Task-oriented spatiotemporal optimal path evaluation method and system

Abstract

The present invention belongs to the technical field of off-road mission path planning, and particularly relates to a task-oriented space-time optimal path evaluation method and system, which constructs a space-time comprehensive environmental impact model and performs rasterization processing, divides space-time into space-time unit cubes with the same space and time interval along the coordinate axis of the space-time coordinate system, and uses the space-time unit cubes as space-time voxels; in the space-time prism that can be reached by the task activity, at a fixed time interval, the geometric center of the space-time voxel is used as a node, and the nodes at different times are connected by uniform linear motion to construct an alternative space-time path section; based on the principle that spatial trajectories can overlap or intersect, in the alternative space-time path sections of different time periods, the comprehensive threat cost is minimized as the goal to search and obtain the optimal path of the comprehensive threat cost in the space-time environment. The present invention uses the space-time prism range, grid scale and space-based mask method to control the amount of calculation within a reasonable limit, realizes the space-time integrated optimal path planning, and is more optimized and efficient.

Description

Technical Field

[0001] The present invention belongs to the technical field of off-road mission path planning, and in particular relates to a task-oriented spatiotemporal optimal path evaluation method and system. Background Art

[0002] In a specific field, a mission refers to a single-stage action in physical space. A mission is a broad concept. It is an orderly collection of a series of interrelated actions carried out by an action unit to fulfill its responsibilities or achieve a specific purpose under certain environmental and spatiotemporal constraints, including actions in multiple stages in physical and non-physical space, and the connection and coordination between the stages. There is a mutual influence or dependency between the actions at each stage. It is a complex union. In addition to the basic environmental impact, the more important thing is the clever strategy. The spatiotemporal environmental effectiveness evaluation for combat missions is not a complex problem that can be solved by any field of knowledge alone. It involves many environmental problems, analysis methods, solutions, and knowledge in related fields. The main purpose of environmental impact analysis is to analyze the objective influence of environmental factors. The corresponding analysis conclusions have no direct relationship with specific tasks, but the results can support further comprehensive impact analysis for specific tasks. The spatiotemporal prism problem studies the boundary problem of human activities, that is, the maximum spatiotemporal range that can appear within a period of time when the speed of human movement is affected by the environment. Studying the space-time prism in off-road activities can evaluate the off-road accessibility as a whole, analyze the space-time range, key points, space-time tolerance and other issues that may occur in off-road activities, so as to achieve the purpose of evaluating the task effectiveness as a whole. In traditional methods, the task-oriented comprehensive space-time environmental analysis adopts a two-dimensional mathematical model of spatial environmental impact to uniformly express the comprehensive impact of the environment on the task. Although the data model of the two-dimensional spatial environmental impact can express the environmental cost law of different spaces, it cannot reflect the impact of environmental changes over time, and has certain limitations in the environmental assessment of task execution plans. The purpose of global path planning in space and time is to find a continuous space-time path connecting the starting and target space-time points. As with the shortest time path planning, it can be approximated by several segments of space straight lines connected end to end. A more important step in this process is to construct a path to search for alternative space-time path segments. In the shortest time path, in order to achieve the purpose of minimizing the time cost, the corresponding space-time straight line segment can be constructed as an alternative route segment using the maximum safe speed in each voxel. This method will result in different spatiotemporal slopes (speeds) of the "spacetime straight line segments" in each voxel, but it is still a boundary problem, and the overall computational effort is controllable. To evaluate the environment from the perspective of the comprehensive threat cost, it is necessary to find the path with the minimum comprehensive cost in spacetime. For the voxel-based minimum comprehensive threat cost spacetime path, although the path selection is locally limited by the maximum speed and overall by the spacetime prism, since the comprehensive threat level cost is not a boundary problem, there are many possible optional paths within the feasible area. If the path search is still performed in units of single voxels, it will lead to a problem of explosive computational effort. Summary of the invention

[0003] To this end, the present invention provides a task-oriented spatiotemporal optimal path evaluation method and system, which constructs an environmental comprehensive impact model based on spatiotemporal voxels, uses spatiotemporal trajectories to obtain the environmental comprehensive cost of the task execution plan, and evaluates the environment from the perspective of comprehensive threat cost based on the principle of searching for the optimal path in the spatiotemporal environment, and finds the path with the minimum comprehensive cost, so as to facilitate the scientific and effective selection of corresponding plan strategies in task execution, use the spatiotemporal prism to limit the range of optional paths, control the calculation amount within reasonable limits, and realize the optimal path planning of off-road mission activities in spatiotemporal integration. Compared with the manual path planning method, the path planning in this case can be more scientific, optimized, and more efficient.

[0004] According to the design scheme provided by the present invention, a task-oriented spatiotemporal optimal path evaluation method is provided, comprising:

[0005] Construct a task-oriented spatiotemporal comprehensive environmental impact model in the spatiotemporal coordinate system to describe the cost of the comprehensive environmental threat level;

[0006] The spatiotemporal integrated environmental impact model is rasterized, and the space-time is divided into space-time unit cubes with the same space and time interval along the coordinate axis of the space-time coordinate system, and the space-time unit cubes are used as space-time voxels;

[0007] In the reachable space-time prism of the task activity, at fixed time intervals, the geometric center of the space-time voxel is used as the node, and the nodes at different times are connected by uniform linear motion to construct the alternative space-time path segment;

[0008] Based on the principle that spatial trajectories can overlap or intersect, the optimal path with the comprehensive threat cost in the space-time environment is searched among the alternative space-time path segments in different time periods with the goal of minimizing the comprehensive threat cost.

[0009] As the task-oriented spatiotemporal optimal path assessment method of the present invention, further, the spatiotemporal comprehensive environmental impact model expands the time dimension on the basis of the spatial environmental cost law, and uses the spatiotemporal coordinate system to record the cumulative growth rate of the environmental cost of the spatial point at the time point due to the environment, situation and ability factors, wherein a point on the surface is determined as the origin, the east direction is the X-axis, the north direction is the Y-axis, and the time T is the vertical Z-axis to establish a three-dimensional rectangular coordinate system, and the three-dimensional rectangular coordinate system is used as the spatiotemporal coordinate system.

[0010] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, in the rasterization processing, the space is rasterized at the same spatial interval along the X-axis and Y-axis of the spatiotemporal coordinate system, and each spatial grid determines a spatial grid center; it is divided at the same time interval along the T-axis, and the minimum space-time unit after the time and space rasterization is used as the space-time voxel for simplifying the complexity of environmental spatial changes and temporal changes.

[0011] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, in a rasterized spatiotemporal voxel environment, according to the shortest time required to travel along a spatial straight line at the maximum safe speed allowed by the environment and reach the center of the target spatial grid, the shortest time required for the corresponding spatiotemporal point to reach the center of the adjacent spatial grid is obtained, the earliest arrival spatiotemporal point at the center of each spatial grid is solved to approximate the lower pyramid surface of the space-time prism, and the latest stay spatiotemporal point at the center of each spatial grid is solved to approximate the upper pyramid surface of the space-time prism; the spatiotemporal trajectory of the corresponding segment movement is used as a continuous spatiotemporal trajectory consisting of multiple spatiotemporal straight line segments connecting the current spatiotemporal point and the center of the target spatial grid and connected end to end; Each space-time straight line segment is in a different space-time voxel, and except for the current space-time point and the center of the target space grid, the space-time starting point and ending point of each segment are the space-time coordinates of the space-time trajectory entering and passing through the space-time voxel; the space-time trajectory tracking uses the space-time coordinates of the last space-time voxel passed through as the space-time starting point of the next space-time straight line segment in sequence, and calculates the space-time coordinates when passing through the space-time voxel in combination with the spatial movement direction, the space-time range of the space-time voxel and the maximum speed of traveling in a specific direction in the space-time voxel until it reaches the center of the target space grid in space; according to the reachable domain range of the target object represented by the space-time prism composed of the lower pyramid surface and the upper pyramid surface, the driving path of the task executor must be within this range.

[0012] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, the spatiotemporal trajectory of the task execution plan is tracked by using the spatiotemporal voxels in the spatiotemporal comprehensive environmental impact model; and the total cost accumulation value of the task execution plan is obtained based on the sum of the cost values ​​generated when the spatiotemporal trajectory passes through each spatiotemporal voxel, and the total cost value is used to evaluate the quality of the execution plan.

[0013] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, the spatiotemporal coordinates of the start and end of the task execution plan are set, and straight line segments are constructed according to the spatiotemporal points entering and exiting the space-time voxels to obtain the spatiotemporal trajectory of the task execution plan; the spatiotemporal coordinates of the space-time points where the task execution enters or exits the space-time voxels in the model are used to obtain the travel length and residence time in each space-time voxel on the space-time trajectory, and the comprehensive environmental cost of the task execution plan is obtained through the cost growth rate of each space-time voxel passed by the space-time trajectory, the residence time in the corresponding space-time voxel, and the travel length.

[0014] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, assuming that the cost generated by the task execution scheme in the i-th spatiotemporal voxel is Then the comprehensive environmental cost of the task execution plan is Among them, n represents the number of space-time voxels that the space-time trajectory passes through in sequence, represents the growth rate of the i-th spatiotemporal voxel cost, represents the time that the space-time trajectory stays in the i-th space-time voxel, Represents the length of the space-time trajectory in the i-th space-time voxel.

[0015] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, the path search adopts time period hierarchical search, and the calculation starts from the first calculation time period inside the reachable domain. When calculating a certain section, each node on the lower boundary of the reachable domain in the time period or the target voxel center point of the previous time period is used as the starting point, and the alternative spatiotemporal paths to the target voxel center point of the time period are calculated on the premise that these nodes meet the speed conditions at that time; multiple alternative paths connected to each target voxel center point in the current calculation time period are evaluated, and only one path with the minimum cumulative comprehensive threat cost is retained, and other alternative paths are discarded, and the minimum cumulative comprehensive threat cost to the current target voxel center point is evaluated in combination with the minimum comprehensive threat cost of the corresponding starting point; if the reachable spatiotemporal domain contains part of the upper boundary of the spatiotemporal prism in the current calculation, each node on the upper boundary is used as the target voxel center point to participate in the calculation, and the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the final target point is evaluated; enter the next calculation time period until all calculations in the reachable domain are completed; find a path with the minimum cumulative comprehensive threat cost of all paths to the off-road end point as the cost optimal path.

[0016] As a task-oriented spatiotemporal optimal path evaluation method of the present invention, further, parallel computing and / or heuristic search are used in path search, or spatial position masks are used to sample the target spatiotemporal voxel center points or reduce the spatiotemporal voxel resolution to increase the spatiotemporal voxel spatiotemporal interval, so as to optimize the path search calculation amount.

[0017] Furthermore, the present invention also provides a task-oriented spatiotemporal optimal path evaluation system, comprising: a model construction module, a model solution module, an alternative construction module and an optimal output module, wherein:

[0018] A model building module is used to build a task-oriented spatiotemporal comprehensive environmental impact model in a spatiotemporal coordinate system for describing the cost of the comprehensive environmental threat level;

[0019] A model solving module is used for rasterizing the spatiotemporal integrated environmental impact model, dividing the space-time into space-time unit cubes with the same space and time interval along the coordinate axis of the space-time coordinate system, and using the space-time unit cubes as space-time voxels;

[0020] An alternative construction module is used to construct an alternative space-time path segment by connecting nodes at different times with uniform linear motion within the reachable space-time prism of the task activity at fixed time intervals and taking the geometric center of the space-time voxel as the node;

[0021] The optimal output module is used to search for the optimal path with the comprehensive threat cost in the space-time environment in the alternative space-time path segments in different time periods based on the principle that spatial trajectories can overlap or intersect, with the goal of minimizing the comprehensive threat cost.

[0022] Beneficial effects of the present invention:

[0023] The present invention constructs an environmental comprehensive impact model based on space-time voxels, uses space-time trajectories to obtain the environmental comprehensive cost of the task execution plan, evaluates the environment from the perspective of comprehensive threat cost based on the principle of searching for the optimal path in the space-time environment, finds the path with the minimum comprehensive cost, expands the time dimension on the basis of the spatial cost law of the task environment to realize the situation where any point in space and the point continuously change in time, obtains the comprehensive environmental cost on the space-time trajectory of the task execution plan by rasterizing the model, and conducts a comprehensive and reasonable task plan environmental evaluation to achieve the purpose of scientifically and effectively selecting corresponding strategies in task execution, uses the space-time prism range to limit the optional paths, effectively controls the amount of calculation, and realizes the optimal path planning of space-time integration. Compared with the manual planning method, the present invention is more optimized and more efficient, and has a good application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The following is a schematic diagram of the task-oriented spatiotemporal optimal path evaluation process in the embodiment;

[0025] Figure 2 It is a schematic diagram of the space-time coordinate system in the embodiment;

[0026] Figure 3 This is a schematic diagram of the space-time environment model after rasterization processing in the embodiment;

[0027] Figure 4 This is a schematic diagram of the space-time trajectory calculation process in the embodiment;

[0028] Figure 5 It is a schematic diagram of the space-time path and the space-time prism space in the embodiment;

[0029] Figure 6 Schematic diagram of a space-time path segment to be selected for a space-time path search node in an embodiment;

[0030] Figure 7 This is a schematic diagram of the process of reducing the amount of calculation by using a spatial position mask in the embodiment;

[0031] Figure 8 This is an illustration of the overlapping or intersecting spatial trajectories in the embodiment;

[0032] Fig. 9 This is a diagram showing the relationship between the time consumption of the time-space minimum cost path planning and the time difference between the starting and target points in the embodiment;

[0033] Fig.10 The experimental results of B133 in the embodiment are shown;

[0034] Fig.11 This is a schematic diagram of the path planning for avoiding danger in time and space with the minimum threat cost in the embodiment;

[0035] Fig.12 This is a schematic diagram of the minimum threat cost path planning for concealed maneuvers in the embodiment;

[0036] Fig.13 This is a schematic diagram of the minimum threat cost path planning for firepower maneuver in the embodiment;

[0037] Fig.14 This is an illustration of the minimum threat cost path planning for rapid fire action in an embodiment. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the present invention clearer and more understandable, the present invention is further described in detail below in conjunction with the accompanying drawings and technical solutions.

[0039] The shortest time path planning in a predictable dynamic space-time environment and the space-time reachable domain evaluation based on space-time prism are both based on the shortest time consumption cost. This type of method provides an overall, general, and rough evaluation of the energy efficiency of the environment, which is very useful for understanding the adaptability of the environment to the task as a whole. However, it lacks a clear energy efficiency evaluation, that is, to what extent the participants in the task activity can best perform in the existing environment. For this problem, it is necessary to evaluate the overall threat level and cost. The environment is evaluated from the perspective of the task, and the path with the lowest comprehensive cost is found to answer the best performance of the task activity participants in the corresponding environment. To this end, the embodiment of the present invention provides a task-oriented spatiotemporal optimal path evaluation method, see Figure 1 As shown, including:

[0040] S101. Construct a task-oriented spatiotemporal comprehensive environmental impact model in a spatiotemporal coordinate system to describe the cost of the comprehensive environmental threat level;

[0041] S102, rasterizing the spatiotemporal integrated environmental impact model, dividing the space-time into space-time unit cubes with the same space and time interval along the coordinate axis of the space-time coordinate system, and using the space-time unit cubes as space-time voxels;

[0042] S103. In the achievable space-time prism of the task activity, at fixed time intervals, with the geometric center of the space-time voxel as the node, at the speed allowed by the environment at that time, use uniform linear motion to connect the nodes at different times to construct an alternative space-time path segment;

[0043] S104. Based on the principle that spatial trajectories may overlap or intersect, the optimal path with the comprehensive threat cost in the space-time environment is searched for with the goal of minimizing the comprehensive threat cost in the alternative space-time path segments in different time periods.

[0044] By constructing an environmental comprehensive impact model based on space-time voxels, using space-time trajectories to obtain the comprehensive environmental cost of the task execution plan, based on the principle of searching for the optimal path in the space-time environment, the environment is evaluated from the perspective of comprehensive threat cost, and the path with the minimum comprehensive cost is found, which facilitates the scientific and effective selection of corresponding plan strategies in task execution. The number of voxels inside the space-time prism is used to limit the optional paths, effectively control the amount of calculation, reduce the combination of computing equipment operations, and achieve low calculation output complexity and high efficiency.

[0045] The two-dimensional space environmental impact model can show the laws of environmental costs in different spaces, but it cannot reflect the impact of environmental changes over time. To this end, in the embodiment of this case, the time-space comprehensive environmental impact model expands the time dimension based on the laws of spatial environmental costs, and uses the space-time coordinate system to record the cumulative growth rate of environmental costs of space points at different time points due to environmental, situational and capability factors in task execution, wherein a point on the surface is determined as the origin, the eastward direction is the X-axis, the northward direction is the Y-axis, and the time T is the vertical upward Z-axis to establish a three-dimensional rectangular coordinate system, and the three-dimensional rectangular coordinate system is used as the space-time coordinate system.

[0046] like Figure 2 As shown in the figure, (a) represents the space-time coordinate system of two-dimensional space, and (b) represents the space coordinate system of three-dimensional space. Both coordinate systems are three-dimensional coordinate systems, but the coordinates have completely different meanings. The definition of the space-time coordinate system of two-dimensional space is consistent with the similar coordinate system in space-time geography. In order to uniformly express the dynamic changes of spatial environment over time, the embodiment of this case uses a space-time comprehensive environmental impact model to establish Figure 2 The space-time coordinate system described in the above is used to record the environmental impact law at any time and space. The space-time growth rate of the comprehensive environmental cost is recorded at any position (x, y, t) in the space-time coordinate system. It is used to represent the rate of change of the cumulative cost of the environment, situation, and ability on task execution at the spatial point (x, y) at time t. The meaning of is related to the cost. For the shortest time planning, It is the minimum time required to move a unit distance, that is, the shortest time consumption cost rate As for the comprehensive path cost analysis, is the time-space comprehensive threat cost rate

[0047] In the rasterization process, the space is rasterized at the same spatial interval along the X-axis and Y-axis of the space-time coordinate system, and each space grid determines a space grid center; it is divided at the same time interval along the T axis, and the smallest space-time unit after the time and space rasterization is used as the space-time voxel for simplifying the complexity of environmental spatial and temporal changes.

[0048] The shortest time required for the corresponding space-time point to reach the center of the adjacent space grid is obtained according to the shortest time required to travel along the space straight line at the maximum safe speed allowed by the environment and reach the center of the target space grid in the gridded space-time voxel environment. The earliest arrival space-time point at the center of each space grid is solved to approximate the lower pyramid surface of the space-time prism, and the latest stay space-time point at the center of each space grid is solved to approximate the upper pyramid surface of the space-time prism; the space-time trajectory of the corresponding segment movement is taken as a continuous space-time trajectory consisting of multiple space-time straight line segments connecting the current space-time point and the center of the target space grid and connected end to end; each space-time straight line segment is at a different time. In the empty voxel, except for the current space-time point and the center of the target space grid, the space-time starting point and ending point of each segment are the space-time coordinates of the space-time trajectory entering and passing through the space-time voxel; the space-time trajectory tracking uses the space-time coordinates of the last space-time voxel passed through as the space-time starting point of the next space-time straight line segment movement in sequence, and calculates the space-time coordinates when passing through the space-time voxel in combination with the spatial movement direction, the space-time range of the space-time voxel and the maximum speed of traveling in a specific direction in the space-time voxel until it reaches the center of the target space grid in space; according to the reachable domain range of the target object represented by the space-time prism composed of the lower pyramid surface and the upper pyramid surface, the path of the task executor must be within this range.

[0049] Mission along L m The total space-time cost Applicable to the task execution plan moving along any space-time trajectory. By integrating the overall space-time cost formula, the relevant environmental cost can be obtained, but the changes in time and space of the environment are irregular, and there are difficulties in the actual calculation process. Therefore, in the embodiment of this case, the space-time environment is rasterized, such as Figure 3 As shown, along the XYT three axes, the space-time is divided into space-time unit cubes (space-time voxels, abbreviated as voxels) with the same space and time interval. It is approximately considered that the space-time growth rate of the comprehensive cost of the environment in the same space-time unit is The environment affiliation problem in the rasterization process can be set as follows: in space, the rasterized environment belongs to the part of the environment that has the greatest impact on the task, and in time, the state of the environment belongs to the state of the environment at the geometric center of the voxel. It expresses the time-space rate of cost accumulation. In addition to being related to the time-space environment, it may also be related to other factors, such as the driving direction during task execution, historical trajectory, etc. Therefore, even after the environment is rasterized, It is not necessarily a numerical value, but represents a law of cost accumulation.

[0050] In theory, the process of understanding the impact of the environment is to But in fact, for all voxels in the space-time region It is unrealistic to perform all the calculations once, and the cost of performing such a huge amount of environmental assessment is huge, which will lead to explosion problems in the calculation process. To address this problem, in the embodiment of this case, the spatiotemporal trajectory can be used, that is, which voxel is used in the assessment process to generate which voxel, how many voxels are used, and how much environmental assessment is performed.

[0051] The significance of discretization is to simplify complex laws of change. When the gridded space-time interval is of appropriate size, it can be approximately considered that the environment in the space-time it represents is similar, and all its cost laws are also consistent. Figure 3 As shown, if (x0, y0, t0) is the geometric center point of the green voxel shown in the figure, w′(·) is the approximate expression of the cost w(·) in the voxel, then w can be approximately expressed by the following formula:

[0052]

[0053] As a task-oriented spatiotemporal optimal path evaluation method in an embodiment of the present invention, further, the spatiotemporal trajectory of the task execution plan is tracked in the spatiotemporal comprehensive environmental impact model using spatiotemporal voxels; and the total cost accumulation value of the task execution plan is obtained based on the sum of the cost values ​​generated at each spatiotemporal voxel in the spatiotemporal trajectory, and the total cost value is used to evaluate the degree of quality of the execution plan. Further, the spatiotemporal coordinates of the start and end of the task execution plan are set, and the spatiotemporal trajectory of the task execution plan is obtained based on the spatiotemporal points entering and exiting the spatiotemporal voxels. Further, the spatiotemporal coordinates of the spatiotemporal points where the task execution enters or exits the spatiotemporal voxels in the model are used to obtain the travel length and residence time in each spatiotemporal voxel on the spatiotemporal trajectory, and the comprehensive environmental cost of the task execution plan is obtained through the cost growth rate of each spatiotemporal voxel passed by the spatiotemporal trajectory and the residence time and travel length in the corresponding spatiotemporal voxel.

[0054] To evaluate the cost of task execution schemes based on the spatiotemporal comprehensive environmental impact model, we first need to obtain which spatiotemporal voxels the known scheme will pass through, as well as the length of time it stays in each spatiotemporal voxel and the spatial distance it moves. This process is the spatiotemporal tracking process of the movement trajectory. Movement in the spatiotemporal comprehensive environmental impact model can be regarded as a series of continuous straight-line motions, and any complex motion trajectory can be approximated in this way. Figure 4 The figure shows a straight-line motion from a spacetime point S to a spacetime point E in a spacetime. A portion of voxels A0, B0, and B1 are drawn in the figure, where A0 and B0 are adjacent in space and time, and B0 and B1 are adjacent in space and time. Spacetime point S is the space point A on its spacetime trajectory L A A spacetime point on the spacetime trajectory L. B A space-time point on the voxel. Assume that the task execution plan starts from S and follows the space-time trajectory shown in the figure, passing through space-time points M0 and M1 to reach E. The space-time coordinates of the endpoints of each straight line segment in the specific plan are known (such as the space-time coordinates of S and E in the figure are known), and the driving speed and direction of this section of the task journey are known. If the space-time coordinates of each space-time point entering or passing through the voxel are calculated (such as the space-time coordinates of M0 and M1 in the figure), the travel length and residence time of the space-time trajectory in each voxel of the path (such as the three voxels of A0, B0, and B1 in the figure) can be calculated.

[0055] When solving the specific model, the driving direction i can be calculated by the spatial coordinates of A and B using the formula i = arctan ((y B -y A ) / (x B -x B )) is calculated. This value does not change when each voxel between these two points moves. The speed in the x and y axes is According to the solution, we can get: Let upper(x) and floor(x) be the downward and upward rounding functions, respectively. We can define forward(x) and backward(x) as the backward and forward rounding functions, respectively. The grid intervals of x, y, and t axes are d x ,d y ,d t , let the coordinates of the start of movement in the j-th space-time voxel be marked as (x j ,y j ,t j ), the maximum remaining movement time within the voxel and the geometric center coordinates of the voxel It can be expressed as:

[0056]

[0057] exist During this period, without considering the spatial boundaries of the voxels, the maximum travel distances in the x and y axes are Without considering the time boundary of space-time voxels, the maximum driving distances in the x and y axes are If other axes are not considered, the shortest time to pass through the T axis, X axis, and Y axis is The corresponding formula is Then the earliest exit time Δt j Expressed as The axial direction corresponds to the shortest time.

[0058] After completing the above calculations, if This means that the end point has not been reached after passing through the jth voxel, and the user will continue to move into the j+1th voxel. The coordinates when entering are Repeat the above steps in the next voxel until the nth voxel can reach the end point, in which case Δt n Corrected to And end the calculation.

[0059] The above method can be used to track the motion of the straight line segment in the voxel and obtain the space-time coordinates of the space-time trajectory when it enters and passes through each voxel. The trajectory is a uniform linear motion in the voxel, that is, the straight line segment connecting the space-time points entering and passing through the voxel (such as Figure 4 The straight line segment M0M1 in the middle) can be used to track which voxels a specific off-road plan will pass through, as well as the length of time it stays in each voxel and the spatial distance it moves.

[0060] If the space-time trajectory of the task execution plan obtained by mobile trajectory tracking will pass through n space-time voxels in sequence, the cost growth rate of the i-th space-time voxel is The trajectory stayed in this spacetime voxel for the time The length of the travel space is Then the cost W that the task execution plan will generate in the i-th voxel of the model is i for The total cost W of the whole scheme is

[0061] Time geography is a universal method for understanding the interdependence between humans, nature, and technology. It is generally studied in a conceptual coordinate system consisting of n-dimensional space and 1-dimensional time. Time and space are orthogonal, and individual activities are unidirectional and irreversible in the time dimension. The basic elements of time geography are space-time paths and space-time prisms. The space-time prism problem studies the boundary problem of human activities, that is, the maximum space-time range that can appear when the speed of human movement is affected by the environment within a period of time. Studying the space-time prism in off-road activities can evaluate the off-road accessibility as a whole, analyze the space-time range size, key points, space-time tolerance capabilities, and other issues that may appear in off-road activities, so as to achieve the purpose of evaluating the mission effectiveness as a whole.

[0062] like Figure 5As shown, the space-time trajectory of a stationary object can be represented in the space-time coordinate system as a straight line parallel to the time axis (such as L s ); the space-time path of a continuously moving object appears as a continuous curve (such as L m ); spacetime point P′ Start and P′ end The space-time region U between the departure space-time point and the arrival space-time point P′ is called end A certain space-time prism, expressing the space-time range in which the moving object may appear during this period of time, P′ end P′ Start They are called the upper and lower anchor points of the space-time prism respectively; PPA is the abbreviation of Potential Path Area, which represents the spatial range in which an individual may appear within the time budget. A (x A ,y A ,t A ),P B (x B ,y B ,t B ) is a uniform linear motion between two points, and the speed calculation formula is:

[0063]

[0064] This value is exactly L m The slope of the space surface. The smaller the slope value, the faster the speed of movement in space. s As shown, if an object is stationary, it should appear as a straight line parallel to the time axis. The expression of reachability by the space-time prism is mainly based on the following principle: if the space-time area that can be reached by any space-time trajectory that does not exceed the maximum speed from the starting point is recorded as N; if the space-time area that can be reached by any space-time trajectory that does not exceed the maximum speed before the end point is recorded as R, then it is as follows:

[0065] U={(x,y,t)|(x,y,t)∈N∩(x,y,t)∈R}

[0066] Anchor point P′ Start and P′ end The determined space-time prism U is a space-time that satisfies both the space-time that can be reached from a specified starting point at a specified time and the space-time that can reach the destination before another specified time. It reflects the possibility of an individual's travel and activities within a certain time budget, starting point, and destination. The space-time area expressed by the space-time prism is the reachable space-time area studied in this case. The space-time area not within this range is subject to time budget, environmental influences, and rule restrictions, and it is either impossible to reach from the starting point or impossible to reach the destination on time.

[0067] The space-time prism is used to obtain the reachable area in the task execution plan under predictable space-time conditions. The space-time prism is used to represent the space-time boundaries that may appear in the task execution plan, which is convenient for evaluating the feasibility, reliability, space-time redundancy and other issues of the task when it is executed in the corresponding space-time. It takes into account the continuous changes of the environment over time, does not lead to a large number of invalid calculations, and solves the problem that a large number of space-times in the traditional algorithm are actually impossible to reach in a changing environment. It also reduces the computational load, has low computational output complexity and high efficiency. In this case, the geometric center of the voxel is used as the node inside the space-time prism, and the integer multiple of the voxel unit time interval is used as the time interval. The method of connecting the starting point and the target space-time point with uniform linear motion is used to construct a search for alternative space-time path segments to perform space-time path search. Since the number of voxels inside the space-time prism is limited, the selected paths are limited, and the amount of calculation can be effectively controlled. When performing the next path search for a certain voxel center, refer to Figure 6 As shown, the selectable space-time path segments can be constructed according to the following steps:

[0068] (1) Taking the starting space-time point as the starting point and the integer multiple voxel unit time interval as a calculation time period, calculate the space-time boundary range that the starting space-time point can reach within this period;

[0069] (2) Find all voxels that contain the target time within the spatial range that can be reached from the starting time and space point, and within the overall cross-country reachable time and space domain, and use the geometric center points of these voxels as the center points of the target voxels;

[0070] (3) Use a space-time straight line (uniform linear motion) to connect the starting space-time point to all target space-time points, and the corresponding connecting line is the alternative space-time trajectory for the next step of searching at the starting space-time point;

[0071] The speed and direction of the vehicle are calculated by the slope of the alternative space-time line, and then the space-time trajectory of the voxels along the space-time line can be calculated. It can be approximately considered that the environmental influence law within a space-time voxel is constant. The cost growth rate of the alternative path segment H in the i-th space-time voxel is The trajectory stayed in this spacetime voxel for the time The length of the travel space is Therefore, the cost incurred in the i-th voxel is If H passes through q voxels, and the cumulative comprehensive threat cost of H's previous segment H-1 is D H-1 , then the cumulative comprehensive threat level cost of H is D H according to Calculate. H It represents the minimum cumulative comprehensive threat level cost required to reach the center of the currently calculated target voxel through path H starting from the starting point of the task activity.

[0072] When searching for a path, the calculation can be performed hierarchically according to the calculation time period. Starting from the first calculation time period inside the reachable domain, the calculation is performed according to the following steps: starting from each node on the lower boundary of the reachable domain within the time period, the alternative spatiotemporal path to the center point of the target voxel in the time period is calculated. If the current calculation time period is not the first time period, the alternative paths from the center points of all target voxels in the previous calculation time period to the center point of the target voxel in the current time period should also be calculated; evaluating that there are multiple alternative paths connecting each target voxel center point in the current calculation time period, only one path with the minimum cumulative comprehensive threat cost is retained, and the other alternative paths are discarded. The minimum cumulative threat cost to the center point of the current target voxel is evaluated in combination with the minimum comprehensive threat cost of the corresponding starting point; if within a calculation time period, the reachable spatiotemporal domain contains part of the upper boundary of the spatiotemporal prism, each node on the upper boundary is used as the center point of the target voxel to participate in the above calculation, and the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the final target point of the off-road is evaluated and recorded; entering the next calculation time period until all calculations in the reachable domain are completed; finding the path with the minimum cumulative comprehensive threat cost of all paths to the end point of the off-road is the path with the optimal cost. Furthermore, in the embodiments of the present case, parallel computing and / or heuristic search may be used in path search, or the center point of the target space-time voxel may be sampled using a spatial position mask or the space-time voxel resolution may be reduced to increase the space-time voxel space-time interval, so as to optimize the path search calculation amount.

[0073] Furthermore, based on the above method, an embodiment of the present invention also provides a task-oriented spatiotemporal optimal path evaluation system, comprising: a model construction module, a model solution module, an alternative construction module and an optimal output module, wherein:

[0074] A model building module is used to build a task-oriented spatiotemporal comprehensive environmental impact model in a spatiotemporal coordinate system for describing the cost of the comprehensive environmental threat level;

[0075] A model solving module is used for rasterizing the spatiotemporal integrated environmental impact model, dividing the space-time into space-time unit cubes with the same space and time interval along the coordinate axis of the space-time coordinate system, and using the space-time unit cubes as space-time voxels;

[0076] An alternative construction module is used to construct an alternative space-time path segment by connecting nodes at different times with uniform linear motion within the reachable space-time prism of the task activity at fixed time intervals and taking the geometric center of the space-time voxel as the node;

[0077] The optimal output module is used to search for the optimal path with the comprehensive threat cost in the space-time environment in the alternative space-time path segments in different time periods based on the principle that spatial trajectories can overlap or intersect, with the goal of minimizing the comprehensive threat cost.

[0078] The principle of searching for the optimal path in a spatiotemporal environment can be implemented based on the following three principles:

[0079] The principle of selecting the travel speed is that driving at the maximum safe speed can achieve the mission goal the fastest, but for many tasks, the fastest does not mean the best. For example, the average fuel consumption of the vehicle will increase significantly after exceeding the economic speed, and deliberately reducing the speed can sometimes obtain better economic benefits; for example, when affected by severe weather conditions, stopping and waiting for the severe weather conditions to improve before continuing to drive can effectively reduce the overall risk. The goal of minimum cost path planning is to reduce the cost. In other words, instead of always using the maximum safe speed, it is better to drive at any free speed that can bring less cost. Just adhere to one principle: the actual speed adopted cannot exceed the maximum safe speed under environmental constraints.

[0080] The relationship between the space-time path and the reachable domain. The space-time reachable domain assessment based on the space-time prism defines the space-time boundary of the reachable domain for the cross-country operation. The space-time area outside this boundary is either inaccessible at the maximum speed or cannot reach the designated mission target on time. Therefore, no matter what kind of space-time trajectory is used, it is impossible to exceed the reachable space-time boundary defined by the space-time prism. Similarly, the planned space-time trajectory cannot pass through any dangerous escape space-time range defined by the space-time prism.

[0081] The principle that spatial trajectories can overlap or intersect. Spatial trajectory overlap or intersection does not refer to the situation of three-dimensional Euclidean space caused by the three-dimensional road network, but refers to the overlap or intersection of different parts of the same trajectory in the original space. Traditional GIS-based path planning almost follows the same principle: spatial trajectories will not overlap or intersect. This is because the intersection of spatial trajectories will lead to closed loop paths in space, which means that the consumption of the corresponding sections will be considered as a waste, resulting in meaningless cost increases. However, the overlap and intersection of spatial trajectories in special cases are meaningful, especially in the face of complex real-world environments. For example, the task undertaker has an important event to complete at a certain "future" moment in the current space, but a serious destructive event is about to occur. The task undertaker must first avoid the destructive event and then come back to complete the relevant tasks. Planning the space-time trajectory of the above tasks is very likely to result in the intersection of spatial trajectories, or even the path to escape the danger completely overlaps with the return path after the danger has passed, but this kind of path planning that overlaps or intersects in space is of special significance. Since any movement takes time, although the trajectories overlap in the spatial coordinate system, such as Figure 8 (b), but they cannot overlap in the space-time coordinate system, as shown in Figure 8 In a spatiotemporal environment, we only need to pay attention to whether the spatiotemporal path avoids the spatiotemporal area where the danger occurs, or achieves a lower cost, without deliberately avoiding the overlap or intersection of the solution trajectory in space.

[0082] This solution can reflect the impact of continuously changing environment, situation, capability and other comprehensive factors, support spatial overlapping and intersecting paths, can cleverly use time difference to solve spatial conflict problems, and can support various tasks with different preferences. This algorithm can be used to evaluate the best performance of the comprehensive threat cost of task activities. Although the amount of calculation is controlled within an affordable range through certain strategies, the amount of calculation is still relatively large because it is necessary to search for the path with the minimum cost in both space and time. This problem can be solved by parallel computing and introducing heuristic search solutions.

[0083] In order to verify the effectiveness of this solution, the following is a further explanation of the effectiveness of this solution combined with test data:

[0084] The first set of experiments: computational complexity experiments

[0085] Scenario: A classic scenario with isotropic speed that does not change over time. It is independent of the situation. The speed in each direction is the same and the speed does not change over time. Environment setting: A gentle isotropic environment with flat terrain and no obstacles. The speed of any task undertaker in each direction is 1m / s and does not change over time. Situation setting: No situation. Only the task undertaker is considered in the environment. Ability setting: A setting used for scenario analysis. The driving speed is independent of the direction and is specified by the environment. No rest or food is allowed during off-road driving. Task type setting: Non-war action. Off-road action is only affected by the environment. It mainly considers issues related to economic consumption, considers fewer risks, and is independent of the situation.

[0086] The above scenario settings are mainly used to verify the characteristics of the algorithm. The algorithm complexity is discussed in the following combinations. Assume that the starting point and the target point have the same spatial position, the off-road time length is 2T, the space-time intervals of the voxels are marked as Δt and Δx respectively, the calculation time period length is selected as nΔt, and the maximum off-road safe speed is v. Then within a calculation time length, the maximum number of possible space-time trajectory segments m for a certain starting space-time point is m=floor(π(vnΔt / Δx) 2 ), then in the kth calculation time period of the lower half pyramid (k=1,2,...,floor(T / (nΔt))), there may be M paths to be selected k =m·floor(π(kvnΔt / Δx) 2 ). Therefore, the equivalent relationship between the total number of paths M that need to evaluate the comprehensive threat cost within the entire space-time prism is M~(v / Δx) 4 T 3 nΔt, its computational complexity is the fourth order of (v / Δx), the third order of the half time T, and the first order of nΔt.

[0087] The above is the equivalent formula for the minimum cost path search when the space-time prism is a regular pyramid. The space-time prism calculation amount for oblique pyramids and other special-shaped pyramids will be even smaller. In general, the minimum cost search in the feasible space-time domain defined by the space-time prism will increase rapidly with the increase of v,T, and decrease rapidly with the increase of Δx.

[0088] In general applications, v, n, and Δt change only to a limited extent, but the change in T is mainly controlled by task requirements, which is difficult to predict and may be large. The only way to reduce the amount of computation is to increase Δx. Fortunately, the order of Δx is greater than the order of T. Although increasing Δx will lead to a decrease in the level of detail of the plan, the rate of decrease is faster than the rate of increase in the amount of computation. This means that no matter how large T is, by adjusting Δx, within a certain amount of computation, a solution that keeps the total level of detail unchanged can be obtained (for example, the number of straight line segments that make up the path remains unchanged). In this case, there are two ways to increase Δx, namely, through the mask method and through the method of reducing the resolution.

[0089] The method of reducing the amount of calculation by using spatial position mask is mainly due to the high amount of search calculation, which is that each starting time-space point corresponds to a large number of target voxel centers, but the cost difference between these candidate paths is not large in most cases. Therefore, the method of discarding non-critical target points in some target voxel centers can be used to quickly reduce the amount of calculation. This method is similar to increasing the distance between the two target voxel centers, which indirectly achieves the effect of increasing Δx.

[0090] The simplest way to exclude the center points of target voxels is to sample the center points of target voxels at equal intervals. However, the regular exclusion of certain points can easily lead to the loss of key information, such as the loss of key nodes such as bridges and river crossings. To solve this problem, this paper proposes a mask-based method to sample the center points of target voxels. Figure 7 As shown in the figure: the spatial center points of the voxels are sampled at an integer multiple of Δx spacing (referred to as the mask reference length in this article, denoted as Δy) in space, and the spatial coordinates of the sampling points are recorded on the mask; the voxel center point coordinates containing key position information are supplemented and recorded additionally on the mask; in the process of space-time path search, when selecting the starting space-time point and the target voxel center point, only the spatial coordinate positions recorded on the mask are considered.

[0091] Since this method still uses the original voxel resolution when calculating the comprehensive threat cost, the calculation accuracy is very high, but at the same time the degree of reduction in calculation amount is not as good as the method of directly reducing the resolution.

[0092] The method of reducing resolution can reduce the computational complexity of spatiotemporal path search. Compared with the method of using masks, a more direct method is to reduce the spatiotemporal resolution of voxels. The method of reducing resolution directly increases Δx, and reduces the computational complexity of comprehensive threat cost evaluation on each path, which quickly reduces the computational complexity at a speed of 4 orders, but the disadvantage is that the accuracy is poorer than the method of using masks.

[0093] The effectiveness of this solution is further illustrated by a computational comparison experiment. There are 8 experiments in total, and the specific parameters are shown in Table 1.

[0094] Table 1. Computational comparison experiment details of minimum comprehensive threat cost path analysis

[0095]

[0096]

[0097] Among them, the aforementioned experiment introduced the classic that isotropic and does not change with time, which will not be repeated here.

[0098] The long-distance off-road complex simulation environment is set as follows: It is independent of the situation, has anisotropic speed, and has a large amount of scene data. The environment will change continuously over time, and the ferry's passability will change periodically, taking into account the biological habits. It is used to verify the complex changes in passability under the influence of the environment and biological habits. It includes the following combinations:

[0099] Environmental setting: Comprehensive simulation environment. The terrain used in this environment setting is ASTER GDEM 2 DEM data, with a latitude and longitude range of E 120°~121°, N 41°~42°. The data of rivers, lakes and artificial building facilities are collected from OpenStreetMap. The assumptions of clay content of the four related soil types are shown in Table 2.

[0100] Table 2 Soil quality settings in environmental settings

[0101]

[0102] Referring to Anderson's related research, it is assumed that the study area is expected to have heavy rain from 8:30 to 11:30, and then the meteorological conditions will be good. The soil moisture content will gradually return to the original level within 28 hours after the rain. The prediction of the moisture content change of the above four soil types under the combined effects of rainfall, temperature, terrain and vegetation is shown in Table 3. The moisture content between given moments can be obtained by interpolation in time.

[0103] Table 3 Predicted values ​​of water content changes in four types of soil

[0104]

[0105]

[0106] There are ferries in some river sections. The ferries are only available during the service time period. The service timetable is shown in Table 4. Bridges can be used at any time (bridge speed limit v b :40km / h).

[0107] Table 4 Ferry service schedule

[0108]

[0109] The real environment is more complex than the above simulation environment, but this setting can already represent various types of environmental impacts to a certain extent. A more complex environment will only show different rules and will not affect the scientific judgment of the experimental results. In order to facilitate data generation and control environmental variables, no more complex environments will be set.

[0110] Situation setting: Same as the no situation setting in the above content;

[0111] Ability setting: long-distance cross-country troop simulation ability setting;

[0112] The main performance parameters of off-road equipment are shown in Table 5, where VCI1 and VCI 50 It is the motor vehicle cone index of off-road vehicles, which refers to the rated cone index that the soil needs to meet at least once and the rated cone index that the soil needs to meet at least 50 times. When the rated cone index (RCI) of the soil is less than VCI1, the vehicle cannot pass through, and when it is greater than VCI 50 The vehicle's passability is not restricted by soil quality. The flat ground speed refers to the speed at which the vehicle generally travels when the road conditions are good, and the minimum speed refers to the minimum speed that the vehicle can accept for long-term driving. The maximum climbing grade refers to the maximum slope that the vehicle can travel in the direction of the slope. In order to ensure driving safety and sufficient power and prevent rollover, it is stipulated that the slope must be less than the maximum climbing grade regardless of the driving direction. Due to the nature of the vehicle, this setting does not allow the vehicle to pass through artificial buildings.

[0113] Table 5 Off-road parameters of motor vehicles

[0114]

[0115] According to the research on path finding of dump trucks in large open pit mines by Choi et al., the speed of the vehicle has different effects on uphill and downhill driving. As shown in Table 6, due to safety and physiological needs, it is usually necessary to arrange meals and rest in time during long-distance off-road driving. In addition, rest for 5 minutes every 55 minutes of driving.

[0116] Table 6 Meals and rest arrangements

[0117]

[0118] Mission type: Non-combat operations. Off-road operations are only affected by the environment and mainly consider issues related to economic consumption. There are fewer risks and they are not related to the situation.

[0119] High spatial resolution, low temporal resolution: spatial resolution: 90 meters; temporal resolution: 120 seconds.

[0120] Low spatial resolution, low temporal resolution: spatial resolution: 500 meters; temporal resolution: 120 seconds.

[0121] Based on the above experimental conditions and parameter settings, experiments A011, A012, A013, A014, and A015 are designed to verify the equivalent relationship between the total number of comprehensive threat cost paths M~(v / Δx) 4 T 3 nΔt, and the only difference between them is the length of the time difference between the starting point and the end point. The first 7 experiments can be repeated 5 times each, and the consumed computing time is listed in Table 7, and the relationship between the computing time of the first 5 experiments and T is plotted as a statistical Fig. 9 As can be seen from the figure, as T increases, the computation time increases very quickly. After fitting, it is exactly a cubic relationship with a very high degree of conformity (R of the fitting curve is 2 Very close to 1), the experiment verified that M~(v / Δx) 4 T 3 The accuracy of nΔt.

[0122] Experiments A035 and A115 respectively implement the method of reducing the amount of calculation by using a mask and the method of reducing the resolution. From the experimental results, it can be concluded that the method of directly or indirectly increasing Δx can quickly and effectively reduce the amount of calculation; by comparing A035 and A115, it can be seen that the mask method can ensure a certain accuracy, but the speed of reducing the amount of calculation is not as fast as the method of directly reducing the resolution; by comparing A011, A015 and A115, it can be seen that the speed of reducing the amount of calculation due to the increase in the spatial resolution Δx of the voxel is faster than the speed of increasing the amount of calculation due to the increase in T.

[0123] Different from the previous seven experiments, experiment B133 was conducted in a long-distance off-road complex simulation environment using a masking method. Its purpose was to verify that the masking method would not cause the loss of important spatial information nodes. Fig.10 The starting point and the end point of the off-road trip are on both sides of the river, connected only by a narrow bridge. If the target voxel center point is simply sampled by a certain spatial spacing method, the bridge may be excluded, resulting in failure to reach the destination. However, when generating a spatial mask, the bridge is retained as a key point through the mask, so it can participate in the path search process. Fig.10 The light-medium thin line represents the space-time path planned by experiment B133, and the semi-transparent space-time prism is the reachable domain of the off-road vehicle. As can be seen from the result diagram (c), the route connects the starting point and the target point through a bridge, and the result is relatively reasonable.

[0124] Table 7 Average calculation time of related experiments repeated 5 times

[0125]

[0126] The second set of experiments: verification experiments on overlapping or intersecting spatial trajectories

[0127] Following the principle that spatial trajectories can overlap or intersect, which has strong practical significance, the experiment of avoiding danger is used to illustrate this problem and show the effect of the algorithm calculation results. This experiment uses a simulation scene of firepower maneuver; space-time rasterization scheme: high spatial resolution, low temporal resolution.

[0128] The simulation scenario of firepower maneuver is set as follows: under the influence of environment and situation, the red side performs a complex simulation scenario of concealed maneuver mission. The environment will change continuously over time, and the ferry's passability will change periodically. Considering the biological habits, the units of the red and blue sides have different ability values ​​and different mission types. It includes the following combinations:

[0129] Environment setting: the same as the comprehensive simulation environment in the above content;

[0130] Situation setting: Red-Blue confrontation situation, the situation setting is based on a comprehensive simulation environment, and the participating troops are divided into two parties, red and blue. The intelligence information held by the red side includes: some blue troops are scheduled to start at 11 o'clock from the suspected deployment area B LArea1 Move to the designated area. In addition, B FPoint1 The troops will carry out patrol missions from 11:00 to 14:00 every day.

[0131] Red side settings:

[0132] Ability setting: The same as the long-distance cross-country troop simulation ability setting in the above content;

[0133] Mission type: Firepower maneuver, cross-country operations have a certain risk-bearing capacity, but the known risks should be minimized. Its characteristics are: avoid encountering the enemy's troops; be cautious when entering the enemy's control area, firepower coverage area, observation post coverage area and suspected deployment area

[0134] Blue unit B LArea1 With B FPoint1 set up:

[0135] Capability setting: The patrol force simulation capability setting is the same as the long-distance cross-country force simulation capability setting. The difference is that the vehicle performance is different, as shown in Table 8.

[0136] Table 8 Off-road parameters of motor vehicles

[0137]

[0138] In addition to the simulation scenario settings for firepower maneuvers, the experiment adds the following settings: The current space-time coordinates of the Red team are (120.378°E, 4 137°N, 8:30am.), and according to the news, the Blue team will attack B between 9:00am. and 9:10am. FArea1 Concentrated artillery attack, but the Red team has an important task to complete at the current coordinates at 9:40am. Please plan a space-time trajectory for the Red team to avoid the artillery attack and return on time in the current scene.

[0139] The scheme in this case is used to plan the path with the minimum time-space comprehensive threat cost. The results are as follows: Fig.11 As shown. The dark space-time prism in the figure is B FArea1 The red team's dangerous escape space-time prism when the area is bombarded, the translucent space-time prism is the red team's reachable space-time prism, and the gray space-time curve is the planned space-time path. From the results in the figure, it can be seen that the space-time path planned for the red team successfully bypassed the dangerous space-time range of the blue team's bombardment and arrived at the departure space position before the deadline. The result is scientific and reasonable. Obviously, the round-trip part of the planned space-time path is completely overlapped in space, which also shows that the overlapping or crossing of spatial trajectories has an important application in space-time path planning.

[0140] The third group of experiments: comparative experiments through the spatiotemporal optimal path planning of different task types

[0141] Four experiments were conducted to verify the differences in the minimum comprehensive threat cost paths for different types of tasks. The threat weights of various types of tasks can be set based on expert experience, or by using hierarchical analysis to set the initial weights of various threats for various tasks, and then fine-tuning the weights based on the characteristics of the tasks or the corresponding activity strategies. Special note on the experimental results: The original calculation results of the experiment are presented by the GLTF model, but Fig.10 As shown in the figure, the planned path drawn by the model is relatively thin and difficult to observe on the original image of the complex scene. The important information in the screenshot is enhanced on the basis of being faithful to the original experimental results. The image enhancement method is limited to bolding and changing the color of the planned path and outlining the key space-time prism. All modifications will not affect the authenticity and scientificity of the results.

[0142] Experiment 1: Non-war action. This experiment is the same as the experiment numbered B133 in the first group of experiments. Planning results analysis: Fig.10 (a) It can be seen that the starting point and destination point of the planning calculation are on both sides of the river and are connected by a bridge. The planned space-time path directly connects the starting point and the destination point through the bridge, and the spatial structure of the path is reasonable. Fig.10 The planning area shown in (a) is flat. During the travel, it is set according to the long-distance off-road complex simulation environment scenario. There is no rainfall and the soil water content is low. In theory, the maximum safe driving speed can be achieved in most environments in the planning area. However, from the perspective of the time axis, the planned path does not stick to the bottom of the space-time prism and does not take the fastest way to reach the destination, but drives at a slightly lower speed. This is due to the characteristics of the mission. The mission is a non-war action. According to the setting of the mission characteristics, it does not have high time requirements, but is more sensitive to fuel economy speed and distance. See Table 9. Therefore, the method of reducing the driving speed is adopted to reduce the cost on the premise of being able to reach the target point on time. The result is scientific and reasonable.

[0143] Table 9 Threat weights of non-war actions

[0144]

[0145] Experiment 2: Concealed maneuver (task characteristics: cross-country operations are only affected by the environment, mainly considering issues related to economic consumption, considering fewer risks, and having nothing to do with the situation). This experiment uses a simulation scenario of concealed maneuver (which is roughly the same as the simulation scenario using firepower maneuver mentioned above, the difference is that the Red Army's mission type is set to concealed maneuver, and the cross-country operation requires the task to be completed as quickly as possible while ensuring concealment. Its characteristics are: avoid encountering the enemy's troops, avoid entering the enemy's firepower coverage area, firepower point coverage area, observation post coverage area, suspected deployment area and control area); spatiotemporal rasterization scheme: low spatial resolution, low temporal resolution; the weight evaluation of the cost of various threats for this task is shown in Table 10; starting point P start and the target point P end The time and space coordinates are (120.263°E, 41.661°N, 9:00am.), (120.538°E, 41.151°N, 13:30pm.). The planning results are as follows Fig.12 As shown, the dark color is the dangerous space-time area, the translucent gray space-time prism is the Red side's situation area or the Red side's off-road reachable area, and the dark gray curve is the planned space-time path. Since the covert maneuver requires the Red side to avoid any contact with the Blue side, all the Blue side's situation areas will form the orange dangerous escape space-time prism shown in the figure for the Red side, so the Red side's route should avoid all orange space-time areas.

[0146] Table 10 Threat weights of concealed maneuvers

[0147]

[0148] Analysis of planning results: combined Fig.12 (a) and (d) show that the Red team crosses the first river and then joins B. FArea1 The dangerous time periods of the areas overlap, so it is necessary to avoid B spatially. FArea1 The escape space-time prism formed; due to the situation, from the space-time reachable domain, it can be seen that the Red team cannot reach the target from the left side, so it can only move towards Bridge A and reach the target from the right side through Bridge A; combined with Fig.12 (a), (b), (c), and (e) show that although the blue side is in B VPoint3 The monitoring area coincides with the planned route, but due to some reasons B VPoint3 The monitoring point shown will only be effective at 12:00am. Therefore, before this time, B VPoint3 The monitoring points will not pose a threat to the Red Army's covert maneuvers; combined with Fig.12 (a), (c), and (e) show that after the Red team passes the second river through Bridge A, they will be at the Blue team's firing point B. FPoint1 Close, but due to continuous rain, Blue B FPoint1 The patrol time and space range is limited, especially from Fig.12 (c) It can be seen that after crossing the river, the Red side has plenty of time to pass through B while maintaining a considerable distance in space and time from the Blue side. FPoint1 The vicinity of patrol units; Fig.12 As shown in (c), in order to avoid the negative effects of hunger, fatigue, etc. caused by the long-distance march of the Red Army, it is planned to have a lunch break before crossing Bridge A.

[0149] From the comprehensive results, it is very difficult for the Red side to perform concealed maneuvers due to the influence of the Blue side's situation. The cross-country reachable area is very narrow and the time window is tight. It is very challenging to reach the target point without any contact with any Blue side. However, the algorithm of this case can cleverly utilize the time difference of situation information and the impact of rainfall on the patrol range of the Blue side. In the continuously changing environment under the comprehensive influence of the situation, a scientific and reasonable concealed maneuvering time and space plan is successfully planned. The result meets the mission expectations and can achieve the comprehensive minimum threat cost.

[0150] Experiment 3: Firepower maneuver (task characteristics, the same as the above-mentioned activity maneuver). This experiment uses the same simulation scenario as the above-mentioned firepower maneuver; the spatiotemporal rasterization scheme: low spatial resolution, low temporal resolution; the weights of various threat costs for this task are shown in Table 11; the spatiotemporal coordinates of the starting point and target point are the same as those in Experiment 2. The planning results are as follows Fig.13As shown in Figure 2, the dark and semi-transparent gray space-time regions and the dark gray space-time curve have the same meaning as those in Experiment 2. Due to the characteristics of firepower mobility and greater risk-bearing capacity, compared with Experiment 2, in Experiment 3, B VPoint1 ,B VPoint2 ,B VPoint3 ,B VPoint4 ,B FPoint1 ,B RArea1 Areas such as the red and yellow areas are no longer considered dangerous space-time areas, but only the yellow semi-transparent risk space-time areas shown in the figure are formed. According to the weight preference of firepower mobility for various threats, the space-time path should avoid all orange areas and enter the yellow area with caution.

[0151] Table 11 Threat weights of firepower mobility

[0152]

[0153] Planning result analysis: Experiment 3 is a firepower maneuver. Although it requires caution in entering the risk area, it does not need to be completely avoided. After considering various costs, given that the blue side has a certain ability to bear the risks in the control area, such as Fig.13 (a) and (c) are the algorithm plans to cross the river from the blue side control area B. RAreal1 The edge passes through and reaches the target directly from the left side of the target point. The planning result shows the different characteristics of firepower maneuver and concealed maneuver. Fig.13 (b) and (d) are both viewed from the bottom up, combined with the top down. Fig.13 (e) It can be found that the algorithm selects a path that is consistent with the blue square B. LAreal1 The routes of the mobile forces have a lot of overlapping time and space routes in space. However, the algorithm takes advantage of the time difference between the two sides' movements. Before meeting the blue side, it bypasses the blue side's troops through the gap formed by the difference in mobility, and then moves to the target after the blue side's troops are far away. Fig.13 (f) Lunch and rest can be arranged while waiting for the blue team to maneuver; this plan cleverly uses the difference in temporal and spatial accessibility to complete the task without taking on situational risks; compared with the narrow time window for covert maneuvers in Experiment 2, Fig.13 As shown in (f), the reachable area for executing fire maneuvers is large, time is ample, and reliability is high.

[0154] From the comprehensive results, since the Red side has a greater risk tolerance when executing firepower maneuvers, it does not need to adopt a maneuver strategy similar to the covert maneuver in Experiment 2, which is to trade time for space. It only needs to take a small amount of risk (from the Blue side's control area B RAreal1 The planning scheme fully considers the characteristics of the continuously changing environment, situation, and capabilities, utilizes the time difference of space activities, conforms to the mission risk preference, and the planning results are scientific and reasonable.

[0155] Experiment 4: Rapid firepower action (Task characteristics: off-road operations have certain firepower capabilities. Under the premise of ensuring certain risks, they are required to reach the target quickly. In addition to meeting the relevant requirements of the firepower maneuver mission type, path planning is also required to prioritize speed and distance). This experiment uses a simulation scenario of rapid firepower action (the red team’s mission type in this scenario is rapid firepower action, and other settings are the same as the simulation scenario of covert maneuver in the above content); spatiotemporal rasterization scheme: low spatial resolution, low temporal resolution; the weights of various threat costs for this task are shown in Table 12; the spatiotemporal coordinates of the starting point and target point are the same as those in Experiment 2. The impact of situation on rapid firepower action and firepower maneuver in Experiment 3 is similar, the only difference is that rapid firepower action pays more attention to travel distance and speed. The calculation results are as follows Fig.14 shown.

[0156] Table 12 Threat weights for rapid fire actions

[0157]

[0158] Analysis of planning results: Fig.14 As shown in (a), (b), and (c), the planned path directly passes through the blue side's control area B RAreal1 This is due to the characteristics of rapid firepower operations, which have certain requirements for travel distances and will choose a more direct and faster route under the premise of taking risks; Fig.14 (c) and (d) show that, in addition to the necessary stage breaks, the planned plan will arrive at the destination before the lunch break, avoiding the conflict between rapid fire action and human biological habits caused by long-term continuous action. From the overall results, the planned path of Experiment 4 reflects the characteristics of rapid fire action, and the results are scientific and reasonable.

[0159] This scheme can be used as part of the spatiotemporal environmental effectiveness evaluation for combat missions, so as to obtain the minimum cost scheme for missions with certain risk preferences in a continuously changing environment under the comprehensive influence of the environment, situation, and capabilities, and evaluate the best performance that the mission can achieve in the corresponding environment. Essentially different from the traditional GIS-based spatial path analysis algorithm, the algorithm of this scheme is a spatiotemporal collaborative calculation method, which completes the analysis and calculation in the environmental comprehensive impact model based on spatiotemporal voxels, and belongs to a type of spatiotemporal analysis. The cost of the planning scheme comes from both the accumulation of space and the accumulation of time, and it is an evaluation method in a spatiotemporal integrated environment. The task-oriented spatiotemporal optimal path planning algorithm can reflect the influence of comprehensive factors such as the continuously changing environment, situation, and capabilities, support spatially overlapping and intersecting paths, can cleverly use time differences to solve spatial conflict problems, and can support various tasks with different preferences. The algorithm can be used to evaluate the best performance of off-road missions.

[0160] Based on the above system, an embodiment of the present invention further provides a server, comprising: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0161] Based on the above system, an embodiment of the present invention further provides a computer-readable medium on which a computer program is stored, wherein the program implements the above method when executed by a processor.

[0162] The device provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned system embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned system embodiment.

[0163] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system and device can refer to the corresponding process in the aforementioned system embodiment, and will not be repeated here.

[0164] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limiting, and thus other examples of the exemplary embodiments may have different values.

[0165] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0166] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, systems and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0167] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and systems can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0168] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0169] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the system described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0170] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A task-oriented spatiotemporal optimal path evaluation method, It is characterized in that Contains the following: A task-oriented spatiotemporal comprehensive environmental impact model for describing the cost of the comprehensive environmental threat level is constructed in the spatiotemporal coordinate system; the spatiotemporal comprehensive environmental impact model expands the time dimension on the basis of the spatial environmental cost law, and uses the spatiotemporal coordinate system to record the cumulative growth rate of the environmental cost of the spatial point at the time point due to the environment, situation and capability factors, wherein a point on the surface is determined as the origin, the eastward direction is the X-axis, the northward direction is the Y-axis, and the time T is the vertical upward Z-axis to establish a three-dimensional rectangular coordinate system, and the three-dimensional rectangular coordinate system is used as the spatiotemporal coordinate system; The spatiotemporal comprehensive environmental impact model is rasterized, and the space-time is divided into space-time unit cubes at the same space and time intervals along the coordinate axis of the space-time coordinate system, and the space-time unit cubes are used as space-time voxels; wherein, in the rasterization process, the space is rasterized at the same space interval along the X-axis and Y-axis of the space-time coordinate system, and a space grid center is determined for each space grid; the space is divided at the same time interval along the T axis, and the smallest space-time unit after the space-time rasterization is used as the space-time voxel for simplifying the complexity of environmental spatial and temporal changes; In the reachable space-time prism of the task activity, at fixed time intervals, the geometric center of the space-time voxel is used as the node, and the nodes at different times are connected by uniform linear motion to construct an alternative space-time path segment; wherein, in constructing the alternative space-time path segment, the shortest time required for the corresponding space-time point to reach the center of the adjacent space grid is obtained according to the shortest time required to travel along the space straight line at the maximum safe speed allowed by the environment and reach the center of the target space grid in the gridded space-time voxel environment, the earliest arrival space-time point at the center of each space grid is solved to approximate the lower pyramid surface of the space-time prism, and the latest stay space-time point at the center of each space grid is solved to approximate the upper pyramid surface of the space-time prism; the space-time trajectory of the corresponding segment motion is used as the connection between the current space-time point and the target space grid. The space-time trajectory is composed of multiple space-time straight line segments connected end to end at the center of the grid; each space-time straight line segment is in a different space-time voxel, and except for the current space-time point and the center of the target space grid, the space-time starting point and ending point of each segment are the space-time coordinates of the space-time trajectory entering and passing through the space-time voxel; the space-time trajectory tracking uses the space-time coordinates of the last space-time voxel passed through as the space-time starting point of the next space-time straight line segment in sequence, and calculates the space-time coordinates when passing through the space-time voxel in combination with the spatial movement direction, the space-time range of the space-time voxel and the maximum speed of traveling in a specific direction in the space-time voxel until it reaches the center of the target space grid in space; the space-time prism composed of the lower pyramid surface and the upper pyramid surface is used to describe the reachable domain range of the target object that the task executor must pass through when exercising the path; Based on the principle that spatial trajectories can overlap or intersect, in the alternative spatiotemporal path segments of different time periods, the spatiotemporal trajectory of task activities is tracked using spatiotemporal voxels in the spatiotemporal comprehensive environmental impact model; and the total cost accumulation value of the task execution plan is obtained according to the sum of the cost values ​​generated when the spatiotemporal trajectory passes through each spatiotemporal voxel, and the total cost value is used to evaluate the pros and cons of the execution plan, and the optimal path of the comprehensive threat cost in the spatiotemporal environment is searched with the goal of minimizing the comprehensive threat cost. In the calculation of the cost value generated when the spatiotemporal trajectory passes through each spatiotemporal voxel, the spatiotemporal coordinates of the starting and reaching points of the task execution plan are set, and the spatiotemporal trajectory of the task execution plan is obtained according to the spatiotemporal points of entering and exiting the spatiotemporal voxel; the spatiotemporal coordinates of the spatiotemporal points where the task execution enters or exits the spatiotemporal voxel in the model are used to obtain the travel length and residence time in each spatiotemporal voxel on the spatiotemporal trajectory, and the comprehensive environmental cost of the task execution plan is obtained through the cost growth rate of each spatiotemporal voxel passed by the spatiotemporal trajectory and the residence time and travel length in the corresponding spatiotemporal voxel.

2. The task-oriented spatiotemporal optimal path evaluation method according to claim 1, It is characterized in that Assume that the cost of the task execution plan in the i-th space-time voxel is Then the comprehensive environmental cost of the task execution plan is Among them, n represents the number of space-time voxels that the space-time trajectory passes through in sequence, represents the growth rate of the i-th spatiotemporal voxel cost, represents the time that the space-time trajectory stays in the i-th space-time voxel, Represents the length of the space-time trajectory in the i-th space-time voxel.

3. The task-oriented spatiotemporal optimal path evaluation method according to claim 1, It is characterized in that The path search is performed in layers of time periods, starting from the first calculation time period inside the reachable domain. When calculating one of the time periods, each node on the lower boundary of the reachable domain within the time period or the target voxel center point of the previous time period is used as the starting point to calculate the alternative space-time path from the node to the target voxel center point of the time period under the premise of satisfying the speed conditions at that time; multiple alternative paths connected to the center point of each target voxel in the current calculation time period are evaluated, and only one path with the minimum cumulative comprehensive threat cost is retained, and other alternative paths are discarded. The minimum cumulative comprehensive threat cost to the current target voxel center point is evaluated in combination with the minimum comprehensive threat cost of the corresponding starting point; if the reachable space-time domain contains part of the upper boundary of the space-time prism in the current time period calculation, each node on the upper boundary is used as the target voxel center point to participate in the calculation, and the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the final target point is evaluated; Enter the next calculation time period until all calculations in the reachable domain are completed; find the path with the smallest cumulative comprehensive threat cost among all paths reaching the off-road end point as the cost-optimal path.

4. The task-oriented spatiotemporal optimal path evaluation method according to claim 1 or 3, It is characterized in that In the path search, parallel computing and / or heuristic search are used, or the spatial position mask is used to sample the center point of the target spatiotemporal voxel or the spatiotemporal voxel resolution is reduced to increase the spatiotemporal voxel spatiotemporal interval to optimize the path search calculation amount.

5. A task-oriented spatiotemporal optimal path evaluation system, It is characterized in that It includes: model building module, model solving module, alternative building module and optimal output module, among which, A model building module is used to build a task-oriented spatiotemporal comprehensive environmental impact model for describing the cost of the comprehensive environmental threat level in a spatiotemporal coordinate system; the spatiotemporal comprehensive environmental impact model expands the time dimension on the basis of the spatial environmental cost law, and uses the spatiotemporal coordinate system to record the cumulative growth rate of the environmental cost of the spatial point at the time point due to the environment, situation and capability factors, wherein a point on the surface is determined as the origin, the eastward direction is the X-axis, the northward direction is the Y-axis, and the time T is the vertical upward Z-axis to establish a three-dimensional rectangular coordinate system, and the three-dimensional rectangular coordinate system is used as the spatiotemporal coordinate system; The model solving module is used to perform rasterization processing on the spatiotemporal integrated environmental impact model, divide the space-time into space-time unit cubes at the same space and time intervals along the coordinate axis of the space-time coordinate system, and use the space-time unit cubes as space-time voxels; wherein, in the rasterization processing, the space is rasterized at the same space interval along the X-axis and Y-axis of the space-time coordinate system, and each space grid determines a space grid center; the space is divided at the same time interval along the T axis, and the smallest space-time unit after the rasterization of time and space is used as the space-time voxel for simplifying the complexity of environmental spatial and temporal changes; The alternative construction module is used to construct an alternative space-time path segment by connecting nodes at different times with uniform linear motion within the reachable space-time prism of the task activity at fixed time intervals and with the geometric center of the space-time voxel as the node; wherein, in constructing the alternative space-time path segment, the shortest time required for the corresponding space-time point to reach the center of the adjacent space grid is obtained according to the shortest time required to travel along the space straight line at the maximum safe speed allowed by the environment and reach the center of the target space grid in the gridded space-time voxel environment, the earliest arrival space-time point at the center of each space grid is solved to approximate the lower pyramid surface of the space-time prism, and the latest stay space-time point at the center of each space grid is solved to approximate the upper pyramid surface of the space-time prism; the space-time trajectory of the corresponding segment motion is used as the connection between the current space-time point and the center of the target space grid, and a continuous space-time trajectory consisting of multiple space-time straight line segments connected end to end; each space-time straight line segment is in a different space-time voxel, and except for the current space-time point and the center of the target space grid, the space-time starting point and ending point of each segment are the space-time coordinates of the space-time trajectory entering and passing through the space-time voxel; the space-time trajectory tracking sequentially uses the space-time coordinates of the last space-time voxel passed through as the space-time starting point of the next space-time straight line segment, and calculates the space-time coordinates when passing through the space-time voxel in combination with the spatial movement direction, the space-time range of the space-time voxel and the maximum speed of traveling in a specific direction in the space-time voxel, until it reaches the center of the target space grid in space; the space-time prism composed of the lower pyramid surface and the upper pyramid surface is used to describe the reachable domain range of the target object that the task executor must pass through when exercising the path; The optimal output module is used to track the spatiotemporal trajectory of task activities in the spatiotemporal comprehensive environmental impact model based on the principle that spatial trajectories can overlap or intersect in the alternative spatiotemporal path segments in different time periods by using spatiotemporal voxels; and obtain the total cost accumulation value of the task execution plan according to the sum of the cost values ​​generated when the spatiotemporal trajectory passes through each spatiotemporal voxel, and use the total cost value to evaluate the pros and cons of the execution plan, and search for the optimal path of the comprehensive threat cost in the spatiotemporal environment with the goal of minimizing the comprehensive threat cost, wherein, in the calculation of the cost value generated when the spatiotemporal trajectory passes through each spatiotemporal voxel, the spatiotemporal coordinates of the start and end of the task execution plan are set, and a straight line segment is constructed according to the spatiotemporal points of entering and exiting the spatiotemporal voxel to obtain the spatiotemporal trajectory of the task execution plan; the spatiotemporal coordinates of the spatiotemporal points of the task execution entering or exiting the spatiotemporal voxel in the model are used to obtain the travel length and residence time in each spatiotemporal voxel on the spatiotemporal trajectory, and the comprehensive environmental cost of the task execution plan is obtained through the cost growth rate of each spatiotemporal voxel passed by the spatiotemporal trajectory and the residence time and travel length in the corresponding spatiotemporal voxel.

Images