto, the measured data of the lunar rover and the prior mechanical parameters of the lunar soil, the global reference path is verified based on the dynamic model, and the global planning strategy is output. In the above process, the path planning method is combined with global reference path planning based on prior information (including image information and lunar soil mechanical parameters, where the image information includes lunar reconnaissance orbiter wide-angle camera images and lunar reconnaissance orbiter narrow-angle camera images) and local planning based on the three-dimensional lunar surface images perceived by the lunar rover in real time. This can effectively solve the current situation that obstacles cannot be accurately identified when using low-resolution historical images for path planning, and the path planning cannot support long-distance movement using the lunar rover's real-time perception images. The lunar soil mechanical parameters are identified on-orbit using measured data, and the global reference path is verified based on the dynamic model. The output global planning strategy is more accurate, which will further improve the reliability of the lunar rover's long-distance movement planning.

[0272] An embodiment of the present invention also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned lunar rover long-distance movement path planning method when executing the computer program.

[0273] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned lunar rover long-distance movement path planning method is implemented.

[0274] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned lunar rover long-distance movement path planning method is implemented.

[0275] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0276] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0277] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0278] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0279] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for planning a long-distance moving path of a lunar rover, characterized in that: include: Based on the typical landform features of the whole moon, the Lunar Reconnaissance Orbiter wide-angle camera images are used to classify the terrain features, and the inaccessible and accessible areas are marked based on the classification results; Based on the passable area, the moving path to be planned is divided into multiple local passable areas; For each local traversable area, digital elevation models and digital orthophotos were generated using Lunar Reconnaissance Orbiter narrow-angle camera images; Use digital elevation models and digital orthophotos to perform global path planning, determine the optimal path based on the cost function, and generate a global reference path; Use the three-dimensional lunar surface images perceived in real time by the lunar rover to build local digital elevation models and local digital orthophotos; According to the local digital elevation model and local digital orthophoto, the measured data of the lunar rover and the prior mechanical parameters of the lunar soil, the global reference path is verified based on the dynamic model and the global planning strategy is output.

2. The method according to claim 1, characterized in that Based on the traversable area, the moving path to be planned is divided into multiple local traversable areas, including: According to the starting position and the end position of the moving path to be planned, the moving direction is marked in sections in the passable area; According to the moving direction at the marked position, the moving path is divided into a plurality of local traversable areas.

3. The method according to claim 1, characterized in that For each local traversable area, the Lunar Reconnaissance Orbiter narrow-angle camera images are used to generate digital elevation models and digital orthophotos, including: According to preset indicators and corresponding thresholds, the lunar reconnaissance orbiter narrow-angle camera images are screened to obtain screened lunar surface images; Construct a strict imaging geometry model for lunar images; According to the strict imaging geometry model of lunar images, a rational function model of lunar images is constructed; According to the rational function model of lunar images, regional block adjustment is performed to obtain the refined imaging geometry model parameters of the page image. Use lunar images for dense matching to generate digital elevation models; The refined imaging geometry model parameters and digital elevation model are used to orthorectify the lunar image to generate digital orthophotos. Perform light and color uniformity processing on digital orthophotos.

4. The method according to claim 1, characterized in that According to the cost function, the global path planning is carried out using the digital elevation model and digital orthophoto to generate a global reference path, including: Using digital orthophotos, identify safe areas within traversable areas; According to the digital elevation model, the slope value of each location point in the safety area is calculated; Determine the slope cost of each location point in the safety area; Determining a cost function according to the slope cost, length cost, comfort cost and safety cost; In the safe area, a path search is performed based on the starting position, the end position and the cost function to obtain a global reference path.

5. The method according to claim 1, characterized in that According to the local digital elevation model and local digital orthophoto, the measured data of the lunar rover and the prior lunar soil mechanical parameters, the global reference path is verified based on the dynamic model, and the global planning strategy is output, including: By using the prior lunar soil mechanical parameters and the lunar rover measured data, the lunar soil internal friction angle is identified on-orbit; the prior lunar soil mechanical parameters and the lunar soil internal friction angle identified on-orbit constitute all the lunar soil mechanical parameters; According to the initial position of the lunar rover and the mechanical structure of the lunar rover, the initial position of each wheel of the lunar rover is calculated; According to the mechanical parameters of lunar soil, the local digital elevation model and the initial positions of the wheels of the lunar rover, the global reference path is displayed on the local digital orthophoto based on the dynamic model; The movement effect of the global reference path is evaluated on the local digital orthophoto, and when the movement effect evaluation result is not greater than the preset threshold, the global planning strategy is output.

6. The method according to claim 5, characterized in that The measured data of the lunar rover include payload telemetry information, motor current telemetry information and lunar rover image information; Using the prior lunar soil mechanical parameters and the measured data of the lunar rover, the internal friction angle of the lunar soil is identified on-orbit, including: The maximum normal stress of the wheel is calculated based on the wheel parameters of the lunar rover, the normal support force of the wheel in the load telemetry information and the image information of the lunar rover; Calculate the maximum tangential stress of the wheel according to the wheel driving resistance torque in the motor current telemetry information; The internal friction angle of the lunar soil is calculated based on the maximum normal stress and maximum tangential stress of the wheel.

7. The method according to claim 6, characterized in that The internal friction angle of lunar soil is calculated using the following formula: θ1′=arccos[(rz) / (r+h)] z=z0+dz′ Among them, τ m is the maximum tangential stress of the wheel, is the internal friction angle of lunar soil, σ m is the maximum normal stress of the wheel, c is the cohesion of the lunar soil, K is the shear deformation modulus, θ1′ is the equivalent entry angle considering the effect of the wheel spur, θ m is the maximum stress angle, s is the wheel slip rate, r is the wheel radius, h is the wheel spur height, z is the wheel settlement, z0 is the prior value of the wheel settlement in the prior lunar soil mechanical parameters, d is the settlement coefficient, and z′ is the settlement estimated based on the lunar rover image information.

8. The method according to claim 5, characterized in that According to the lunar soil mechanical parameters, the local digital elevation model and the initial positions of the wheels of the lunar rover, the global reference path is displayed on the local digital orthophoto based on the dynamic model, including: Read the local digital elevation model and calculate the wheel-ground interaction force and wheel-ground interaction torque of each wheel based on the lunar soil mechanical parameters and the initial position of each wheel of the lunar rover; Convert the global reference path into a control strategy; Calculate the position and speed of the lunar rover body and the position and speed of each wheel according to the control strategy, the wheel-ground interaction force and wheel-ground interaction torque of each wheel; The positions of the lunar rover body and wheels are displayed on the local digital orthophoto.

9. The method according to claim 8, characterized in that Read the local digital elevation model, and calculate the wheel-ground interaction force and wheel-ground interaction torque of each wheel based on the lunar soil mechanical parameters and the initial position of each wheel of the lunar rover, including: Read the local digital elevation model, and calculate the state parameters of each wheel of the lunar rover according to the mechanical parameters of the lunar soil and the initial position of each wheel of the lunar rover, wherein the state parameters include the actual wheel sinking amount, slip rate and sideslip angle; According to the wheel-ground interaction force model and the state parameters of each wheel of the lunar rover, the wheel-ground interaction force and wheel-ground interaction torque of each wheel are calculated.

10. The method according to claim 8, characterized in that According to the control strategy, the wheel-ground interaction force and wheel-ground interaction torque of each wheel, the position and speed of the lunar rover body and the position and speed of each wheel are calculated, including: Calculate the forces on the active joints of the lunar rover based on the control strategy, the wheel-ground interaction force and wheel-ground interaction torque of each wheel, and the contact angle; the contact angle includes the external friction angle and the lunar soil internal friction angle; According to the forces on the active joints of the lunar rover, the expected acceleration of the active joints of the lunar rover is calculated; The expected acceleration of the lunar rover's active joints is linearly integrated to obtain the position and velocity of the lunar rover body and the position and velocity of each wheel.

11. The method according to claim 5, characterized in that Also includes: When the mobility effect evaluation result is greater than the preset threshold, the following steps are performed again until the mobility effect evaluation result is not greater than the preset threshold, and a new global planning strategy is output: Use digital elevation models and digital orthophotos to replan the global path, determine the optimal path based on the cost function, and generate a new global reference path; converting the new global reference path into a new control strategy; According to the new control strategy, the wheel-ground interaction force and wheel-ground interaction torque of the current lunar rover wheels, the new position and new speed of the lunar rover body, and the new position and new speed of each wheel are calculated; The new positions of the lunar rover body and wheels are displayed on the local digital orthophoto, and the movement effect is evaluated on the global reference path.

12. The method according to claim 5, characterized in that The preset threshold is that the path does not crush stones that do not meet the driving requirements or the path does not fall into a deep pit that does not meet the driving requirements.

13. A lunar rover long-distance movement path planning device, characterized in that: include: The terrain classification module is used to classify the terrain features based on the typical landform features of the whole moon and the images of the Lunar Reconnaissance Orbiter Wide-angle Camera, and mark the inaccessible and accessible areas based on the classification results; A path division module is used to divide the moving path to be planned into multiple local passable areas based on the passable area; The model generation module is used to generate digital elevation models and digital orthophotos for each local traversable area using the Lunar Reconnaissance Orbiter narrow-angle camera images; A global reference path generation module is used to perform global path planning using a digital elevation model and digital orthophotos, determine the optimal path based on a cost function, and generate a global reference path; The local model building module is used to build a local digital elevation model and local digital orthophoto using the stereoscopic lunar surface images perceived by the lunar rover in real time; The global planning strategy output module is used to verify the global reference path based on the dynamic model according to the local digital elevation model and local digital orthophoto, the lunar rover's measured data and the prior lunar soil mechanical parameters, and output the global planning strategy.

14. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 12 is implemented.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

16. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

Images