A star wheel long-distance path planning method and system in a time-varying environment

Abstract

The application discloses a star ball vehicle long-distance path planning method and system in a time-varying environment, comprising the following steps: constructing a three-dimensional space-time cost map varying with time based on a digital elevation map and sun position information; in the three-dimensional space-time cost map, expanding nodes in a way that the start point and the end point are taken as targets, nodes are expanded in the spatial dimension and the time dimension at the same time, and the time dimension is unidirectionally increased; searching for a path based on the calculation of the movement cost and the time cost between nodes in the three-dimensional space-time cost map and the principle of minimum cost, to obtain an optimal path; extracting key path points from the optimal path, generating a smooth movement trajectory meeting the kinematic constraints of the star ball vehicle based on the key path points, and generating a navigation point sequence for controlling the movement of the star ball vehicle according to the smooth movement trajectory. The application solves the problem that long-distance path planning is difficult to adapt to the dramatic change of the time-varying environment light, and improves the driving safety and the global path planning ability of the star ball vehicle in a complex time-varying environment.

Description

A method and system for long-distance path planning of planetary rovers in time-varying environments Technical Field

[0001] This invention relates to the field of planetary rover path planning technology, and in particular to a method and system for long-distance path planning of planetary rovers under time-varying environments. Background Technology

[0002] Long-distance path planning for planetary rovers refers to the spatiotemporal path planning that adapts to changes in terrain and time-varying factors such as light intensity on the surface of extraterrestrial bodies during long-term, long-distance travel. For example, at the lunar poles, the solar altitude angle is low, and the undulating terrain causes drastic changes in light intensity. In the past, for short-term, short-distance travel planning, the conditions of sunlight and shadow remained almost constant over several days, and only terrain obstacles needed to be considered. However, in long-distance planning under time-varying light conditions, the time dimension exceeds ten days, the distance is hundreds of meters, and the light conditions change drastically. Time-varying light becomes a necessary factor to consider in path planning, thus transforming it into a spatiotemporal path planning method that considers both geographical location and time variations.

[0003] Long-distance path planning is a crucial aspect of large-scale scientific exploration missions by rovers. The unique, time-varying environment of extraterrestrial surfaces (such as the lunar poles, where the solar altitude angle is low and the solar azimuth angle changes dynamically over time, coupled with undulating celestial terrain leading to drastic changes in the location and extent of shadowed areas, causing abrupt changes in the state of some passable areas) presents rovers with more complex safety challenges and path feasibility risks during long-distance travel. Currently, mainstream rover path planning algorithms mainly construct path search models based on static environment assumptions, applicable to short-distance, low-latitude regions where lighting conditions do not change significantly, and do not consider the impact of time-varying environmental information on path costs during the planning process. Furthermore, while some algorithms consider environmental dynamism, they only focus on real-time adjustments to dynamic obstacles along short-distance local paths, failing to complete global top-level path planning for long distances, and thus failing to meet the path planning requirements of large-scale scientific exploration missions by rovers.

[0004] As lunar exploration missions expand to more complex terrains and larger areas, extraterrestrial rovers need to complete more scientific exploration tasks within larger exploration areas, placing higher demands on the time-varying environmental adaptability, global optimization, and mission suitability of long-distance path planning. Traditional path planning methods that only consider terrain factors are difficult to adapt to time-varying environments and long-distance driving scenarios, and cannot meet actual engineering needs.

[0005] Therefore, there is an urgent need to provide a technical solution to address the above problems. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method and system for long-distance path planning for planetary rovers in time-varying environments.

[0007] In a first aspect, the present invention provides a long-distance path planning method for a planetary rover under time-varying environments, the technical solution of which is as follows:

[0008] Based on digital elevation maps and solar position information, the distribution of illumination and shadow areas for multiple time slices within the planning period is generated.

[0009] Based on the digital elevation map, a static movement cost representing terrain accessibility is generated;

[0010] The distribution of illumination and shadow regions in each time slice is fused with the static movement cost to form a three-dimensional spatiotemporal cost map that changes dynamically over time.

[0011] In the three-dimensional spatiotemporal cost graph, the starting point and ending point containing spatial coordinates and time information are taken as the starting and ending targets, the nodes are expanded in both spatial and time dimensions and the time dimension is increased in one direction, the movement cost between nodes is calculated based on the three-dimensional spatiotemporal cost graph and the path search is performed according to the principle of minimizing the cost, so as to obtain the optimal path composed of a series of nodes containing spatial coordinates and time information.

[0012] Key path points are extracted from the optimal path, a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover is generated based on the key path points, and a sequence of navigation points for controlling the movement of the planetary rover is generated based on the smooth movement trajectory.

[0013] The beneficial effects of the long-distance path planning method for a planetary rover under time-varying environments proposed in this invention are as follows:

[0014] The method of this invention constructs a three-dimensional spatiotemporal cost map that integrates terrain and dynamic lighting, and simultaneously expands search nodes in the spatial and temporal dimensions to perform global path optimization. This solves the problem that long-distance path planning is difficult to adapt to drastic changes in lighting in time-varying environments, and improves the driving safety and global path planning capability of the planetary rover in complex time-varying environments.

[0015] Secondly, the present invention provides a long-distance path planning system for a planetary rover under time-varying environments, the technical solution of which is as follows:

[0016] The first generation module is used to generate the distribution of illumination and shadow areas for multiple time slices within the planned time period based on digital elevation maps and solar position information;

[0017] The second generation module is used to generate a static movement cost that characterizes terrain accessibility based on the digital elevation map.

[0018] The fusion generation module is used to fuse the illumination and shadow region distribution of each time slice with the static movement cost to form a three-dimensional spatiotemporal cost map that changes dynamically over time.

[0019] The path search module is used in the three-dimensional spatiotemporal cost map to expand nodes by taking the starting point and ending point containing spatial coordinates and time information as the starting and ending targets, expanding nodes in both spatial and time dimensions and increasing the time dimension in one direction, and performing path search based on the three-dimensional spatiotemporal cost map to calculate the movement cost between nodes and follow the principle of minimizing the cost, so as to obtain the optimal path consisting of a series of nodes containing spatial coordinates and time information.

[0020] The path planning module is used to extract key path points from the optimal path, generate a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points, and generate a sequence of navigation points for controlling the movement of the planetary rover based on the smooth movement trajectory.

[0021] The beneficial effects of the long-distance path planning system for a planetary rover under time-varying environments according to the present invention are as follows:

[0022] The system of this invention constructs a three-dimensional spatiotemporal cost map that integrates terrain and dynamic lighting, and simultaneously expands search nodes in the spatial and temporal dimensions to perform global path optimization. This solves the problem that long-distance path planning is difficult to adapt to drastic changes in lighting in time-varying environments, and improves the driving safety and global path planning capability of the planetary rover in complex time-varying environments.

[0023] Thirdly, the technical solution of an electronic device according to the present invention is as follows:

[0024] It includes a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of the long-distance path planning method for a planetary rover in a time-varying environment as described in this invention.

[0025] Fourthly, the technical solution of a computer-readable storage medium provided by the present invention is as follows:

[0026] The computer-readable storage medium stores instructions that, when read, cause the computer-readable storage medium to perform the steps of the long-distance path planning method for planetary rovers in time-varying environments as described in this invention.

[0027] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0028] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0029] Figure 1 is a flowchart illustrating an embodiment of a long-distance path planning method for a planetary rover under time-varying conditions according to the present invention;

[0030] Figure 2 is a schematic diagram of the construction of the occlusion angle database;

[0031] Figure 3 is a schematic diagram of lighting and shadows;

[0032] Figure 4 is a schematic diagram of the slope;

[0033] Figure 5 is a schematic diagram of slope cost;

[0034] Figure 6 is a schematic diagram of roughness;

[0035] Figure 7 is a schematic diagram of roughness cost;

[0036] Figure 8 is a schematic diagram of the three-dimensional spatiotemporal cost;

[0037] Figure 9 is a structural schematic diagram of an embodiment of a long-distance path planning system for a planetary rover under time-varying conditions according to the present invention;

[0038] Figure 10 is a schematic diagram of an embodiment of an electronic device according to the present invention. Detailed Implementation

[0039] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0040] Figure 1 shows a flowchart of an embodiment of a long-distance path planning method for a planetary rover in a time-varying environment provided by the present invention. This method can be executed by electronic devices such as terminal devices or servers. The terminal device can be any fixed or mobile terminal, such as user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, or wearable device. The server can be a single server or a server cluster composed of multiple servers. Any electronic device can implement the long-distance path planning method for a planetary rover in a time-varying environment by having its processor call computer-readable instructions stored in its memory. As shown in Figure 1, the method includes the following steps:

[0041] S1. Based on digital elevation maps and solar position information, generate the distribution of illumination and shadow areas for multiple time slices within the planning period.

[0042] Digital elevation maps (DEMs) refer to digital maps that record the elevation of each location on a planet's surface using a regular grid. For example, a DEM covering the lunar south pole region S with a resolution of 1 m / pixel stores the elevation value of the terrain in each grid cell. Solar position information refers to celestial motion data describing the azimuth and elevation angle of the sun as seen from a specific observation point on the planet's surface at a particular moment. For example, by calling the orbital ephemeris calculation library B, the azimuth and elevation angles of the sun at midnight each day during a planned 10-day period starting from the mission start time at a predetermined landing point on the lunar south pole can be obtained. The planned period refers to the continuous time period covered by the path planning. For example, if a rover plans to perform a 10-day movement mission starting from mission start time T0, then this continuous 10-day period is the planned period. A time slice refers to an independent time segment obtained by discretizing a continuous planning period at fixed intervals. For example, discretizing the above 10-day planning period at 1-day intervals yields 10 time slices, corresponding to day 1, day 2, and so on up to day 10. Illumination and shadow area distribution refers to the spatial distribution of each location within a target area that is determined to be either illuminated by sunlight or shaded by terrain within a specific time slice. For example, for the time slice "day 3," a binary distribution map is generated based on the sun's position and digital elevation map at that moment, where each unit is labeled as either "illuminated area" or "shaded area."

[0043] S2. Based on the digital elevation map, generate a static movement cost that characterizes terrain accessibility.

[0044] Terrain accessibility refers to the inherent property of terrain features on the ease with which a rover can move through them; for example, a gently sloping, flat area has high accessibility, while a steep slope covered with large rocks has low accessibility or is impassable. Static movement cost refers to a value determined solely by the location's terrain features and remaining constant over time during the planning period, used to quantify the difficulty or energy consumption of the rover's movement at that location; for example, a cost matrix of the same size as the digital elevation map is calculated, where each accessible cell stores a movement cost between 0 and 1.

[0045] S3. The distribution of illumination and shadow regions in each time slice is fused with the static movement cost to form a three-dimensional spatiotemporal cost map that changes dynamically over time.

[0046] Among them, the three-dimensional spatiotemporal cost map refers to a three-dimensional data model composed of a two-dimensional spatial grid with the addition of a time dimension, in which each (x, y, t) cell stores the comprehensive movement cost or passage status at time t and located at position (x, y); for example, a data cube with dimensions of 1000 (rows) * 1000 (columns) * 10 (time layers).

[0047] S4. In the three-dimensional spatiotemporal cost graph, the starting point and ending point containing spatial coordinates and time information are taken as the starting and ending targets, the nodes are expanded in both spatial and time dimensions and the time dimension is increased in one direction, the movement cost between nodes is calculated based on the three-dimensional spatiotemporal cost graph and the path search is performed according to the principle of minimizing the cost, so as to obtain the optimal path composed of a series of nodes containing spatial coordinates and time information.

[0048] Spatial coordinates refer to numerical pairs used to uniquely identify a location on a two-dimensional plane; for example, in a lunar map using UTM projection, the spatial coordinates of a point can be represented as (500,123.4 m east, 3,456,789.0 m north). Time information refers to a specific moment or point in time associated with a path point; for example, an intermediate point obtained through path searching might contain the time information "Day 3, 14:30". The starting point refers to the spatial location and time of the rover at the start of path planning; for example, the starting point can be defined as spatial coordinates (500,000 m east, 3,456,000 m north) and time "Day 1, 08:00". The ending point refers to the expected spatial location and time the rover will reach at the end of path planning; for example, the ending point can be defined as spatial coordinates (502,000 m east, 3,458,000 m north) and time "Day 10, 18:00". The starting and ending points refer to the starting and ending states that need to be connected in the path search problem. For example, the path search algorithm needs to find a feasible trajectory from "starting point (position P1, day 1 08:00)" to "ending point (position P2, day 10 18:00)".

[0049] In this context, spatial dimension refers to the two directions (usually the X and Y axes) used to describe positional changes in the path search state space. For example, in a three-dimensional spatiotemporal cost graph, a node (x, y, t) can be expanded to adjacent spatial nodes, such as (x+1, y, t) and (x, y+1, t). Temporal dimension refers to the single direction used to describe temporal changes in the path search state space. For example, in a three-dimensional spatiotemporal cost graph, a node (x, y, t) can be expanded to the next node in the time dimension (x, y, t+Δt). Unidirectional increment means that in path search, the expansion of the time dimension can only proceed in the future and cannot go back to the past; for example, from a node representing "day 2," expansion can only extend to nodes representing "day 3" and later. Node expansion refers to the process in path search algorithms of exploring all possible next states from the currently examined state (node); for example, starting from node (x, y, t), the algorithm simultaneously explores the nodes in its 8-neighborhood at the same time and the next time slice node at the same location. Movement cost refers to the quantified value of the total cost incurred by the rover to move from one state (node) to the next; for example, the movement cost from node A to node B may combine terrain access costs and waiting time costs. The minimum cost principle refers to the optimization criterion followed by path search algorithms, which is to select the path with the minimum cumulative movement cost from the starting point to the destination among all feasible paths; for example, when using the A* search algorithm, the aim is to find the path that minimizes the cumulative cost g(n) upon reaching the destination. Path search refers to the process of finding a feasible path from the starting point to the destination using a specific algorithm under given state space and constraints; for example, in the state space of a three-dimensional spatiotemporal cost graph, an improved A* algorithm is used to find the cost-optimal path from the starting point to the destination. The optimal path refers to the spatiotemporal path with the minimum total movement cost found through path search based on a preset cost model and optimization principles. For example, the search yields a path consisting of a sequence of multiple spatiotemporal nodes, whose total cost is the minimum among all feasible paths.

[0050] S5. Extract key path points from the optimal path, generate a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points, and generate a sequence of navigation points for controlling the movement of the planetary rover based on the smooth movement trajectory.

[0051] Critical path points refer to a few important points selected from the discrete optimal path node sequence that play a decisive role in the path geometry. For example, after processing a path containing multiple points using the Douglas-Peucker algorithm, the line connecting the remaining critical points can approximately represent the main direction of the original path. Planetary rover kinematic constraints refer to the limitations determined by the physical structure and motion characteristics of the planetary rover that must be considered when planning a smooth trajectory. For example, the radius of curvature of the planned trajectory cannot be less than the minimum turning radius of the planetary rover. A smooth trajectory refers to a continuous spatial curve with a gentle curvature change, described by a mathematical function, that conforms to the kinematic constraints of the planetary rover. For example, fitting critical path points with a cubic spline curve generates an infinitely differentiable curve in a two-dimensional plane. A navigation point sequence refers to a series of discrete position points sampled from the smooth trajectory according to certain rules, used to directly control the movement of the planetary rover. For example, according to the control cycle of the planetary rover, a target point is sampled from the smooth trajectory every 10 seconds, and these target points are arranged in chronological order to form a navigation point sequence.

[0052] The technical solution of this embodiment constructs a three-dimensional spatiotemporal cost map that integrates terrain and dynamic lighting, and simultaneously expands search nodes in the spatial and temporal dimensions to perform global path optimization. This solves the problem that long-distance path planning is difficult to adapt to drastic changes in lighting in time-varying environments, and improves the driving safety and global path planning capability of the planetary rover in complex time-varying environments.

[0053] In one alternative approach, S1 specifically includes:

[0054] Based on the digital elevation map, a terrain occlusion angle database is constructed; wherein, the terrain occlusion angle database is used to record the terrain occlusion angle of each location in the digital elevation map in different directions.

[0055] The terrain shading angle database refers to a pre-calculated and stored collection of data on the maximum elevation angle of terrain shading at each location in each azimuth direction within a digital elevation map. For example, for a 1000x1000 DEM of a target area, the maximum elevation angle of view in each unit without terrain shading is calculated at 0.5° azimuth intervals, forming a multi-layered database. The terrain shading angle refers to the maximum elevation angle formed by the terrain outline when viewed from a target location along a specific azimuth direction. For example, at a certain point, looking due east (azimuth 90°), if the angle between the highest point of a distant ridge and the horizon is 5°, then the terrain shading angle due east at that point is 5°.

[0056] Based on the solar position information, calculate the solar azimuth and solar altitude angle corresponding to each time slice within the planned time period.

[0057] The solar azimuth angle refers to the horizontal angle of the sun as seen from the observation point, usually measured clockwise from true north. For example, at a certain observation point on a certain day and at a certain time, if the sun is located 150° clockwise from true north, then the solar azimuth angle at that moment is 150°. The solar altitude angle refers to the angle between the sun and the horizon as seen from the observation point. For example, at the south pole of the moon, the solar altitude angle at a certain moment may be only 1.5°.

[0058] Based on the solar azimuth angle corresponding to each time slice, the corresponding terrain occlusion angle is queried from the terrain occlusion angle database.

[0059] By comparing the solar altitude angle and terrain occlusion angle corresponding to any time slice, it is determined whether each location in the digital elevation map is in the light or shadow of that time slice, and each location is marked as a lit area or a shadow area, generating the light and shadow area distribution of that time slice, until the light and shadow area distribution of each time slice is obtained.

[0060] In this context, a sunny area refers to a region where, within a specific time slice, the solar altitude angle is greater than the terrain shading angle at that location in the direction of the solar azimuth. For example, in a certain time slice, the solar altitude angle at a point is 1.8°, while the terrain shading angle at that point in the direction of the solar azimuth is 1.0°. Since 1.8 > 1.0, this point is identified as a sunny area. A shadowed area refers to a region where, within a specific time slice, the solar altitude angle is less than or equal to the terrain shading angle at that location in the direction of the solar azimuth. For example, in the same time slice, another point also has a solar altitude angle of 1.8°, but its terrain shading angle in the direction of the solar azimuth is 2.5°. Since 1.8 < 2.5, this point is identified as a shadowed area.

[0061] In the above-mentioned optional methods, a terrain shading angle database is further constructed to store shading information of different azimuths at each location. The shading angle corresponding to the solar azimuth angle is queried and compared with the solar altitude angle to quickly determine the shadow status of each location at different times, thereby improving the efficiency and accuracy of the calculation of the distribution of light and shadow areas.

[0062] In one alternative approach, S2 specifically includes:

[0063] Calculate the terrain slope and terrain roughness values ​​for each location in the digital elevation map.

[0064] The terrain slope value refers to the degree of inclination of the earth's surface at a certain point, usually expressed as an angle. For example, by calculating the elevation of a point and its surrounding pixels in a digital elevation map, the slope value of the local area where that point is located is found to be 8°. The terrain roughness value is a measure describing the degree of local surface undulation, usually characterized by the maximum elevation difference within a certain range. For example, calculating the difference between the maximum and minimum elevation values ​​of 9 points within a 3x3 window around a point yields a roughness value of 0.15 m for that point.

[0065] Locations where the terrain slope value is greater than the climbing ability threshold of the rover or the terrain roughness value is greater than the obstacle crossing ability threshold are identified as obstacle areas.

[0066] The climbing ability threshold refers to the maximum gradient that a rover can safely climb; for example, if a rover's designed climbing ability is 15°, then 15° is the climbing ability threshold. The obstacle crossing ability threshold refers to the maximum height difference between local obstacles on the ground that a rover can safely overcome; for example, if a rover's obstacle-crossing wheels are designed to overcome a rock with a height of 0.2 m, then 0.2 m is the obstacle-crossing ability threshold. An obstacle area refers to an area deemed impassable due to terrain conditions exceeding the rover's capabilities; for example, areas with a gradient greater than 15° or a roughness greater than 0.2 m on the map are marked as obstacle areas in the static movement cost map.

[0067] The static movement cost is generated by normalizing the terrain slope and terrain roughness values ​​of each location in the digital elevation map that does not belong to the obstacle area.

[0068] Normalization cost calculation refers to the process of mapping the original terrain feature values, which may have different dimensions and ranges, to a unified standard numerical range (such as 0 to 1) according to certain rules; for example, mapping a region with a slope value between 0° and 15° to a moving cost value between 0 and 0.8 through a linear function.

[0069] Among the above-mentioned optional methods, the terrain accessibility is further assessed by comprehensively considering slope and roughness, a capability threshold is set to clearly delineate obstacle areas, the movement cost of accessible areas is calculated in a normalized manner, and a detailed static cost map is generated to provide a scientific quantitative basis for terrain difficulty for path planning.

[0070] In one alternative approach, S3 specifically includes:

[0071] For any time slice within the planned time period, the shaded area is set as impassable based on the distribution of light and shadow areas in that time slice, and the obstacle area is set as impassable based on the static movement cost. Each other location in the digital elevation map is then assigned a corresponding movement cost based on the static movement cost, generating a two-dimensional cost layer for that time slice. This process continues until a two-dimensional cost layer for each time slice is obtained. Each two-dimensional cost layer for each time slice consists of the accessibility status or movement cost of each location corresponding to that time slice.

[0072] In this context, the movement cost value refers to the specific cost assigned to each passable location in a two-dimensional cost layer or a three-dimensional spatiotemporal cost map, quantifying the relative difficulty of movement at that location. For example, in the two-dimensional cost layer of a certain time slice, the movement cost value stored at location (x, y) is 0.35. The two-dimensional cost layer refers to a data layer that, for a specific time slice, expresses the passability status, incorporating lighting and shadow information and static terrain costs, in the form of a two-dimensional grid. For example, for the time slice "Day 3," a 1000*1000 grid is generated, with each grid storing either a movement cost value or an "impeded" marker. The passability status refers to the most basic classification of each grid cell in the two-dimensional cost layer: passable or impeded. For example, in the final generated two-dimensional cost layer, each cell contains either a specific movement cost value (indicating passability) or a special marker (indicating impediment).

[0073] According to the time sequence, the two-dimensional cost layers corresponding to each time slice within the planned time period are combined to form the three-dimensional spatiotemporal cost map.

[0074] In the above-mentioned optional methods, the lighting and shadow distribution of each time slice is further integrated with the static cost. The shadow area and obstacle area are uniformly set as impassable, and the remaining positions are assigned corresponding movement costs to generate a two-dimensional cost layer. A three-dimensional spatiotemporal cost map is constructed by combining them in time sequence to achieve unified modeling of dynamic lighting and static terrain.

[0075] In one alternative approach, S4 specifically includes:

[0076] Starting from the aforementioned starting point, nodes are expanded in the three-dimensional spatiotemporal cost graph; wherein, the expansion of nodes is carried out simultaneously in the spatial and temporal dimensions, the expansion of the spatial dimension is based on a predefined geometric neighborhood, and the expansion of the temporal dimension is unidirectionally increasing along the time axis to the next time slice.

[0077] In this context, the predefined geometric neighborhood refers to the set of nodes spatially adjacent to the current node, defined beforehand during the spatial dimensional expansion of the path search. For example, when using a raster map, a commonly used predefined geometric neighborhood is the 8-neighborhood, which consists of eight adjacent raster cells: the top, bottom, left, right, and four diagonal directions of the central node. Unidirectional time axis increment means that during the time dimensional expansion of the path search, a node can only move a fixed time interval in the positive direction (future) of the time axis. For example, from a node representing "day 3," the node can only expand to a node representing "day 4." The next time slice refers to the time segment immediately following the current time slice in the discretized time series. For example, if the current time slice is "day 5" and the slice interval is 1 day, then the "next time slice" refers to "day 6."

[0078] The cumulative cost during node expansion is calculated based on the movement cost of the corresponding position of the node in the three-dimensional spatiotemporal cost map.

[0079] The node expansion process refers to the iterative steps in a path search algorithm that sequentially explore all possible subsequent nodes from the current node and calculate the corresponding costs. For example, in each step of the A* algorithm, the node with the smallest evaluation function is selected from the open list for expansion, and the costs of all its neighboring nodes are calculated. The cumulative cost refers to the sum of the movement costs of all road segments traversed from the starting point of the path search to the currently examined node. For example, during the search process, the cumulative cost g(n) to reach each node n is recorded.

[0080] Based on the accumulated cost, and in accordance with the principle of minimizing cost, a path is searched from the starting point to the ending point, and the path with the minimum accumulated cost is determined as the optimal path.

[0081] It should be noted that the movement cost used in the path search process of this embodiment has a specific meaning. Unlike the movement cost that generally only considers terrain suitability, the movement cost constructed in this embodiment is a combination of spatial terrain access cost and temporal waiting cost.

[0082] To achieve spatiotemporal joint optimization, this embodiment proposes a spatiotemporal search method that integrates time and spatial movement costs. The core of this method lies in designing a unified cost function that quantifies and integrates the spatial movement cost δ(x,y,t) and the time cost ζ(t). The spatial cost is calculated by weighting the slope cost S(x,y) and roughness cost R(x,y), i.e., δ(x,y,t) = α*S(x,y) + β*R(x,y). The time cost ζ(t) = λT(t) represents the time cost incurred in moving from one time slice to the next. To enable comparison and accumulation of spatial and time costs at the same scale, both need to be normalized and weighted by coefficients α, β, and λ.

[0083] When expanding nodes in the 3D spatiotemporal cost map, the search algorithm employs a 9-neighborhood search strategy, including 8 spatial directions (adjacent grids within the same time slice) and 1 temporal direction (the next time slice at the same spatial location). This design allows the algorithm to weigh the options of "immediately moving to the surrounding space" versus "staying in place and waiting until the next moment" at each decision point. The key to comprehensive optimization lies in setting weight coefficients to make the maximum spatial cost of moving within a time slice comparable to the temporal cost of waiting for a complete time slice. Specifically, by combining the rover's maximum movement speed and the time slice length, the theoretically maximum distance that can be moved within a time slice can be calculated, and then the upper limit of the spatial cost generated by this maximum distance under the worst terrain conditions can be estimated. By correlating this upper limit value with the cost of a time slice (λT(t)), the algorithm can proactively choose to "wait" until the next time slice when the travel cost in the surrounding area is high at the current moment, hoping for a better travelable path to emerge after the lighting conditions change. This mechanism ensures the global adaptability of path planning in time-varying environments and achieves the joint optimal allocation of time and space resources.

[0084] In the above-mentioned optional methods, the spatial and temporal dimension nodes are further expanded simultaneously in the three-dimensional spatiotemporal cost graph. The spatial expansion is based on the geometric neighborhood, and the temporal expansion is unidirectionally increasing. The movement cost is accumulated and the optimal path is searched according to the minimum principle to achieve long-distance spatiotemporal joint global optimization.

[0085] In one alternative approach, the step of extracting key path points from the optimal path and generating a smooth trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points includes:

[0086] The Douglas-Peucker algorithm is used to extract critical path points from a series of location points contained in the optimal path.

[0087] The Douglas-Peucker algorithm is a geometric algorithm for curve simplification that reduces the number of points that make up a curve by retaining the points that have the greatest impact on the overall shape. For example, when processing a polyline composed of multiple points, the algorithm outputs a set of key points to approximate the original curve after setting a distance tolerance ε.

[0088] Based on the key path points, a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover is generated using a cubic spline curve fitting method.

[0089] The cubic spline curve fitting method refers to a mathematical method that uses piecewise cubic polynomial functions to make a curve smoothly pass through a series of given key points. For example, given multiple key path points, a smooth curve that is continuously differentiable of the second order can be constructed by calculating the cubic polynomial coefficients between every two adjacent points.

[0090] Among the above-mentioned optional methods, the Douglas-Peucker algorithm is further used to extract key path points to reduce data redundancy, and cubic spline curve fitting is used to generate a continuous and smooth trajectory to meet the kinematic constraints of the planetary rover, thereby improving trajectory smoothness and practical control feasibility.

[0091] In one alternative approach, the step of generating a sequence of navigation points for controlling the movement of the planetary rover based on the smoothed trajectory includes:

[0092] The sampling interval on the smooth movement trajectory is determined based on the distance of the planetary rover's single-step movement control.

[0093] In this context, "planetary rover" refers to a robotic vehicle designed for mobile exploration on the surface of extraterrestrial bodies. For example, a certain type of planetary rover, powered by solar energy, possesses autonomous or teleoperated movement capabilities. The distance of a single-step movement control refers to the expected length of a linear movement that the planetary rover will execute and complete in a single operation after its navigation and control system issues a movement command. For example, based on the design of a certain type of planetary rover's movement control system, its positioning update and control command cycle is 10 seconds, and it moves at a preset safe speed of 0.5 m / s within this cycle; therefore, the distance of a single-step movement control is set to 5 meters. The sampling interval refers to the distance or time difference between two adjacent sampling points when performing discrete sampling on a continuous curve. For example, to generate navigation points, based on the planetary rover's movement speed of 200 m / h, if it is desired to issue a navigation point every 10 seconds, the sampling interval is approximately 0.56 meters.

[0094] Multiple location points are extracted from the smooth movement trajectory according to the sampling interval.

[0095] Here, a location point refers to a point on a two-dimensional plane represented by a coordinate pair (x, y); for example, a point in a navigation point sequence with coordinates (501235.6 m east and 3457123.4 m north).

[0096] The plurality of location points are arranged in order along the smooth movement trajectory to form the navigation point sequence.

[0097] In the above-mentioned optional methods, the smooth trajectory sampling interval is further determined according to the rover's movement speed, the corresponding position points are extracted and arranged in time sequence to generate a navigation point sequence, thereby realizing the conversion from continuous trajectory to discrete control commands and enhancing the adaptability and practicality of navigation control.

[0098] To achieve long-distance movement path planning in time-varying environments, this embodiment establishes a spatiotemporal integrated environmental cost model for the planetary rover, constructs a spatiotemporal movement path search method, and generates navigation points that satisfy the rover's movement constraints to address path planning requirements under scenarios with drastic changes in illumination. Long-distance movement path planning is based on a high-precision digital elevation map constructed from remote sensing imagery. The planning process mainly includes the following five steps:

[0099] Step 1: Modeling the time-varying lighting and shadow environment.

[0100] To accurately calculate the illumination and shadow conditions over a large area, the core task is to accurately calculate the terrain's shading of the sun, i.e., the maximum terrain shading angle at each location in all directions. As shown in Figure 2, the terrain shading angle is calculated point-by-point along a 360° azimuth. The azimuth angles are discretized at specified intervals (e.g., 0.5°) to generate a series of terrain shading angle map data. These maps together constitute a terrain shading angle database, which records the terrain shading angle at each location in the digital elevation map in different azimuths.

[0101] Step 2: Modeling the movement costs of time-varying and time-invariant factors.

[0102] The spatiotemporal integrated modeling of movement costs comprises two parts. The first part is the construction of a time-varying illumination and shadow map. Planetary rovers are constrained by illumination conditions and can typically only move within illuminated areas, which change rapidly over time. Based on the terrain occlusion angle database constructed in step one, combined with solar position information obtained from a celestial orbit kinematics database, the solar altitude angle and azimuth angle corresponding to each time slice within the planning period can be calculated. By comparing the terrain occlusion angle and solar altitude angle of each location in the digital elevation map along the solar azimuth direction for that time slice, it is determined whether the location is in illumination or shadow, and each location is marked as either an illuminated or shadowed area, thus generating the illumination and shadow region distribution for that time slice. Figure 3 shows an example of an illumination and shadow map for a certain time slice.

[0103] The second part is the static geometric movement cost modeling, which is determined by the terrain and does not change over time. The movement cost is mainly analyzed from two terrain factors: slope and roughness. First, the terrain slope and roughness values ​​for each location in the digital elevation map are calculated. Slope is calculated by analyzing the elevation values ​​of the central pixel and its eight neighboring pixels. Roughness can be calculated using the difference between the maximum and minimum elevation values ​​in the neighborhood. Figure 4 shows a slope diagram, Figure 5 shows a slope cost diagram, Figure 6 shows a roughness cost diagram, and Figure 7 shows a roughness cost diagram. Second, the terrain slope value is compared with the rover's climbing ability threshold, and the terrain roughness value is compared with the rover's obstacle-crossing ability threshold. Locations with terrain slope values ​​greater than the climbing ability threshold or terrain roughness values ​​greater than the obstacle-crossing ability threshold are identified as obstacle areas. Finally, for each location in the digital elevation map that does not belong to an obstacle area, the terrain slope and terrain roughness values ​​are normalized for cost calculation, generating a static movement cost map.

[0104] Step 3: Integrated modeling of spatiotemporal mobility costs.

[0105] The time-varying illumination and shadow information obtained in step two is fused with the static terrain cost information to form a three-dimensional spatiotemporal cost map that dynamically changes over time. For any time slice within the planning period, the following operations are performed: shadow areas are set as impassable based on the illumination and shadow area distribution of that time slice; obstacle areas are set as impassable based on the static movement cost; and each other location in the digital elevation map is assigned a corresponding movement cost based on the static movement cost. This generates a two-dimensional cost layer corresponding to that time slice, which consists of the accessibility status or movement cost of each location corresponding to that time slice. The two-dimensional cost layers corresponding to each time slice within the planning period are combined in chronological order to form a three-dimensional spatiotemporal cost map. Figure 8 illustrates the concept of a three-dimensional spatiotemporal cost volume, where the time dimension extends from the planning start time to the end time. The cost map of each Ti section consists of a comprehensive terrain cost map and a shadow obstacle map at time Ti.

[0106] Step 4: Search for long-distance movement paths in a dynamic spatiotemporal environment.

[0107] Path search is performed in a 3D spatiotemporal cost graph to obtain the optimal path consisting of a series of nodes containing spatial coordinates and temporal information. The search uses a starting point and an ending point containing spatial coordinates and temporal information as the starting and ending targets. The search process starts from the starting point and expands nodes in the 3D spatiotemporal cost graph; node expansion occurs simultaneously in both spatial and temporal dimensions. Spatial expansion is based on predefined geometric neighborhoods (e.g., eight neighborhoods), while temporal expansion is unidirectionally increasing along the time axis to the next time slice. During node expansion, the movement cost needs to be calculated. The movement cost consists of spatial and temporal costs. The spatial cost δ(x,y,t) is calculated by weighting the slope cost S(x,y) and roughness cost R(x,y) at position (x,y), i.e., δ(x,y,t)=α*S(x,y)+β*R(x,y). The temporal cost ζ(t) is related to the time increment from the previous time slice to the current time slice, i.e., ζ(t)=λT(t). Here, α, β, and λ are normalized weighting coefficients. The total cost g(n) of node (x,y,t) is the cumulative sum of the movement costs of all segments on the path from the starting point to that node, i.e., g(n) = ∑[α*S(x,y) + β*R(x,y) + λT(t)]. Based on the movement cost of the corresponding position of the node in the 3D spatiotemporal cost map, the cumulative cost during the node expansion process is calculated. Based on the cumulative cost, and according to the principle of minimizing cost, a path from the starting point to the ending point is searched, and the path with the minimum cumulative cost is determined as the optimal path.

[0108] Step 5: Fitting long-distance movement paths and generating navigation points.

[0109] The optimal path obtained in step four is fitted to satisfy the kinematic constraints of the planetary rover, generating a smooth trajectory. The Douglas-Peucker algorithm is used to extract critical path points from a series of locations within the optimal path. Based on these extracted critical path points, a cubic spline curve fitting method is employed to generate a smooth trajectory that satisfies the kinematic constraints of the planetary rover.

[0110] A sequence of navigation points for controlling the rover's movement is generated based on the smooth trajectory. The sampling interval on the smooth trajectory is determined based on the rover's speed. Multiple location points are extracted from the smooth trajectory according to this sampling interval. These extracted location points are then arranged in the order they appear on the smooth trajectory to form the navigation point sequence.

[0111] This embodiment solves the problem of long-distance "light-chasing" movement path planning for a planetary rover in a time-varying environment. This embodiment achieves integrated spatiotemporal cost environment modeling under conditions of drastic lighting changes. By constructing a three-dimensional spatiotemporal cost map that integrates time-varying lighting shadows and static terrain accessibility, a unified dynamic environment representation is provided for path planning. Based on this, a global spatiotemporal path search method calculated using this three-dimensional spatiotemporal cost map is established, capable of jointly optimizing in both spatial and temporal dimensions to generate the spatiotemporal path with the optimal cost. Simultaneously, a method is provided to extract key path points from discrete spatiotemporal path points and further generate smooth navigation trajectories, achieving comprehensive optimization and adaptive adjustment of the movement path in both temporal and spatial dimensions.

[0112] This embodiment effectively supports robust path planning under the influence of spatiotemporal variations in long-distance driving tasks. By iteratively calculating the geometric features of path points and optimizing distance thresholds, it achieves comprehensive optimization and extraction of key path points. This embodiment constructs an integrated solution for mobile path search suitable for complex time-varying environments. Through iterative optimization of the key point extraction and navigation point generation process, it significantly improves the applicability of the planned mobile path over a long period and its adaptability to time-varying environmental factors, demonstrating high engineering application value.

[0113] Figure 9 shows a schematic diagram of an embodiment of a long-distance path planning system 200 for a planetary rover in a time-varying environment provided by the present invention. As shown in Figure 9, the long-distance path planning system 200 for a planetary rover in a time-varying environment includes:

[0114] The first generation module 201 is used to generate the distribution of light and shadow areas of multiple time slices within the planned time period based on digital elevation maps and solar position information.

[0115] The second generation module 202 is used to generate a static movement cost representing terrain accessibility based on the digital elevation map.

[0116] The fusion generation module 203 is used to fuse the illumination and shadow region distribution of each time slice with the static movement cost to form a three-dimensional spatiotemporal cost map that changes dynamically over time.

[0117] The path search module 204 is used to perform path search in the three-dimensional spatiotemporal cost map, taking the starting point and ending point containing spatial coordinates and time information as the starting and ending targets, expanding the nodes in both spatial and time dimensions and increasing the time dimension in one direction, calculating the movement cost between nodes based on the three-dimensional spatiotemporal cost map and following the principle of minimizing the cost, to obtain an optimal path consisting of a series of nodes containing spatial coordinates and time information.

[0118] The path planning module 205 is used to extract key path points from the optimal path, generate a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points, and generate a sequence of navigation points for controlling the movement of the planetary rover based on the smooth movement trajectory.

[0119] In one alternative embodiment, the first generation module 201 is specifically used for:

[0120] Based on the digital elevation map, a terrain occlusion angle database is constructed; wherein, the terrain occlusion angle database is used to record the terrain occlusion angle of each location in the digital elevation map in different directions;

[0121] Based on the solar position information, calculate the solar azimuth and solar altitude angle corresponding to each time slice within the planned time period;

[0122] Based on the solar azimuth angle corresponding to each time slice, the corresponding terrain occlusion angle is queried from the terrain occlusion angle database respectively;

[0123] By comparing the solar altitude angle and terrain occlusion angle corresponding to any time slice, it is determined whether each location in the digital elevation map is in the light or shadow of that time slice, and each location is marked as a lit area or a shadow area, generating the light and shadow area distribution of that time slice, until the light and shadow area distribution of each time slice is obtained.

[0124] In one alternative embodiment, the second generation module 202 is specifically used for:

[0125] Calculate the terrain slope and terrain roughness values ​​for each location in the digital elevation map;

[0126] Locations where the terrain slope value is greater than the climbing ability threshold of the rover or the terrain roughness value is greater than the obstacle crossing ability threshold are identified as obstacle areas.

[0127] The static movement cost is generated by normalizing the terrain slope and terrain roughness values ​​of each location in the digital elevation map that does not belong to the obstacle area.

[0128] In an alternative embodiment, the fusion generation module 203 is specifically used for:

[0129] For any time slice within the planned time period, the shadowed area is set as impassable based on the distribution of light and shadow areas in that time slice, and the obstacle area is set as impassable based on the static movement cost. Each other location in the digital elevation map is then assigned a corresponding movement cost based on the static movement cost, generating a two-dimensional cost layer for that time slice. This process is repeated until a two-dimensional cost layer for each time slice is obtained. Each two-dimensional cost layer for each time slice consists of the accessibility status or movement cost of each location corresponding to that time slice.

[0130] According to the time sequence, the two-dimensional cost layers corresponding to each time slice within the planned time period are combined to form the three-dimensional spatiotemporal cost map.

[0131] In an alternative embodiment, the path search module 204 is specifically used for:

[0132] Starting from the aforementioned starting point, nodes are expanded in the three-dimensional spatiotemporal cost graph; wherein, the expansion of nodes is carried out simultaneously in the spatial and temporal dimensions, the expansion in the spatial dimension is based on a predefined geometric neighborhood, and the expansion in the temporal dimension is unidirectionally increasing along the time axis to the next time slice.

[0133] Based on the movement cost of the corresponding position of the node in the three-dimensional spatiotemporal cost map, calculate the cumulative cost during the node expansion process;

[0134] Based on the accumulated cost, and in accordance with the principle of minimizing cost, a path is searched from the starting point to the ending point, and the path with the minimum accumulated cost is determined as the optimal path.

[0135] In an alternative embodiment, the path planning module 205 is specifically used for:

[0136] The Douglas-Peucker algorithm is used to extract key path points from a series of location points contained in the optimal path;

[0137] Based on the key path points, a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover is generated using a cubic spline curve fitting method.

[0138] In an alternative embodiment, the path planning module 205 is specifically used for:

[0139] Based on the distance of the planetary rover's single-step movement control, the sampling interval on the smooth movement trajectory is determined;

[0140] According to the sampling interval, extract multiple location points from the smooth movement trajectory;

[0141] The plurality of location points are arranged in order along the smooth movement trajectory to form the navigation point sequence.

[0142] It should be noted that the beneficial effects of the long-distance path planning system 200 for planetary rovers in time-varying environments provided in the above embodiments are the same as those of the long-distance path planning method for planetary rovers in time-varying environments described above, and will not be repeated here. Furthermore, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation processes are detailed in the method embodiments, and will not be repeated here.

[0143] The long-distance path planning system 200 for a planetary rover in a time-varying environment of the present invention can be a computer program (including program code) running on a computer device. For example, the long-distance path planning system 200 for a planetary rover in a time-varying environment of the present invention is an application software that can be used to execute the corresponding steps in the long-distance path planning method for a planetary rover in a time-varying environment of the present invention.

[0144] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned long-distance path planning methods for planetary rovers in time-varying environments. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the long-distance path planning method for planetary rovers in time-varying environments shown in any embodiment of the present invention by calling the computer program.

[0145] In one optional embodiment, an electronic device is provided, as shown in FIG10. The electronic device 4000 shown in FIG10 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0146] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0147] Bus 4002 may include a path for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent bus 4002 in Figure 10, but this does not mean that there is only one bus or one type of bus.

[0148] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0149] The memory 4003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0150] Among them, electronic devices can also be terminal devices. A terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0151] It should be noted that the electronic device shown in Figure 10 is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0152] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned long-distance path planning methods for planetary rovers under time-varying environments.

[0153] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0154] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0155] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A long-distance path planning method for a planetary rover under time-varying conditions, characterized in that, include: Based on digital elevation maps and solar position information, the distribution of illumination and shadow areas for multiple time slices within the planning period is generated. Based on the digital elevation map, a static movement cost representing terrain accessibility is generated; The distribution of illumination and shadow regions in each time slice is fused with the static movement cost to form a three-dimensional spatiotemporal cost map that dynamically changes over time. In this three-dimensional spatiotemporal cost map, starting and ending points containing spatial coordinates and time information are used as the starting and ending targets. Node expansion is performed by simultaneously expanding nodes in both spatial and temporal dimensions, with the temporal dimension increasing unidirectionally. Path search is conducted by calculating the movement cost between nodes based on the three-dimensional spatiotemporal cost map and following the principle of minimizing cost, resulting in an optimal path composed of a series of nodes containing spatial coordinates and time information. Key path points are extracted from the optimal path, and a smooth movement trajectory satisfying the rover's kinematic constraints is generated based on these key path points. A sequence of navigation points for controlling the rover's movement is then generated based on the smooth movement trajectory. The step of generating the distribution of illumination and shadow areas for multiple time slices within a planned time period based on a digital elevation map and solar position information includes: constructing a terrain occlusion angle database based on the digital elevation map; wherein the terrain occlusion angle database is used to record the terrain occlusion angle of each location in the digital elevation map at different azimuths; calculating the solar azimuth angle and solar altitude angle corresponding to each time slice within the planned time period according to the solar position information; querying the corresponding terrain occlusion angle from the terrain occlusion angle database according to the solar azimuth angle corresponding to each time slice; determining whether each location in the digital elevation map is in illumination or shadow in that time slice by comparing the solar altitude angle and terrain occlusion angle corresponding to any time slice, and marking each location as an illuminated area or a shadowed area, generating the distribution of illumination and shadow areas for that time slice, until the distribution of illumination and shadow areas for each time slice is obtained. The step of fusing the distribution of illumination and shadow areas of each time slice with the static movement cost to form a three-dimensional spatiotemporal cost map that dynamically changes over time includes: for any time slice within the planning period, setting shadow areas as impassable based on the distribution of illumination and shadow areas of that time slice, setting obstacle areas as impassable based on the static movement cost, and assigning a corresponding movement cost to each of the remaining locations in the digital elevation map based on the static movement cost, generating a two-dimensional cost layer for that time slice, until a two-dimensional cost layer for each time slice is obtained; wherein, the two-dimensional cost layer for each time slice consists of the access status or movement cost of each location corresponding to the corresponding time slice; and combining the two-dimensional cost layers corresponding to each time slice within the planning period in chronological order to form the three-dimensional spatiotemporal cost map.

2. The long-distance path planning method for a planetary rover under time-varying conditions according to claim 1, characterized in that, The step of generating a static movement cost characterizing terrain accessibility based on the digital elevation map includes: calculating the terrain slope value and terrain roughness value at each location in the digital elevation map; identifying locations where the terrain slope value is greater than the climbing ability threshold of the rover or the terrain roughness value is greater than the obstacle crossing ability threshold as obstacle areas; and performing normalized cost calculation on the terrain slope value and terrain roughness value of each location in the digital elevation map that does not belong to the obstacle area to generate the static movement cost.

3. The long-distance path planning method for a planetary rover under time-varying conditions according to claim 1 or 2, characterized in that, The steps of obtaining an optimal path consisting of a series of nodes containing spatial coordinates and time information in the three-dimensional spatiotemporal cost graph include: starting from the starting point, expanding nodes in the three-dimensional spatiotemporal cost graph; expanding nodes simultaneously in both spatial and temporal dimensions with unidirectional increments in the temporal dimension; calculating the movement cost between nodes based on the three-dimensional spatiotemporal cost graph and performing path search according to the principle of minimizing cost; starting from the starting point, expanding nodes in the three-dimensional spatiotemporal cost graph; wherein, node expansion is performed simultaneously in both spatial and temporal dimensions, with spatial expansion based on a predefined geometric neighborhood and temporal expansion unidirectionally incrementing along the time axis to the next time slice; calculating the cumulative cost during node expansion based on the movement cost of the corresponding position of the node in the three-dimensional spatiotemporal cost graph; and searching for a path from the starting point to the ending point based on the cumulative cost and the principle of minimizing cost, and determining the path with the minimum cumulative cost as the optimal path.

4. The long-distance path planning method for a planetary rover under time-varying conditions according to claim 3, characterized in that, The step of extracting key path points from the optimal path and generating a smooth trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points includes: using the Douglas-Peucker algorithm to extract key path points from a series of location points contained in the optimal path; and using a cubic spline curve fitting method to generate the smooth trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points.

5. The long-distance path planning method for a planetary rover under time-varying conditions according to claim 4, characterized in that, The step of generating a navigation point sequence for controlling the movement of the planetary rover based on the smooth movement trajectory includes: determining a sampling interval on the smooth movement trajectory based on the distance of the planetary rover's single-step movement control; extracting multiple position points from the smooth movement trajectory according to the sampling interval; and arranging the multiple position points in order on the smooth movement trajectory to form the navigation point sequence.

6. A long-distance path planning system for a planetary rover under time-varying conditions, employing the long-distance path planning method for a planetary rover under time-varying conditions as described in any one of claims 1 to 5, characterized in that, include: The first generation module is used to generate the distribution of illumination and shadow areas for multiple time slices within the planned time period based on digital elevation maps and solar position information; The second generation module is used to generate a static movement cost that characterizes terrain accessibility based on the digital elevation map. The fusion generation module is used to fuse the illumination and shadow region distribution of each time slice with the static movement cost to form a three-dimensional spatiotemporal cost map that changes dynamically over time. The path search module is used in the three-dimensional spatiotemporal cost map to expand nodes by taking the starting point and ending point containing spatial coordinates and time information as the starting and ending targets, expanding nodes in both spatial and time dimensions and increasing the time dimension in one direction, and performing path search based on the three-dimensional spatiotemporal cost map to calculate the movement cost between nodes and follow the principle of minimizing the cost, so as to obtain the optimal path consisting of a series of nodes containing spatial coordinates and time information. The path planning module is used to extract key path points from the optimal path, generate a smooth movement trajectory that satisfies the kinematic constraints of the planetary rover based on the key path points, and generate a sequence of navigation points for controlling the movement of the planetary rover based on the smooth movement trajectory.

7. An electronic device, characterized in that, The electronic device includes a processor coupled to a memory, the memory storing at least one computer program, which is loaded and executed by the processor to enable the electronic device to implement the long-distance path planning method for a planetary rover in a time-varying environment as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which, when executed by a processor, implements the long-distance path planning method for a planetary rover in a time-varying environment as described in any one of claims 1 to 5.

Images