Task-oriented Multi-dimensional Efficiency Evaluation Method

Abstract

The present invention belongs to the technical field of mission environment effectiveness evaluation, and particularly relates to a task-oriented multi-dimensional effectiveness evaluation method, which constructs an effectiveness evaluation framework for multi-dimensional indicators of a mission under the influence of the environment. The effectiveness evaluation framework is implemented based on a task-oriented spatio-temporal comprehensive environment impact model. Among them, the comprehensive environment impact model uses a function model that records the continuous change of the comprehensive environment with spatial position and time in a spatio-temporal coordinate system to quantify the impact of the environment on the mission. By rasterizing the model, path search based on the shortest time cost and the minimum threat degree cost is used to obtain the reachable spatio-temporal range of mission activities, the degree of adaptation of the mission to the environment, and the optimal execution plan within the spatio-temporal range of feasible solutions, thereby realizing the prediction of mission effectiveness evaluation in multiple dimensions. The present invention can achieve an all-round evaluation of the mission under the influence of the environment, including extreme performance, mission spatio-temporal range, and optimal comprehensive performance, which is more scientific, reasonable, and convenient for practical scenario applications.

Description

Technical Field

[0001] The present invention belongs to the technical field of mission effectiveness evaluation, and particularly relates to a task-oriented multi-dimensional effectiveness evaluation method, which comprehensively evaluates the effectiveness of a mission from multiple levels and multiple perspectives. Background Art

[0002] Task-oriented multi-dimensional effectiveness evaluation does not refer to the research in a certain field, but involves large-scale comprehensive analysis from basic environmental research such as terrain feature extraction, visibility analysis, soil quality and rainfall analysis, to task comprehensive research such as traffic maps and various thematic map analyses, and then to application research such as cross-country navigation and airborne analysis. Task-oriented multi-dimensional effectiveness evaluation is the basis for wargame confrontation, calculation, strategy formulation, etc., and is one of the core topics in the science of geographical environment. The cognitive process of environmental impact is complex, with rich cognitive levels and intertwined analysis processes. There are both clear and specific analysis objectives such as "can it be seen", and vague and abstract understandings such as "is this road dangerous". Most of the related research on environmental effectiveness evaluation has developed from traditional GIS analysis, and the related technologies are widely used, but there are also deficiencies in the actual application field. Most of the related research is based on the analysis of space, and the accumulation of costs or benefits is basically only related to space, ignoring the cumulative effect of time on risks and costs; some problem studies are too simplified, rarely considering or even ignoring the influence of the situation and the capabilities of personnel and equipment of all parties involved, ignoring the influence of the changing environment and situation over time, and not analyzing in detail the influence of different task risk preferences; traditional GIS-based methods are difficult to perform spatio-temporal collaborative analysis and calculation, and their results generally only express the conclusions of spatial analysis.

[0003] In traditional methods, the calculation of the cost or benefit of a task is only related to space, ignoring the cumulative effect of time on cost or benefit. Generally, only spatial conflicts are considered, while spatio-temporal conflicts are ignored. Spatial conflict does not necessarily equal spatio-temporal conflict. In some special cases, spatial conflict can be resolved by the time difference between the actions of both sides. Moreover, the influencing factors are not comprehensively considered and the evaluation index is single. This problem can be analyzed from three aspects: (1) Traditional research tends to ignore the influence of situation, ability, and the biological habits of participants. Compared with the simple environmental influence, in some cases, factors such as situation and ability are the main factors affecting task performance. However, in traditional research, most relevant factors are not comprehensively considered and are even directly ignored. (2) Traditional research faces the problem of a single evaluation index. For example, the analysis result of a traditional off-road task is generally a recommended optimal route, and the performance of the task in the corresponding environment can only be inferred from the performance of this route. The evaluation index of such analysis is single, which is not convenient for analyzing the risk distribution of the task area, difficult to accurately obtain the feasible spatial and temporal ranges of the task plan, unable to evaluate the reliability of the task plan and the overall impact of the situation on the task, etc. (3) Some planning principles of traditional research are contrary to task requirements. Currently, most path analyses for off-road are based on spatial analysis and follow the basic principle that the planned paths should not cross or overlap. However, this is contrary to the real military requirements. In military activities, due to the changes in the battlefield situation and environment, paths with spatial overlap or intersection have their significance. Moreover, the significance of task planning for task effectiveness evaluation lies in that it can find and, through the evaluation of typical plans, understand various performance situations of the task under the comprehensive influence of the environment. However, the effectiveness evaluation of a task is not a process of listing the feasible plans of the task in an unordered manner. A typical plan can only reflect one possible performance of completing the task, establishing a "point-like" understanding of the task performance. No matter how many typical plans are considered, it is difficult to form an overall and continuous evaluation of the task performance, that is, a "surface-like" understanding. The effectiveness evaluation of a task should grasp the essence of the understanding of task effectiveness, gradually transitioning from surface problems to internal problems, from specific problems to abstract problems, and from lower-level cognition to higher-level cognition. Summary of the Invention

[0004] Therefore, the present invention provides a multi-dimensional effectiveness evaluation method for tasks. By using the comprehensive environmental impact model, it fully reflects the continuous changes of the environment and situation over time, uses the time difference of spatial activities to solve the spatial conflict problem, takes factors such as situation and ability as equally important as environmental impact, and gradually establishes a multi-level cognition of task effectiveness from point to surface and then back to point, from specific to abstract and then back to specific, based on extreme performance, the range of reachable spatio-temporal regions, and the minimum comprehensive cost. It is close to actual application, realizes multi-level, multi-angle, and all-round effectiveness evaluation in off-road tasks, is convenient for formulating the overall optimal and minimum-cost task execution plan in terms of time and space, is scientific, reasonable, highly efficient, and convenient for actual scenario application.

[0005] According to the design solution provided by the present invention, a task-oriented multi-dimensional effectiveness evaluation method is provided, including:

[0006] Construct an effectiveness evaluation framework for multi-dimensional indicators of a task under the influence of the environment. This effectiveness evaluation framework is implemented based on a task-oriented spatio-temporal comprehensive environmental impact model. Among them, the comprehensive environmental impact model uses a function model that records the continuous change of the comprehensive environment with spatial position and time in a spatio-temporal coordinate system to quantify the impact of the environment on the task. The multi-dimensional indicators include: evaluating the extreme performance ability of the task with the shortest time to complete the task as an indicator based on the comprehensive environmental impact model, predicting and evaluating the spatio-temporal region boundaries that should be avoided for task activities formed by environmental and situation influences, evaluating the spatio-temporal range that task activities can reach under the influence of environmental capabilities and situations, and the best performance of task activities under the comprehensive influence of the environment.

[0007] Perform rasterization processing on the spatio-temporal comprehensive environmental impact model. Divide the spatio-temporal into spatio-temporal unit cubes along the coordinate axes of the spatio-temporal coordinate system at the same spatial and time intervals, and use the spatio-temporal unit cubes as spatio-temporal voxels.

[0008] Take the starting point and the target point of the spatio-temporal coordinates where the task execution plan has been determined as anchor points, and use the spatio-temporal trajectory tracking algorithm to predict the shortest time cost of the straight line segment of the task execution object moving along a straight line in space-time; solve the spatio-temporal prism through path search based on the shortest time cost, obtain the spatio-temporal range that task activities can reach based on the solved spatio-temporal prism, and search for the optimal path of task activities in the spatio-temporal environment with the minimum comprehensive threat cost as the goal in the spatio-temporal range sections where task activities can reach in different time periods based on the principle that spatial trajectories can overlap or cross.

[0009] Perform multi-dimensional evaluation on the spatio-temporal environmental impact effectiveness of the task based on the reachable spatio-temporal range of task activities, the adaptability of the task to the environment, and the optimal execution plan within the spatio-temporal range of the feasible plan. Among them, the adaptability of the task to the environment is obtained through the shortest time cost exerted by the task in the environment.

[0010] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, the spatio-temporal comprehensive environmental impact model is expressed as a mathematical model F = M(E(T), S(T), P(T)) of environmental variable E(T), situation variable S(T), and ability variable P(T) that changes continuously with time T. Among them, by determining a point on the ground surface as the origin, taking the eastward direction as the X-axis, the northward direction as the Y-axis, and time T as the vertically upward Z-axis, a three-dimensional rectangular coordinate system is established, and this three-dimensional rectangular coordinate system is used as the spatio-temporal coordinate system.

[0011] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, the spatio-temporal prism is solved through path search based on the shortest time cost, and the spatio-temporal range reachable by the task activity is obtained based on the solved spatio-temporal prism. Among them, the spatio-temporal prism includes: a dangerous escape spatio-temporal prism for describing the spatio-temporal area that the task activity should avoid and a reachable domain spatio-temporal prism for describing the reachable spatio-temporal area. The difference set of the spatio-temporal area ranges represented by the reachable domain spatio-temporal prism and the dangerous escape spatio-temporal prism is used to evaluate and predict the spatio-temporal range reachable by the task activity.

[0012] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, in an environment that changes continuously over time, it is assumed that each spatial point that may be passed during cross-country is composed of an earliest arrival time and a latest stay time to form the earliest arrival spatio-temporal point and the latest stay spatio-temporal point of this passing point in the spatio-temporal coordinate system; the earliest arrival spatio-temporal points and the latest stay spatio-temporal points of all possible passing points respectively form continuous lower and upper pyramidal spatio-temporal surfaces for describing the future and past spatio-temporal surfaces; taking the starting spatio-temporal point as the lower anchor point and the arrival spatio-temporal point as the upper anchor point, and using the lower and upper pyramidal spatio-temporal surfaces as the boundary surfaces, the spatio-temporal area between the two boundary surfaces is used as the reachable domain spatio-temporal prism area for representing the spatio-temporal range in which cross-country runners who can complete the task on time may appear.

[0013] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, for the prediction and evaluation of the reachable area, taking the starting spatio-temporal point as the lower anchor point and the arrival spatio-temporal point as the upper anchor point, starting from the starting spatio-temporal point, calculate the earliest arrival time of each spatial point that may be passed during task execution, and use the spatio-temporal points corresponding to these earliest arrival times to form the lower pyramidal boundary surface of the reachable domain spatio-temporal prism; starting from the arrival spatio-temporal point, calculate in reverse the latest stay time that each spatial point that may be passed during task execution can stay on the premise of being able to reach the end point on time, and use the spatio-temporal points corresponding to these latest stay times to form the upper pyramidal boundary surface of the reachable domain spatio-temporal prism; for the obtained upper and lower pyramidal boundary surfaces, perform an intersection operation to obtain the target spatio-temporal range represented by the reachable domain spatio-temporal prism.

[0014] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, the dangerous escape spatio-temporal prism region is divided into an occurrence area for describing the spatio-temporal range where a dangerous event occurs, a affected area for describing the spatio-temporal range where it is impossible to escape from the spatio-temporal range affected by the dangerous event due to limited mobility, and a vacuum area for describing the spatio-temporal range that cannot be reached without experiencing the dangerous event due to limited mobility. Among them, the lower spatio-temporal boundary of the affected area, the spatio-temporal boundaries of the occurrence area except for the upper and lower boundaries, and the upper spatio-temporal boundary of the vacuum area together divide a closed spatio-temporal region. Using this closed spatio-temporal region as the dangerous escape spatio-temporal prism for representing the spatio-temporal region range that must be avoided to complete the task or cannot be reached due to the dangerous event, the lower anchor point of the dangerous escape spatio-temporal prism is the spatio-temporal edge at the start time of the dangerous event, and the upper anchor point is the spatio-temporal edge at the end time of the dangerous event.

[0015] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, in the solution of the dangerous escape spatio-temporal prism, both the upper and lower boundary surfaces are within the spatial range where the dangerous event occurs. Among them, when solving the lower spatio-temporal boundary surface, starting from the moment when the dangerous event occurs, with the maximum speed allowed by the environment and the optimal path and using the spatial edge of the dangerous event as the destination, calculate backward in time the latest departure time for all spatial points within the dangerous event range to escape from danger. The corresponding spatio-temporal points form a continuous pyramidal spatio-temporal surface, and this pyramidal spatio-temporal surface is the lower spatio-temporal edge of the dangerous escape spatio-temporal prism for representing that when the spatio-temporal position of the off-road vehicle is below this spatio-temporal surface, it can escape from the upcoming dangerous event; when solving the upper spatio-temporal boundary surface, starting from the moment when the dangerous event ends, with the maximum speed allowed by the environment and the optimal path and driving inward from the spatial edge of the dangerous event, calculate forward in time the earliest time that all spatial points within the dangerous event range can reach after the dangerous event ends. The corresponding spatio-temporal points form a continuous pyramidal spatio-temporal surface, and this continuous pyramidal spatio-temporal surface is the upper spatio-temporal edge of the dangerous escape spatio-temporal prism for representing that if the current off-road vehicle has not been affected by the dangerous event, it is necessarily impossible to appear below this spatio-temporal surface.

[0016] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, when calculating the spatio-temporal boundary surface to the future or the past in the spatio-temporal voxel, an open list and a determined list are established. The point-like anchor points, scattered multi-point-like anchor points, line-like anchor points, and spatio-temporal points located at the center of the spatial grid near the edge line of the surface-like anchor points are directly added to the determined list. The spatio-temporal points located at the center of the spatial grid inside the surface-like anchor points and with time near the surface-like anchor points are also added to the determined list. The shortest time cost evaluation is carried out on the central points of the spatial grids adjacent to the existing determined list, and the formed spatio-temporal points after evaluation are added to the open list. Search for the spatio-temporal point with the earliest time in the open list and add it to the determined list. Use the path search algorithm to search in eight directions adjacent to the space based on this point, update the earliest arrival time of the adjacent spatio-temporal points in the open list, or evaluate the central points of the spatial grids that have not been evaluated and add them to the open list. Search for the spatio-temporal point with the shortest time in the open list again and add it to the determined list. Repeat this cycle until the earliest possible arrival time of each central point of the spatial grid is calculated. When calculating the latest staying spatio-temporal point at the position of the central point of the spatial grid, start from the spatio-temporal point with the latest staying time in the open list and search backward for the latest departure time of each central point of the spatial grid.

[0017] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, when calculating the spatio-temporal boundary surface to the future or the past in the spatio-temporal voxel, an open list and a determined list are established. The point-like anchor points, scattered multi-point-like anchor points, line-like anchor points, and spatio-temporal points located at the center of the spatial grid near the edge line of the surface-like anchor points are directly added to the determined list. The spatio-temporal points located at the center of the spatial grid inside the surface-like anchor points and with time near the surface-like anchor points are also added to the determined list. The shortest time cost evaluation is carried out on the central points of the spatial grids adjacent to the existing determined list, and the formed spatio-temporal points after evaluation are added to the open list. Search for the spatio-temporal point with the earliest time in the open list and add it to the determined list. Use the path search algorithm to search in eight directions adjacent to the space based on this point, update the earliest arrival time of the adjacent spatio-temporal points in the open list, or evaluate the central points of the spatial grids that have not been evaluated and add them to the open list. Search for the spatio-temporal point with the shortest time in the open list again and add it to the determined list. Repeat this cycle until the earliest possible arrival time of each central point of the spatial grid is calculated. When calculating the latest staying spatio-temporal point at the position of the central point of the spatial grid, start from the spatio-temporal point with the latest staying time in the open list and search backward for the latest departure time of each central point of the spatial grid.

[0018] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, in the solution of the spatio-temporal pyramid, the spatio-temporal trajectory of the movement between the centers of adjacent spatial grids is used as the continuous spatio-temporal trajectory connecting the current spatio-temporal point and the center of the target spatial grid, and the continuous spatio-temporal trajectory is composed of multiple spatio-temporal straight line segments that are connected end to end; each spatio-temporal straight line segment is in a different spatio-temporal voxel, and except for the current spatio-temporal point and the center of the target spatial grid, the spatio-temporal start point and end point of each segment are the spatio-temporal coordinates at which the spatio-temporal trajectory enters and exits the spatio-temporal voxel; the spatio-temporal trajectory tracking sequentially uses the spatio-temporal coordinates of the last time when exiting the spatio-temporal voxel as the spatio-temporal start point of the movement of the next spatio-temporal straight line segment, and combines the spatial movement direction, the spatio-temporal range of the spatio-temporal voxel, and the maximum speed of traveling in a specific direction within the spatio-temporal voxel to calculate the spatio-temporal coordinates when exiting the spatio-temporal voxel, until reaching the center of the target spatial grid in space.

[0019] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, in the rasterized spatio-temporal voxel environment, set the spatio-temporal coordinates at the start and arrival of the task execution plan, and construct straight line segments based on the spatio-temporal points when entering and exiting the spatio-temporal voxel to obtain the spatio-temporal trajectory of the task execution plan; use the spatio-temporal coordinates of the spatio-temporal points when the task execution enters or exits the spatio-temporal voxel in the model to obtain the travel length and residence time within each spatio-temporal voxel passed by the spatio-temporal trajectory, and obtain the comprehensive environmental cost of the task execution plan through the cost growth rate of each spatio-temporal voxel passed by the spatio-temporal trajectory, the residence time and travel length within the corresponding spatio-temporal voxel; and perform hierarchical path search using time periods, starting from the first calculation time period within the reachable domain, using each node on the lower boundary of the reachable domain within this time period or the center point of the target voxel in the previous time period as the starting point, and calculate the alternative spatio-temporal paths from the node to the center point of the target voxel in this time period that meet the current speed condition; evaluate multiple alternative paths connected to the center point of each target voxel in the current calculation time period, only retain the path with the smallest cumulative comprehensive threat cost among them, discard other alternative paths, and combine the minimum comprehensive threat cost of the corresponding starting point to evaluate the minimum cumulative comprehensive threat cost to the center point of the current target voxel; if in the current time period calculation, the reachable spatio-temporal domain contains the upper boundary of the spatio-temporal prism, use each node on the upper boundary as the center point of the target voxel to participate in the calculation, and evaluate the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the final target point; enter the next calculation time period until all calculations within 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 optimal path, and obtain the optimal execution plan within the spatio-temporal range of the feasible plan corresponding to the minimum threat degree cost based on this optimal path.

[0020] As the task-oriented multi-dimensional effectiveness evaluation method of the present invention, further, in the rasterized spatio-temporal voxel environment, based on the time consumed by the linear movement between any two adjacent spatio-temporal nodes, the spatio-temporal trajectory between the starting point and the target point of the task execution plan is obtained through path search to obtain the spatio-temporal trajectory with the least time consumption for the linear movement between the starting point and the target point, and the adaptability of the task to the environment corresponding to the shortest time cost is obtained from this spatio-temporal trajectory.

[0021] Advantages of the present invention:

[0022] 1. Aiming at the problem that the traditional methods have insufficient research on the analysis and expression of risk events, the present invention analyzes and classifies the situation and environmental risks in the task, and believes that the risks in the task execution process are mainly divided into general risks that can be borne, avoidance risks that cannot be borne, and reliability risks that affect the spatio-temporal tolerance ability of the task plan. Aiming at the problems of the expression and evaluation of these risks, through the spatio-temporal expression of risk events, the concepts and evaluations of reliability risks and avoidance risks, a unified mathematical model is constructed for the expression and evaluation of risk events, especially short-term risk events, as the spatio-temporal comprehensive environmental impact model in the effectiveness evaluation framework to describe the continuous changes of task spatio-temporal environmental impact factors over time, and it can scientifically and accurately understand, analyze and evaluate the performance of the task in a complex, changeable and multi-factor dynamic environment.

[0023] 2. Aiming at the problem that some costs accumulated over time are usually ignored in the traditional GIS-based spatial task planning analysis, using the accumulation law of costs in space and time, all planning behaviors should fully consider the cost accumulation in time and the cost accumulation in space. By using the time-accumulated cost, space-accumulated cost and the cost calculation of the task plan in space and time, in the process of evaluating the off-road analysis problem, two spatio-temporal cost indicators, the shortest time cost and the comprehensive threat degree cost, are used to lay the foundation for the off-road planning integrating space and time.

[0024] 3. Aiming at the problem that the comprehensive environmental impact law may change irregularly in both time and space, the comprehensive environmental impact is expressed in the form of spatio-temporal voxels, a comprehensive environmental impact model based on spatio-temporal voxels is constructed, and the spatio-temporal cost evaluation of the off-road task plan is realized in this impact model. Among them, the model can express the impact of the continuous changes of the environment and the situation, and can uniformly express the occurrence and accumulation law of the cost when performing tasks locally under the influence of the environment, the situation, the ability and other factors. Its unified expression of time and space costs can flexibly use the time difference to solve space problems and obtain the basis for the minimum spatio-temporal cost.

[0025] 4. Aiming at the problem that it is difficult for traditional minimum cost path planning algorithms to reflect the impact of continuously changing environments, based on the comprehensive environmental impact model based on spatio-temporal voxels, the shortest time path planning using predictable environments can fully reflect the impact of continuously changing environments on off-road operations. Compared with traditional space-based analysis, the evaluation results are closer to the actual situation, the planning scheme can always achieve a lower comprehensive cost, and the computational complexity is comparable to that of traditional algorithms.

[0026] 5. Conduct research on the reachable domain for off-road in the spatio-temporal domain of continuously changing environments. Based on the comprehensive environmental impact model based on spatio-temporal voxels, aiming at the planning requirements of avoiding dangerous events, use the dangerous spatio-temporal prism and combine other methods to conduct spatio-temporal prediction of the battlefield situation, and then evaluate the off-road maneuverable reachable spatio-temporal domain through the spatio-temporal prism method. Experimental results show that this spatio-temporal reachable domain evaluation method for off-road tasks can reflect the influence of various factors such as environment, situation, capabilities, biological habits, and task characteristics. The evaluation results can very intuitively reflect various important factors such as the activity spatio-temporal range, overall adaptability, risk distribution, and key event nodes of the task, and are suitable for the overall performance analysis of the task.

[0027] 6. Aiming at the problems that traditional space-based GIS analysis methods cannot reflect the accumulation of cost over time and cannot achieve spatio-temporal integrated path planning, conduct minimum comprehensive cost path planning in a continuously changing spatio-temporal environment. Through a large number of experiments, it is proved that the relevant algorithms of the solution in this case can reflect the influence of comprehensive factors such as continuously changing environments, situations, capabilities, and biological habits, support paths with spatial overlaps and intersections, can cleverly use time differences to solve spatial conflict problems, can support various tasks with different preferences, and are suitable for complex spatio-temporal path planning.

[0028] 7. By constructing an integrated environmental impact model, the present invention can fully reflect the continuous changes of the environment and the situation over time, can make full use of the time difference of space activities to solve space conflict problems, and the calculation result of the task effectiveness reflects the unity of time and space; it emphasizes that battlefield situation, the capabilities and living habits of all parties involved, etc. are all factors equally important as environmental impacts, and are included in the comprehensive impact law of the task, so as to realize a more combat-like environmental impact effectiveness evaluation; it emphasizes multi-level, multi-angle and all-round evaluation. The effectiveness evaluation is not an evaluation of a certain aspect of the task, but a multi-angle and all-round evaluation of the task under the influence of the environment. It is a repeated process and a process from specific low-level cognition to gradually rising to abstract high-level cognition. By evaluating the task effectiveness from multiple angles, a comprehensive evaluation of the task's performance ability under the influence of the environment is realized, including evaluating the extreme performance of the task in some aspects, evaluating the spatio-temporal range of the task activities, and evaluating the best comprehensive performance of the task, reflecting a comprehensive understanding of all levels of the task, so as to achieve the purpose of scientifically and effectively formulating strategies during task execution, reducing the operation load and complexity of computing devices, improving the efficiency of model evaluation, and having good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Schematic diagram of the multi-dimensional effectiveness evaluation process for tasks in the embodiment;

[0030] Figure 2 Schematic diagram of spatio-temporal comprehensive environmental impact factors in the embodiment;

[0031] Figure 3 Schematic diagram of the spatio-temporal trajectories of moving objects and stationary objects in the embodiment;

[0032] Figure 4 Schematic diagram of the spatio-temporal coordinate system in the embodiment;

[0033] Figure 5 Schematic diagram of the spatio-temporal path and spatio-temporal prism space in the embodiment;

[0034] Figure 6 Schematic diagram of the spatio-temporal environmental model after rasterization processing in the embodiment;

[0035] Figure 7 Schematic diagram of the spatio-temporal trajectory calculation process in the embodiment;

[0036] Figure 8 Schematic diagram of the spatio-temporal prism structure for dangerous escape in the embodiment;

[0037] Figure 9 Schematic diagram of the calculation process of the lower boundary surface of the spatio-temporal prism in the embodiment;

[0038] Figure 10Schematic of the calculation process of the upper boundary surface of the spatio-temporal prism in the embodiment;

[0039] Figure 11 Schematic of the calculation process of the upper and lower pyramid surfaces of the dangerous escape spatio-temporal prism in the embodiment;

[0040] Figure 12 Schematic of the candidate spatio-temporal path region of the spatio-temporal path search node in the embodiment;

[0041] Figure 13 Schematic of the overlap or intersection of spatial trajectories in the embodiment. Specific implementation manner

[0042] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and technical solutions.

[0043] An embodiment of the present invention provides a task-oriented multi-dimensional effectiveness evaluation method, as shown in Figure 1 shown, including:

[0044] S101. Construct an effectiveness evaluation framework for multi-dimensional indicators of a task under the influence of the environment. The effectiveness evaluation framework is implemented based on a task-oriented spatio-temporal comprehensive environment impact model. Among them, the comprehensive environment impact model uses a function model that records the continuous change of the comprehensive environment with spatial position and time in a spatio-temporal coordinate system to quantify the impact of the environment on the task. The multi-dimensional indicators include: evaluating the extreme performance ability of the task with the shortest time to achieve the task as an indicator based on the comprehensive environment impact model, predicting and evaluating the spatio-temporal region boundaries that the task activities should avoid due to the influence of the environment and the situation, evaluating the spatio-temporal range that the task activities can reach under the influence of the environmental ability and the situation, and the best performance of the task activities under the comprehensive influence of the environment;

[0045] S102. Perform rasterization processing on the spatio-temporal comprehensive environment impact model. Divide the spatio-temporal into spatio-temporal unit cubes along the coordinate axes of the spatio-temporal coordinate system at the same spatial and time intervals, and use the spatio-temporal unit cubes as spatio-temporal voxels;

[0046] S103. Use the starting point and the target point with determined spatio-temporal coordinates of the task execution plan as anchor points, and use the spatio-temporal trajectory tracking algorithm to predict the shortest time cost of the straight line segment of the task execution object moving along a straight line in space-time; Solve the spatio-temporal prism through path search based on the shortest time cost, obtain the spatio-temporal range that the task activities can reach based on the solved spatio-temporal prism, and search for the optimal path of the task activities in the spatio-temporal environment with the minimum comprehensive threat cost as the goal in the spatio-temporal range sections that the task activities can reach in different time periods based on the principle of overlap or intersection of spatial trajectories;

[0047] S104. Conduct multi-dimensional evaluation on the impact effectiveness of the task spatio-temporal environment based on the reachable spatio-temporal range of task activities, the adaptability of the task to the environment, and the optimal execution plan within the spatio-temporal range of feasible solutions. Among them, the adaptability of the task to the environment is obtained through the shortest time cost incurred by the task in the environment.

[0048] The evaluation of task effectiveness is essentially an evaluation of the expected effect of the task execution plan, which is comprehensively affected by the environment, situation, and capabilities, and the latter three will change continuously over time. Among the factors affecting effectiveness, the environment is the most fundamental. The environment may limit the range of choices of feasible solutions and affect the execution effect of the task. Capability is also one of the main reasons affecting task effectiveness. Different task performers are restricted by their respective capabilities in the environment, which will lead to significant differences in the execution effects of the same plan. However, the influence of capability cannot be discussed without considering the environment, and similarly, the influence of the environment cannot be evaluated in isolation from the capability factor. The situation may not exist, but if it does, its impact on the task is huge. However, the situation does not affect the task in isolation, and its impact effect is also restricted by the environment factor and the capability factor. Time is the only main line running through the changes of the environment, situation, and capabilities, and it is the unified variable for the continuous change of the environment during task execution. In the embodiments of this case, the task effectiveness is comprehensively evaluated from four aspects: the extreme performance of the task, the avoidance spatio-temporal region formed by the environment and the situation, the reachable spatio-temporal domain range of task activities, and the minimum comprehensive cost of the task. By alternately evaluating using the spatio-temporal cost accumulation rate at the local, overall, low-level, and high-level levels, it is possible to evaluate the extreme performance ability, overall adaptability, and optimal result of the task under the influence of the environment from various angles. The model solving complexity is low and the efficiency is high, which is convenient for decision-makers to make more scientific and reasonable task execution strategies and is convenient for application in actual scenarios.

[0049] As the task-oriented multi-dimensional effectiveness evaluation method in the embodiments of the present invention, further, the spatio-temporal comprehensive environment impact model is expressed as a mathematical model F = M(E(T), S(T), P(T)) of environmental variable E(T), situation variable S(T), and capability variable P(T) that change continuously with time T. Among them, by determining a point on the ground surface as the origin, with the eastward direction as the X-axis, the northward direction as the Y-axis, and time T as the vertically upward Z-axis, a three-dimensional rectangular coordinate system is established, and this three-dimensional rectangular coordinate system is used as the spatio-temporal coordinate system.

[0050] There are many environmental factors related to each task, and each environmental factor can be marked as a continuous change function or a constant function environment(t) with time as the independent variable. Assume that in the time period T = [t 1 , t 2A total of n environmental factors that may affect the execution of the task are considered for the task to be executed. As shown in formula (1), the relevant environmental variables can be expressed as the set E(T). Similarly, the relevant situation factor situation(t) and ability factor power(t) can be expressed as formulas (2) and (3) respectively. In particular, here S(T) is the original information of the situation, without including the detailed inference of the situation by combining environmental and ability factors.

[0051] E(T) = {environment 1 (t), environment 2 (t),..., environment n (t) | t ∈ T} (1)

[0052] S(T) = {situation 1 (t), situation 2 (t),..., situation m (t) | t ∈ T} (2)

[0053] P(T) = {power 1 (t), power 2 (t),..., power k (t) | t ∈ T} (3)

[0054] If F is used to express the energy efficiency of the task under the comprehensive influence of the environment, then obviously, F is a function of the continuously changing environment E(T), situation S(T), and ability P(T) over time T. The mathematical model of F can be expressed as formula (4). Among them, for different tasks, due to the preference categories of risks and benefits during the task implementation, the evaluation methods may vary greatly, and their evaluation function M(x, y, z) should also be different.

[0055] F = M(E(T), S(T), P(T)) (4)

[0056] Formula (4) is a summary of the effectiveness evaluation variables and dependency relationships. The conceptual model of the time-space environmental impact on effectiveness evaluation for tasks can be as Figure 2 shown. The environmental impact effectiveness of the task comes from the combined action of the environment, situation, and ability, and is a function of time T. The effectiveness evaluation methods for different tasks vary greatly and need to be analyzed specifically..

[0057] The cost accumulation rate within the spatio-temporal domain is represented by a spatio-temporal vector. The spatio-temporal comprehensive cost growth rate distribution and variation law are analyzed using the cost accumulation rate to evaluate the activity cost within the corresponding range of the task. Among them, the cost accumulation rate is a spatio-temporal vector, including the time accumulation rate (time component) and the space accumulation rate (space component). The evaluation of the environmental impact effectiveness of the task can be achieved by evaluating the effect of the implementation plan. However, different implementation plans have different costs or benefits, and the performance of a single plan will show significant differences due to the influence of the plan planning level and cannot represent the effectiveness of the task. The environmental impact effectiveness should be an evaluation of the degree to which the overall task is affected by the environment without including the quality of the plan itself. The spatio-temporal range of the feasible plan, the performance of the extreme plan and the comprehensive optimal plan can, to a certain extent, represent the overall impact of the environment on the task and can be used as indicators for evaluating the environmental impact effectiveness. For example, for the off-road problem, extreme problems include: when to arrive earliest, when to depart latest, etc. Such extreme problems can evaluate the limit of the degree to which the task can play in the corresponding environment. The boundary problem is a general extension of the extreme problem. For example, under the premise that the task can be completed, which spatio-temporal regions may be reached, the earliest arrival time and the latest departure time of each passing location, etc. These boundary problems give the spatio-temporal range of the implementation plan and can evaluate the adaptability of the task to the environment as a whole. The optimal problem is to select the optimal plan within the above feasible range to minimize the cost or maximize the benefit, which is closely related to the characteristics of the task and expresses the adaptability of the task to the environment under the best choice. Whether it is an extreme problem, a boundary problem or an optimization problem, there is a specific plan supporting each evaluation value behind it. That is to say, the process of evaluating the spatio-temporal environmental impact effectiveness of the task is to continuously plan specific plans that can represent these extreme problems, boundary problems or optimal problems and evaluate the performance of the plans. The process of environmental effectiveness evaluation can be transformed into a process of a large number of task planning.

[0058] The key to performance evaluation is to scientifically quantify and express the impact of the environment on tasks. In traditional minimum-cost paths, concepts such as two-dimensional relative cost grids or friction surfaces in space are generally used to express the distribution of local environmental costs in space, and then the entire plan is evaluated by accumulating these costs along the planned path. In the embodiments of this case, the spatio-temporal environmental impact performance evaluation framework through minimum-cost planning also evaluates the performance of the plan by analyzing cost accumulation. Specifically, it may include the following: establishing a spatio-temporal environmental impact model, by understanding the occurrence and accumulation laws of various local costs or benefits, and expressing this law as a comprehensive spatio-temporal environmental impact model that can correctly evaluate the costs of any plan; based on the comprehensive spatio-temporal environmental impact model, the performance evaluation of the task is achieved through searching for the minimum-cost plan, searching in the feasible spatio-temporal domain to find the minimum-cost plan for various costs, and evaluating the performance of the task through the performance of the plan.

[0059] The main goal of evaluating the local spatio-temporal environment is to correctly express the "cost" of the environment (benefits can be converted into "costs" through certain methods) with respect to the occurrence speed and conditions over time and space. This analysis process is the spatio-temporal cost differentiation process. Cost is an accumulative concept, that is, the total amount of cost will gradually accumulate with human activities. This accumulation process can be regarded as an analysis process of integrating costs in a continuously changing environment.

[0060] There are many types of costs, and each type of cost corresponds to one or more performance evaluation types (for example, the cost of the shortest time consumption corresponds to the shortest time path analysis and the reachable domain analysis; the cost of the comprehensive threat level corresponds to the spatio-temporal optimal path planning method). Traditional algorithms mainly focus on the costs accumulated with spatial activities. In the embodiments of this case, it is considered that in addition, there are also types that accumulate with time, and for cost types related to both time and space, they can be decomposed into time-related parts and space-related parts.

[0061] Time-accumulative cost: A cost whose accumulation process is only related to the length of the activity time. For example, the cost of being discovered in an area prone to exposure is only related to the time spent staying in that area.

[0062] Space-accumulative cost: A cost whose accumulation process is only related to the size of the activity space. For example, the risk cost of crossing a minefield is only related to the distance of passing through the minefield. Among them, the concept of space has different meanings for different types of tasks. For route planning tasks, it mainly refers to the length of the route extended in space in the action plan; for spatio-temporal site selection tasks, it mainly refers to the selected two-dimensional plane area or three-dimensional space volume.

[0063] Due to the uneven environmental distribution, the cost accumulation rates at different times and spaces are different. In the time dimension, since the passage of time is irreversible, there is no difference in the direction of the cost accumulation rate related to time; while in the space dimension, some cost accumulation rates are related to direction, such as the driving time cost caused by the slope effect is different in each direction, and some costs are not related to direction, such as the risk cost brought by low temperature. Denote the cost accumulation rate at a certain time-space position by the time-space vector as shown in the definition formula (5). Among them, respectively represent its time accumulation rate and space accumulation rate, which may be directed variables or undirected variables according to the characteristics of different cost types. In the embodiments of this case, for the convenience of unified description form, the undirected variables can be regarded as isotropic directed variables, and all are expressed by directed variable symbols.

[0064]

[0065] The evaluation purpose of the above "comprehensive cost growth rate" is not the specific "value" of the accumulation rate, but the "change law" of the cost accumulation rate. This is because The change is not only related to the environment, situation, ability and task type of the local time and space, but also related to different parameters according to different cost types, such as the historical track of off-road, the driving direction, the results of lower-level effectiveness analysis, etc. Let i 1 , i 2 ,..., i n be the parameters required to evaluate , then at the time-space coordinates (x, y, t), can be expressed as a function formula (6) of the time-space position and related parameters. Through the time-space coordinates, the environment, situation and ability information can be queried, and other parameters are related to the cost type and need specific analysis.

[0066]

[0067] After mastering the distribution and change law of the time-space comprehensive cost growth rate of the task, the cost of the task in local activities can be evaluated. As shown in formulas (4) and (7), taking the off-road maneuver task as an example, see Figure 3 shown, when moving along the time-space trajectory L m , the cost value at P A should be the product of the time-space comprehensive cost growth rate at P A and the time-space trajectory. Formula (7) is the embodiment of formula (4) in the local time and space, which can describe the occurrence and accumulation law of the local cost during the task execution, and can also be called the differential formula of the local time and space cost.

[0068]

[0069] The solution process of the overall cost of the task implementation plan can be regarded as the integral of the cost accumulation rate according to the plan in time and space. Based on the differential formula (7), for the task along L m the overall spatio-temporal cost W is the integral formula (8). Figure 4 As shown, this formula is applicable to the plans moving along any spatio-temporal trajectory. Among them, on the spatio-temporal trajectory L s of a stationary object, since is always 0, it finally only has the time-accumulative cost value.

[0070]

[0071] Suppose the set of feasible task solutions is Solutions = {L 1 , L 2 ,..., L n ,...}. When evaluating k kinds of effectiveness indicators, the set of k cost growth rates is PriceRates = {w 1 , w 2 ,..., w k}. Then the corresponding plan cost matrix is expressed as represents the kth cumulative cost value of the nth feasible plan, and W wk represents the kth cumulative cost. Further, according to the task type and domain-related planning methods, the task effectiveness is evaluated by evaluating the extreme values, value ranges of each cost, and effectiveness evaluation indicators such as the comprehensive minimum cost.

[0072] According to the evaluation requirements, there can be many definitions of cost. Each cost type combined with multiple evaluation methods can evaluate multiple task effectiveness indicators. When the form of the cost is determined, through formula (8), the cumulative value of this kind of cost in any plan can be evaluated, and thus the overall cost of any plan can be evaluated. Taking off-road path planning as an example, theoretically the set of feasible plans contains countless plans Solutions = {L 1 , L 2 ,..., L n ,...}. If k kinds of effectiveness indicators are evaluated, it will correspond to a set of k cost growth rates PriceRates = {w 1 , w 2 ,..., w k} (where w 1 , w 2 ,..., w kSome of them may be the same because even for the same cost type, different performance metrics can be obtained by combining different evaluation methods. For each solution, k different costs can be calculated, forming the cost matrix shown in formula (9).

[0073] The essence of performance evaluation is to analyze the extreme, general, or best performance of a task through formula (9). Among them, the extreme performance can be evaluated by finding the minimum cost that each cost type can achieve in these cost matrices. For example, if the y = min(x) function is used to represent taking the minimum value, the extreme performance can be calculated according to formula (10); the general performance can be evaluated by finding the general metrics (such as value range, average value, etc.) of each cost in the matrix. For example, if the [y 1 ,y 2 = r(x) function is used to represent the value range function, the cost value range can be calculated according to formula (11); the best performance requires integrating various costs to form a unified evaluation system. If the formula y = m(x 1 ,x 2 ,...,x n ) is used to represent the function of integrating multiple types of costs, and m represents the minimum integrated cost value, the best performance can be calculated according to formula (12).

[0074]

[0075]

[0076]

[0077]

[0078] Formulas (9) to (12) illustrate the theoretical basis for the spatio-temporal environmental impact effectiveness evaluation through minimum-cost planning in the embodiments of this case. Formulas (10), (11), and (12) are all specific implementation methods of Formula (4). These three formulas can include information on the environment E(T), situation S(T), ability P(T), and task characteristics through Formula (6), and finally achieve the effectiveness evaluation. In actual calculations, according to the task type, domain-related planning methods should be used to search for the minimum-cost solutions for each cost. Taking cross-country mobility as an example, compared with the traditional cross-country analysis based on GIS methods, the embodiment solution of this case can pay attention to the unified expression of the continuous change of the environment, and the evaluation results need to reflect the impact of the continuous change of the environment; and the process of cost accumulation is not only carried out in space, but also includes cost accumulation in time, so that the process of scheme planning is not only carried out in space, but is a spatio-temporal integrated planning. The evaluation results reflect the role of the situation and ability in environmental evaluation, and can also reflect the effectiveness evaluation differences brought by different tasks. Secondly, the effectiveness evaluation process in the solution of this case is not completed in one step, but alternates between the local and the whole, from low-level understanding to high-level understanding, and can evaluate the extreme performance ability, overall adaptability, and optimal results of the task under the influence of the environment from various angles, and use the evaluation model for solution analysis, which is convenient for assisting the execution of the task execution strategy and improving efficiency.

[0079] Time geography is a general method for understanding the interdependence between humans, nature, and technology. It is generally studied in a conceptual coordinate system that includes an n-dimensional space and a 1-dimensional time. Time and space are orthogonal, and the activities of individuals are unidirectional and irreversible in the time dimension. The basic elements of time geography are spatio-temporal paths and spatio-temporal prisms. The study of spatio-temporal prism problems is about the boundary problems of human activities, that is, within a certain time range, the maximum spatio-temporal range that can occur when the human movement speed is affected by the environment. Studying spatio-temporal prisms in cross-country activities can evaluate the cross-country accessibility as a whole, analyze a series of problems such as the possible spatio-temporal range size, key points, and spatio-temporal tolerance ability of cross-country activities, so as to achieve the purpose of evaluating the task effectiveness as a whole, which is scientific, reasonable, effective, and closer to the actual application environment.

[0080] To uniformly express the dynamic changes of the spatial environment over time, the spatio-temporal comprehensive environmental impact model is established on Figure 4 the spatio-temporal coordinate system shown to record the environmental impact laws at any spatio-temporal point, Figure 4In (a), it represents the space-time coordinate system of a two-dimensional space, and in (b), it represents the space coordinate system of a three-dimensional space. Both coordinate systems are three-dimensional coordinate systems, but their coordinate meanings are completely different. The definition of the space-time coordinate system in the two-dimensional space is consistent with the similar coordinate system in space-time geography. To uniformly express the dynamic change of the spatial environment over time when studying the problem of ground cross-country, in the embodiments of this case, further, the environmental comprehensive impact model is expressed as: at any position (x, y, t) in the space-time coordinate system, record the space-time cost growth rate of the environmental comprehensive cost To record the rate of change of the cost accumulation of the environment, situation, and ability on task execution at the space point (x, y) at time t. The meaning of is related to the definition of "cost". For the shortest time planning, is the minimum time required to move a unit distance, that is, the shortest time consumption cost rate And for the analysis of the comprehensive cost of the path, is the space-time comprehensive threat cost rate

[0081] Such as Figure 5 As shown, the space-time trajectory of a stationary object in the space-time coordinate system can be represented as a straight line parallel to the time axis (such as L s ); the space-time path of a continuously moving object presents as a continuous curve (such as L m ); the space-time region U between the space-time points P' Start and P' end is called the space-time prism determined by the starting space-time point and the arriving space-time point P' end , expressing the possible space-time range where the moving object may appear during this period of time. P' end P' Start are respectively called the upper and lower anchor points of the space-time prism; PPA is the abbreviation of Potential Path Area, representing the possible space range where an individual may appear within the time budget. If P 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 calculation formula for its speed is:

[0082]

[0083] This value is exactly the slope of L m and the space plane. The smaller the slope value, the faster the moving speed in the space. Such as L sAs shown, if an object is stationary, it should be presented as a straight line parallel to the time axis. The expression of reachability by the spatio-temporal prism is mainly based on the following principle: If the spatio-temporal region that can be reached from the starting spatio-temporal point by any spatio-temporal trajectory not exceeding the maximum speed is denoted as N; and the spatio-temporal region that may be passed through by any spatio-temporal trajectory not exceeding the maximum speed and can reach the end point before the specified time is denoted as R, then it is as follows:

[0084] U = {(x, y, t)|(x, y, t) ∈ N ∩ (x, y, t) ∈ R} (14)

[0085] Anchor point P' Start With P' end The determined spatio-temporal prism U is the spatio-temporal that satisfies both the spatio-temporal that can be reached starting from the specified starting point at the specified moment and the spatio-temporal that can reach the end point before another specified moment. It reflects the possibility of an individual traveling and moving within the starting point, the end point, and a certain time budget. The spatio-temporal region expressed by the spatio-temporal prism is the reachable spatio-temporal region studied in this case. The spatio-temporal regions outside this range are restricted by the time budget, environmental impacts, and rules, or it is impossible to reach from the starting point or impossible to reach the end point on time.

[0086] Task along L m The overall spatio-temporal cost Applies to the task execution plan for moving along any spatio-temporal trajectory. By integrating through the overall spatio-temporal cost formula, the relevant environmental cost can be obtained. However, the changes of the environment in time and space are both irregular, and there are difficulties in the actual operation process. Therefore, in the embodiments of this case, by rasterizing the spatio-temporal environment, as Figure 6 shown, the spatio-temporal is divided into spatio-temporal unit cubes (spatio-temporal voxels, simply called voxels) along the X-Y-T three axes respectively at the same spatial and time intervals. It is approximately considered that the spatio-temporal growth rate of the comprehensive environmental cost in the same spatio-temporal unit is consistent. The environmental membership problem in the rasterization process can be set as: spatially, the rasterized environment belongs to the part of the environment that has the greatest impact on the task; temporally, the state of the environment belongs to the environmental state of the geometric center of the voxel. What is expressed is the spatio-temporal rate of cost accumulation. In addition to being related to the spatio-temporal environment, it may also be related to other factors, such as the driving direction during task execution, historical trajectories, etc. Therefore, even after the environmental rasterization, it is not necessarily a numerical value, but represents a cost accumulation law.

[0087] Theoretically, the process of understanding the environmental impact is the calculation process for each voxel . But in fact, for all voxels within the spatio-temporal region under investigation It is unrealistic to perform all calculations, and the cost of such a huge amount of environmental assessment is enormous, which will lead to explosion problems during the calculation process. In the embodiments of this case, spatio-temporal trajectories can be utilized, that is, generate voxels as needed during the assessment process, determine the quantity used, and perform environmental assessment accordingly.

[0088] The significance of discretization lies in simplifying complex change laws. When the spatio-temporal interval size of rasterization is appropriate, it can be approximately considered that the environment is similar within the spatio-temporal range it represents, and all its cost laws are also the same. For example, Figure 6 as shown, if (x 0 , y 0 , t 0 ) is the geometric center point of the green voxel shown in the figure, and w′(·) is the approximate expression of the cost w(·) within the voxel, then w can be approximately expressed by the following formula:

[0089]

[0090] To evaluate the cost of the task execution plan based on the spatio-temporal comprehensive environmental impact model, it is first necessary to obtain which spatio-temporal voxels the known plan will pass through, as well as the time length of stay and the spatial distance of movement in each spatio-temporal voxel. This process is the spatio-temporal tracking process of the movement trajectory. The movement in the spatio-temporal environmental comprehensive impact model can be regarded as a series of continuous straight-line motions, and any complex movement trajectory can be approximated in this way. For example, Figure 7 as shown is a straight-line motion from spatio-temporal point S to spatio-temporal point E in a period of time and space. Some voxels A 0 , B 0 , B 1 are drawn in the figure, where A 0 , B 0 are spatially adjacent and have the same time, and B 0 , B 1 have the same space and are time-adjacent. Spatio-temporal point S is a spatio-temporal point of spatial point A on its spatio-temporal trajectory L A ; spatio-temporal point E is a spatio-temporal point of spatial point B on its spatio-temporal trajectory L B . Assume that the task execution plan starts from S and follows the spatio-temporal trajectory shown in the figure, passing through spatio-temporal points M 0 , M 1 to reach E. The spatio-temporal coordinates of the endpoints of each straight-line segment in the specific plan are known (such as the spatio-temporal coordinates of S and E in the figure). If the driving speed and direction of this section of the task itinerary are known, and then the spatio-temporal coordinates of each spatio-temporal point entering or exiting the voxel are obtained (such as the spatio-temporal coordinates of M 0 , M 1 in the figure), the spatio-temporal trajectory passing through each voxel (such as A 0 , B 0 , B 1 in the figure) can be calculated.The travel length and residence time within the three voxels of the path.

[0091] When performing specific model solving calculations, the driving direction i can be calculated using the spatial coordinates of A and B with the formula i = arctan((y B - y A ) / (x B - x B ). This value remains unchanged when moving through each voxel between these two points. The velocities in the x and y axis directions Can be obtained according to the scheme. Let upper(x) and floor(x) be the floor and ceiling functions respectively, and forward(x) and backward(x) can be defined as the backward and forward rounding functions respectively. Denote the grid intervals of the x, y, and t axes as d x , d y , d t . Suppose the coordinate marker at the start of movement within the j-th spatio-temporal voxel is (x j , y j , t j ). The maximum remaining movement time And the geometric center coordinates of the voxel Can be expressed as:

[0092]

[0093] During This period of time, without considering the spatial boundaries of the voxels, the maximum travel distances in the x and y axis directions are respectively Without considering the time boundaries of the spatio-temporal voxels, the maximum travel distances in the x and y axis directions are respectively If other axes are not considered, denote the shortest times to exit from the T axis, X axis, and Y axis as respectively The corresponding formulas are Then the earliest exit time Δt j Is expressed as The axis is the axis corresponding to the shortest time.

[0094] After completing the above calculations, if and It indicates that the end point has not been reached when passing through the j-th voxel, and it will continue to move into the j + 1-th voxel. The entry coordinates are Repeat the above steps in the next voxel until the n-th voxel can reach the end point. In this case, Δt n Is corrected to And the calculation ends.

[0095] The tracking of the straight-line segment movement in the voxel can be completed through the above method, and the spatio-temporal coordinates when the spatio-temporal trajectory enters and exits each voxel can be obtained. The trajectory is a uniform straight-line movement within the voxel, that is, the straight-line segment connecting the spatio-temporal points where it enters and exits the voxel. Through the above content, it is possible to track which voxels the specific cross-country plan will pass through, as well as the length of time staying in each voxel and the spatial distance traveled.

[0096] Suppose the spatio-temporal trajectory of the task execution plan is obtained through the movement trajectory tracking and will sequentially pass through n spatio-temporal voxels. Denote the cost growth rate in the i-th spatio-temporal voxel as The trajectory stays in this spatio-temporal voxel for a time The length of the traveling space is Then the cost W generated by the task execution plan in the i-th voxel of the model is i For The overall cost W of the entire plan is

[0097] As the task-oriented multi-dimensional effectiveness evaluation method in the embodiment of the present invention, further, in the evaluation of the extreme performance of the task through the shortest time path planning, based on the time consumed by the straight-line movement between any two adjacent spatio-temporal points, the spatio-temporal trajectory between the starting point and the target point of the task execution plan is obtained through path search to obtain the spatio-temporal trajectory with the least time consumption from the starting point to the target point.

[0098] The straight-line movement in the spatio-temporal environment comprehensive influence model can be represented by a function. Assume that the spatio-temporal coordinates of the starting point are (x 1 , y 1 , t), and the spatio-temporal coordinates of the arrival at the target space point are (x 2 , y 2 , t + Δt), where Δt is the shortest time consumption for moving between the two points. In most traditional analysis methods, Δt is only related to the spatial positions of the two points and has nothing to do with t, that is, Δt is the same no matter when the movement occurs. This process corresponds to the functional formula Δt = f 0 (x 1 , y 1 , x 2 , y 2 ). However, this obviously does not conform to the actual situation. Since the environment is constantly changing and the speeds at different times are different, Δt is related to the geographical positions (x 1 , y 1 ), (x 2 , y 2 ) of the starting and ending points and the departure time t. A more reasonable expression should be the functional formula Δt = f′ N (x 1 , y 1 , x 2 , y2 , t); Correspondingly, if the time t + Δt is known but t is unknown, the functional equation Δt = f' R (x 1 , y 1 , x 2 , y 2 , t + Δt) should be used.

[0099] Compared with the traditional static method, the biggest difference in the solution of this case is that the time-independent Δt = f in the traditional method is replaced by a time-related one. 0 (x 1 , y 1 , x 2 , y 2 ). Through this change, the problem of the continuous change of the environment over time can be effectively expressed. For example, when encountering bad weather and off-road operations are forced to stop, the time cost will immediately accumulate to the marginal time when the bad weather disappears. As the bad weather disappears, spatial movement can be resumed. On the contrary, if Δt = f 0 (x 1 , y 1 , x 2 , y 2 ) is used to describe this process, in a mathematical sense, there is only one climate state, and off-road operations will stop forever, or the state of bad weather will be ignored. Using the formula Δt = f' N (x 1 , y 1 , x 2 , y 2 , t) or Δt = f' R (x 1 , y 1 , x 2 , y 2 , t + Δt) can always reflect the impact of environmental changes on the driving speed, which is a mathematical model closer to the actual situation.

[0100] Traditional path search under off-road conditions is generally completed by using a path search algorithm on a "cost" surface. Different from traditional minimum-cost path planning, the minimum cost required for each movement in a spatio-temporal environment is related to the moment when the movement occurs. The process is relatively complex, but it is still possible to perform path search on a spatio-temporal comprehensive environmental impact model with the goal of reducing the cost accumulation value through a path search algorithm.

[0101] If the spatio-temporal comprehensive environment impact model is sliced at a certain moment, its profile is equivalent to the "cost" surface of the traditional algorithm. Different from the traditional method, in each step of the search, in addition to calculating the cumulative cost, the required time also needs to be accumulated simultaneously. The cost magnitude can be used as the optimization objective of the planning, while the time can be used to calculate the spatio-temporal environment of the starting point of the next search. Through iterative calculation, the minimum cost search problem of the continuously changing environment can be completed by steps similar to the traditional algorithm. In the calculation process, the "cost" adopts the shortest time consumption cost. That's all.

[0102] See Figure 8 As shown, the left and right subgraphs are the dangerous spatio-temporal region and the escape spatio-temporal prism respectively. As shown in the figure, the structure of the escape spatio-temporal prism is mainly divided into three parts, which can be defined as the affected area, the occurrence area and the vacuum area. These three parts are combined into a closed spatio-temporal region, expressing the spatio-temporal region range that must be avoided or cannot be reached due to dangerous events to complete the task.

[0103] Affected area: Although not currently affected in this spatio-temporal region, due to the problem of mobility, it will not be able to escape the upcoming danger in the future. This region divides the spatio-temporal by the lower pyramid into an external region and an internal region. The time represented by the spatio-temporal points on the pyramid surface refers to the latest time when one should leave at the corresponding spatial point. If later than this point, it will surely not be able to escape the danger that will occur in the "future".

[0104] Occurrence area: In this spatio-temporal region, dangerous events are occurring. If this region is a regular graph, it completely coincides with the dangerous spatio-temporal region.

[0105] Vacuum area: If the task bearer is not affected by dangerous events, it is impossible to reach this region. This region divides the spatio-temporal by the upper inverted pyramid into an external region and an internal region. The danger in its internal region has disappeared, but due to the limitation of the moving speed, it is impossible to appear in this region without passing through the occurrence area.

[0106] In the solution of this case, the reachable region range in the task execution plan under predictable spatio-temporal conditions is obtained through the spatio-temporal prism, and the spatio-temporal prism is used to represent the possible spatio-temporal boundaries in the task execution plan, which is convenient for evaluating issues such as the feasibility, reliability, and spatio-temporal redundancy of the task when executed in the corresponding spatio-temporal. Considering the continuous change of the environment over time, it will not lead to a large number of invalid calculations, and the operation load and the complexity of the operation output are controllable. The spatio-temporal domain evaluation efficiency of the task activities is high, and it is more scientific and reasonable.

[0107] Furthermore, in an environment that changes continuously over time, it is assumed that each possible spatial point that may be passed during cross-country travel consists of an earliest arrival time and a latest stay time, which form the earliest arrival spacetime point and the latest stay spacetime point of this waypoint in the spacetime coordinate system; the earliest arrival spacetime points and the latest stay spacetime points of all possible waypoints respectively form the lower and upper pyramidal spacetime surfaces, which are the spacetime surfaces used to describe the "future" and "past"; taking the starting spacetime point as the lower anchor point and the arrival spacetime point as the upper anchor point, with the lower and upper pyramidal spacetime surfaces as the boundary surfaces, the spacetime region between the two boundary surfaces is the reachable spacetime prism region that is solved and used to represent the spacetime range where cross-country travelers who can complete the task on time may appear.

[0108] For all the earliest arrival spacetime points starting from the same starting spacetime point, they present a continuous surface, and for all the latest departure spacetime points for the same arrival spacetime point, they present a continuous surface. In this patent application, for the prediction and evaluation of the reachable region, taking the starting spacetime point as the lower anchor point and the arrival spacetime point as the upper anchor point, starting from the starting spacetime point, calculate the earliest arrival time of each spacetime point in the spacetime trajectory of the task execution plan, and use the spacetime points corresponding to this earliest arrival time to form the lower pyramidal boundary surface of the reachable domain spacetime prism; starting from the arrival spacetime point, calculate backward the latest departure time for each spatial point in the spacetime trajectory of the task execution plan to reach the end point on time, and use the spacetime points corresponding to this latest departure time to form the upper pyramidal boundary surface of the reachable domain spacetime prism; for the obtained upper and lower pyramidal boundary surfaces, obtain the target spacetime range represented by the reachable domain spacetime prism through intersection operation.

[0109] Using an accessible spatio-temporal prism to represent the accessible spatio-temporal domain ignores the impact of extreme dangers on the accessible domain. In particular, unexpected events within the accessible domain can greatly affect the evaluation scope of the accessible domain, resulting in the evaluation results obtained solely using the accessible spatio-temporal prism deviating from the actual situation. Task activities may be disrupted by extremely dangerous events, and task plans should avoid being affected by dangerous events. In the present application, the danger escape spatio-temporal prism is divided according to time and the spatio-temporal scope of influence into an occurrence area for describing the spatio-temporal scope where a dangerous event occurs, a affected area for describing the spatio-temporal scope where it is impossible to escape from the dangerous event due to restricted mobility, and a vacuum area for describing the spatio-temporal scope that could not be reached without experiencing the dangerous event due to restricted mobility. Among them, the lower spatio-temporal boundary of the affected area, the spatio-temporal boundaries of the occurrence area except for the upper and lower boundaries, and the upper spatio-temporal boundary of the vacuum area together divide a closed spatio-temporal region. Using this closed spatio-temporal region as the danger escape spatio-temporal prism for representing the spatio-temporal region range that must be avoided to complete the task or cannot be reached due to the dangerous event, the lower anchor point of the danger escape spatio-temporal prism is the spatio-temporal edge at the start time of the dangerous event. The significance of the lower spatio-temporal boundary of the affected area is equivalent to the upper boundary of the accessible spatio-temporal prism and is a spatio-temporal surface describing the "past". The upper anchor point is the spatio-temporal edge at the end time of the dangerous event. The significance of the upper spatio-temporal boundary of the vacuum area is equivalent to the lower boundary of the accessible spatio-temporal prism and is a spatio-temporal surface describing the "future".

[0110] Furthermore, as shown in Figure 11 During the solution of the danger escape spatio-temporal prism, both the upper and lower boundary surfaces are within the spatial range where the dangerous event occurs. Among them, when solving the lower spatio-temporal boundary surface ("past" spatio-temporal boundary), starting from the moment when the dangerous event occurs, using the maximum speed allowed by the environment and the optimal path and taking the spatial edge of the dangerous event as the destination, calculate backward in time the latest departure time at which all spatial points within the dangerous event range can escape danger. The corresponding spatio-temporal points form a continuous pyramidal spatio-temporal surface, and this pyramidal spatio-temporal surface is the lower spatio-temporal edge of the danger escape spatio-temporal prism for representing that when the spatio-temporal position of the off-road vehicle is below this spatio-temporal surface, it can escape the upcoming dangerous event; when solving the upper spatio-temporal boundary surface ("future" spatio-temporal boundary), starting from the moment when the dangerous event ends, using the maximum speed allowed by the environment and the optimal path and driving inward from the spatial edge of the dangerous event, calculate forward in time the earliest time at which all spatial points within the dangerous event range can reach after the dangerous event ends. The corresponding spatio-temporal points form a continuous pyramidal spatio-temporal surface, and this continuous pyramidal spatio-temporal surface is the upper spatio-temporal edge of the danger escape spatio-temporal prism for representing that if the current off-road vehicle has not been affected by the dangerous event, it is necessarily impossible to be below this spatio-temporal surface.

[0111] Further, when solving for future or past spacetime surfaces in spacetime voxels, an open list and a closed list are established. Point - like anchors, hashed multi - point - like anchors, line - like anchors, and spacetime points located at the center of the spatial grid near the edges of planar anchor lines are directly added to the closed list. Spacetime points located inside the planar anchor and at the center of the spatial grid with a time near that of the planar anchor are also added to the closed list. The shortest - time cost of the central points of the spatial grids adjacent to the periphery of the existing closed list is evaluated, and the resulting spacetime points are added to the open list. The spacetime point with the earliest time is found from the open list and added to the closed list. Using a path - search algorithm, search in eight directions adjacent to the space based on this point, update the earliest arrival time of the adjacent spacetime points in the open list, or evaluate the unevaluated central points of the spatial grid and add them to the open list. Then, find the spacetime point with the shortest time in the open list again and add it to the closed list. Repeat this cycle until the earliest possible arrival time of each central point of the spatial grid is calculated. When calculating the latest staying spacetime point at the central position of the spatial grid, start from the spacetime point with the latest staying time in the open list and search backward for the latest departure time of each central point of the spatial grid.

[0112] Further, in the solution of the spacetime pyramid, the spacetime trajectory of the corresponding segment of motion is taken as a continuous spacetime trajectory composed of multiple spacetime straight - line segments that connect the current spacetime point and the central point of the target spatial grid and are connected end - to - end. Each spacetime straight - line segment is in a different spacetime voxel. Except for the current spacetime point and the central point of the target spatial grid, the starting and ending spacetime coordinates of each segment are the spacetime coordinates where the spacetime trajectory enters and exits the spacetime voxel. The spacetime trajectory tracking sequentially uses the spacetime coordinates where it exited the spacetime voxel last time as the starting spacetime coordinates for the next segment of the spacetime straight - line motion, and combines the spatial movement direction, the spacetime range of the spacetime voxel, and the maximum speed when traveling in a specific direction within the spacetime voxel to calculate the spacetime coordinates when exiting the spacetime voxel until it reaches the central point of the target spatial grid in space. The reachable spacetime range of the target object's task activities can be predicted by the spacetime range that is within the task - reachable spacetime prism but outside the dangerous - escape spacetime prism, which is a feasible spacetime region where the task can be safely completed and extreme - danger events can be avoided.

[0113] An off - road task can be described as follows: Given the starting spatial coordinates, the target spatial coordinates, the earliest departure time, and the latest arrival time, it is required to start from the starting point and reach the target point within the specified time range through continuous spatial activities. Its travel trajectory can be expressed as a continuous spatial trajectory, and this continuous spatial trajectory can be approximated by a large number of shorter spatial straight - line segments that are connected end - to - end and connect the starting point and the ending point. Therefore, the key to solving the shortest - time cost of the trajectory is to calculate the shortest time difference for each small - segment spatial straight - line motion when traveling at the maximum speed allowed by the environment at each moment.

[0114] When analyzing the spatio-temporal trajectory of each small segment of linear motion in space, given the spatio-temporal coordinates of the starting point of the motion and the spatial coordinates of the end point of the linear segment in space, this segment of linear motion in space must occur in one or more spatio-temporal voxels. We approximately assume that the environmental impact is consistent within the same voxel. Therefore, the maximum driving speed within the same voxel is the same. If we do not consider the speed mutation between different voxels, we can approximately regard the spatio-temporal trajectory of this segment of linear motion in space as a multi-segment shorter spatio-temporal straight line segment that is connected end to end and connects the starting spatio-temporal of this segment of linear motion in space and the end space of this segment of linear motion in space. Each spatio-temporal straight line segment is in a different voxel, and its starting spatio-temporal and ending spatio-temporal are exactly the spatio-temporal coordinates where the spatio-temporal trajectory penetrates and exits the voxel. During specific calculations, in sequence, we use the spatio-temporal coordinates of the last time exiting the voxel as the starting spatio-temporal of this segment of spatio-temporal linear motion, and calculate the spatio-temporal coordinates when exiting the spatio-temporal voxel by combining factors such as the direction of spatial movement, the spatio-temporal range of the voxel, and the maximum driving speed that can be obtained in the corresponding voxel, until reaching the spatial coordinates of this segment of linear motion in space. Then, sum up the time consumption of each segment of spatio-temporal straight line segment along the way, and the obtained value is the shortest time consumption cost of this segment of linear motion in space. Finally, sum up the shortest time consumption costs of each segment of linear motion in space, and the obtained value is the shortest time cost of the cross-country plan.

[0115] In the solution of the spatio-temporal prism of the reachable region from environmental data to the final representation, first, analyze the objective environmental impact as needed, and give the shortest time consumption rate of the corresponding spatial point by synthesizing the states of various environments during evaluation. Form an integrated environmental impact model based on spatio-temporal voxels, and then calculate the shortest time for linear movement between two points starting at a certain moment as needed; through the search with the minimum time cost, find the upper and lower pyramid surfaces of the spatio-temporal prism; and take the union of the upper and lower pyramids. The spatio-temporal range expressed by this union is the spatio-temporal pyramid.

[0116] In the calculation of the upper and lower pyramid surfaces, the Dijkstra algorithm can be used to search in eight directions adjacent to the space to implement the search for the shortest arrival time of each spatial point. In addition to the simplest eight-way adjacent search, other types of search strategies can still be supported in the embodiments, such as the search strategy of the travel route with a certain width studied by Shirabe, or the cross-grid search strategy studied by Saha et al., and the specific application can be replaced according to requirements. Figure 9 For the process of calculating the lower pyramid boundary surface, Figure 10 For the process of calculating the upper pyramid surface, the reverse calculation idea can be used for the calculation of the upper pyramid surface, that is, taking the end point as the first point to start the calculation, and calculating the latest departure time from the adjacent point to the previous point. The calculation process and method are similar to those of the lower pyramid surface.

[0117] The key to calculating the upper and lower boundary surfaces of the spatio-temporal prism lies in using a The spatio-temporal comprehensive environmental impact model is used to search and calculate the path. Taking the Dijkstra algorithm as an example, the calculation of linear anchor points is analyzed. When calculating the traditional spatio-temporal prism, the feasible paths closer to the starting point in space-time are divided into an open list and a determined list, which refer to the paths with preliminarily determined time costs and the paths with completely determined time costs respectively. All paths are gradually searched by continuously finding the minimum time cost. In the embodiments of this case, the lower interface calculation of the anchor point can be completed according to the following steps:

[0118] All spatio-temporal points on the linear "anchor point" are directly added to the determined list; all points near the linear "anchor point" are evaluated for the shortest time cost once, and all the evaluated adjacent points are added to the initial open list. Analyze the objective impact of the environment as needed, and give the shortest time consumption rate of the corresponding spatial points by synthesizing the states of various environments during evaluation, forming an environmental comprehensive impact model based on spatio-temporal voxels; calculate the shortest time for a straight-line movement to start between two points at a certain moment; through the search of the minimum time cost, find the upper and lower pyramid surfaces of the spatio-temporal prism, and obtain the spatio-temporal range expressed by taking the union of the upper and lower pyramids. This spatio-temporal range is the spatio-temporal pyramid.

[0119] The calculation of scattered multi-point "anchor points" is the same as that of the linear ones. The calculation method of the spatio-temporal prism of the planar anchor point is mostly the same as that of the linear "anchor point", but the points inside the plane should be avoided from being evaluated. Therefore, before calculating the spatio-temporal prism of the planar "anchor point", the points inside the plane and the points on the plane boundary line should be added to the initial determined list together.

[0120] The environment is the main factor leading to the change of off-road accessibility. By means of path analysis, it is only possible to analyze the performance of a single line, and it is difficult to scientifically and intuitively analyze the overall off-road performance of the investigated area. However, using the method of spatio-temporal prism to study the accessibility boundary of the off-road environment can intuitively express and analyze the reachable space domain, the feasible time range at each space, the passability characteristics of key nodes (such as bridges, river crossings), etc., and realize the overall evaluation of the off-road performance of the environment. In the embodiments of the solution of this case, the spatio-temporal range of off-road activities is evaluated from two perspectives:

[0121] One is for the unreachable spatio-temporal area. To avoid being affected or involved by the dangerous events that will occur in the "future" and cannot enter the boundary of the smallest spatio-temporal range, use the dangerous escape spatio-temporal prism to cooperate with the spatio-temporal representation of the risk area to predict the future situation in detail.

[0122] The other is to evaluate the reachable spatio-temporal domain of the mission operation under the joint influence of the environment, situation, ability, and mission characteristics through the improved spatio-temporal prism. With the help of the analysis results of the dangerous escape spatio-temporal prism, the overall adaptability of the mission to the environment can be reflected through the evaluation results.

[0123] Both the ordinary space-time prism and the dangerous escape space-time prism are based on voxel-based analysis and simulation, which can reflect the influence of a continuously changing environment. The "anchor points" are not limited to space-time points and can be space-time lines, space-time planes, or multiple scattered space-time points, making the evaluation scheme more scientific, reasonable, and intuitive to meet the requirements of actual activity applications.

[0124] The reachable area range in the task execution plan under predictable space-time conditions is obtained through the space-time prism. The space-time prism is used to represent the possible space-time boundaries in the task execution plan, facilitating the evaluation of issues such as the feasibility, reliability, and space-time redundancy of the task when executed in the corresponding space-time. Considering the continuous change of the environment over time, the cost caused by the environment and other factors is quantified in a scientific way to solve the situation where a large amount of space-time in traditional algorithms is impossible to reach in a changing environment. Moreover, due to the limited number of voxels inside the space-time prism and the limited number of selected paths, the computational amount can be effectively controlled, and the evaluation efficiency is higher.

[0125] The evaluation of the space-time range of task activities by the space-time prism is based on the precise quantification of the minimum time cost in time and space, and the reachable space-time domain and the extreme performance of the task are predicted through path search. The optimal path based on the minimum threat level in the embodiments of this case is a relatively optimal path, which is obtained by non-dimensionalizing and fuzzily optimizing the cost in a rasterized space-time environment to obtain an execution plan with a relatively minimum cost optimal path.

[0126] The optimization solution process of the optimal path based on the minimum threat level can be described as follows: In the rasterized spatio-temporal environment, set the spatio-temporal coordinates for the start and end of the task execution plan. Construct line segments based on the spatio-temporal points entering and exiting the spatio-temporal voxels to obtain the spatio-temporal trajectory of the task execution plan. Use the spatio-temporal coordinates of the spatio-temporal points where the task execution enters or exits the spatio-temporal voxels in the model to obtain the travel length and residence time within each spatio-temporal voxel passed by the spatio-temporal trajectory. Obtain the comprehensive environmental cost of the task execution plan through the cost growth rate of each spatio-temporal voxel passed by the spatio-temporal trajectory, the residence time, and the travel length within the corresponding spatio-temporal voxel. And perform hierarchical path search using time periods. Start calculating from the first calculation time period within the reachable domain. Use each node on the lower boundary of the reachable domain within this time period or the center point of the target voxel in the previous time period as the starting point, and calculate the alternative spatio-temporal paths from the node to the center point of the target voxel in this time period that meet the current speed condition. Evaluate multiple alternative paths connected to the center point of each target voxel in the current calculation time period, and only retain the path with the minimum cumulative comprehensive threat cost among them, discard other alternative paths, and combine the minimum comprehensive threat cost of the corresponding starting point to evaluate the minimum cumulative comprehensive threat cost to the center point of the current target voxel. If, during the calculation of the current time period, the reachable spatio-temporal domain contains the upper boundary of the spatio-temporal prism, use each node on the upper boundary as the center point of the target voxel to participate in the calculation, and evaluate the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the final target point. Enter the next calculation time period until all calculations within the reachable domain are completed. Find the path with the minimum cumulative comprehensive threat cost among all paths reaching the off-road end point as the optimal path, and obtain the optimal execution plan within the spatio-temporal range of the feasible plan corresponding to the minimum threat level cost based on this optimal path.

[0127] See Figure 12 As shown, when performing the next path search for the center of a certain voxel, the alternative spatio-temporal path segments can be constructed according to the following steps:

[0128] (1) Taking the starting spatio-temporal point as the starting point and an integer multiple of the voxel unit time interval as a calculation time period, calculate the spatio-temporal boundary range that the starting spatio-temporal point can reach within this time period;

[0129] (2) Within the spatial range reachable by the starting spatio-temporal point and within the overall off-road reachable spatio-temporal domain, find all voxels containing the target time, and use the geometric center points of these voxels as the center points of the target voxels;

[0130] (3) Connect the starting spatio-temporal point to all target spatio-temporal points with a spatio-temporal straight line (uniform linear motion), and the corresponding connection lines are the alternative spatio-temporal trajectories for the next search at the starting spatio-temporal point;

[0131] Calculate information such as the driving speed and direction through the slope of the alternative space-time straight line. Then, the space-time trajectory of the relevant voxels passed along the space-time straight line can be calculated. It can be approximately considered that the environmental impact law within a space-time voxel is unchanged. Denote the cost growth rate of the alternative path segment H in the i-th space-time voxel as The trajectory stayed in this space-time voxel for a time The traveling space length is Therefore, the cost generated in the i-th voxel will be If H passes through q voxels, and the cumulative comprehensive threat cost of the previous segment H-1 of H is D H-1 , then the cumulative comprehensive threat degree cost D of H H Calculate according to D H represents the minimum cumulative comprehensive threat degree cost required to reach the center point of the current calculated target voxel from the starting point of the mission activity through H.

[0132] When performing path search, it can be calculated layer by layer according to the calculation time period. Starting from the first calculation time period inside the reachable domain, calculate according to the following steps: taking each node on the lower boundary of the reachable domain in this time period as the starting point, calculate the alternative space-time paths to the center point of the target voxel in this time period. If this 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 this time period should also be calculated; evaluate that there will be multiple alternative paths connected to the center point of each target voxel in the current calculation time period, and only retain the one with the minimum cumulative comprehensive threat cost, discard other alternative paths, and combine the minimum comprehensive threat cost of the corresponding starting point to evaluate the minimum cumulative threat cost to the center point of the current target voxel; if within a calculation time period, the reachable space-time domain contains a part of the upper boundary of the space-time prism, then take each node on the upper boundary as the center point of the target voxel to participate in the above calculation, and evaluate the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the off-road final target point, and record it; enter the next calculation time period until all calculations within the reachable domain are completed; find the one with the minimum cumulative comprehensive threat cost among all paths reaching the off-road end point, which is the optimal cost path. Further, in the embodiments of this case, parallel calculation and / or heuristic search can be adopted in the path search, or the center points of the target space-time voxels can be sampled using a spatial position mask or the space-time voxel resolution can be reduced to increase the space-time interval of the space-time voxels to optimize the path search calculation amount.

[0133] The principle of searching for the optimal path in the space-time environment can be realized based on the following three principles:

[0134] The principle for selecting the traveling speed is that traveling at the maximum safe speed can achieve the mission goal fastest. However, for many missions, the fastest is not necessarily the best. For example, when a vehicle exceeds its economic speed, its average fuel consumption will increase significantly, and deliberately reducing the speed can sometimes result in better economic benefits; another example is that when affected by adverse weather conditions, waiting for the weather to improve before continuing to drive can effectively reduce the overall risk. The goal of minimum-cost path planning is to reduce costs. That is to say, instead of always using the maximum safe speed, it is better to drive at a free speed that can bring smaller costs. Just adhere to one principle: the actual speed used should not exceed the maximum safe speed under environmental constraints.

[0135] The relationship between the spatio-temporal path and the reachable domain. The spatio-temporal reachable domain evaluation based on the spatio-temporal prism defines the spatio-temporal boundary of the reachable domain for off-road operations. The spatio-temporal regions outside this boundary can either not be reached at the maximum speed or cannot reach the specified mission goal on time. Therefore, no matter what spatio-temporal trajectory is adopted, it is impossible to exceed the reachable spatio-temporal boundary delimited by the spatio-temporal prism. Similarly, the planned spatio-temporal trajectory cannot pass through the spatio-temporal range delimited by any dangerous escape spatio-temporal prism.

[0136] The principle that spatial trajectories can overlap or cross. The overlap or crossing of spatial trajectories does not refer to the situation in three-dimensional Euclidean space caused by a three-dimensional road network, but rather the overlap or crossing of different parts of the same trajectory in the original space. Traditional GIS-based path planning almost follows the same principle: that is, spatial trajectories do not overlap or cross. This is because the crossing of spatial trajectories will result in a closed loop path in space, meaning that the consumption of the corresponding section will be considered a wasted part, thus leading to a meaningless increase in cost. However, the overlap and crossing of spatial trajectories in special cases are meaningful, especially in the face of a complex real environment. For example, the mission bearer has an important event to complete at a certain "future" moment in the current space, but a severely destructive event is about to occur. The mission bearer must first avoid this destructive event and then come back to complete the relevant task. Planning the spatio-temporal trajectory of the above mission is very likely to result in the crossing of spatial trajectories, and even the escape dangerous path completely coincides with the return path after the danger has passed. However, this kind of path planning that overlaps or crosses in space has special significance. Since any movement takes time, although the trajectories coincide in the space coordinate system, as shown in Figure 13 (b), they cannot coincide in the spatio-temporal coordinate system, as shown in Figure 13 (a). In the spatio-temporal environment, only need to pay attention to whether the spatio-temporal path avoids the spatio-temporal region where the danger occurs, or obtains a lower cost, and there is no need to deliberately avoid the overlap or crossing of the planned trajectory in space.

[0137] In the solution of this case, the shortest time to complete the task is used as an indicator to evaluate the extreme performance ability of the task. The only comprehensive cost of the task considered is the cost of the shortest time consumed by off-road under the influence of the environment and capabilities. The evaluation is relatively simple and closest to traditional analysis, and can form a "point-like" understanding of the task performance.

[0138] The shortest time path planning takes the maximum moving speed as the object of environmental impact assessment and the time consumption as the optimization goal, and finds the shortest time and the corresponding path to reach the destination within a certain time and space range. It is a typical minimum cost path planning. In the solution of this case, the impact of environmental changes on the driving speed is reflected at all times, which is a mathematical model closer to the actual situation.

[0139] The boundaries of the time and space regions that may pose a devastating risk to the task and should be avoided due to the environment and situation, including the action-blocking regions caused by harsh climate or geographical environment, such as dangerous swamp regions, and the high-risk regions of landslides or collapses temporarily formed by rainfall; the dangerous regions caused by the situation, such as the concentrated fire coverage regions of the enemy and the ambushing regions of the enemy determined by reconnaissance; the time and space regions where the task actions are not suitable to appear, such as the regions that should be avoided during the covert maneuver passing through the coverage regions of the enemy observation posts and the regions overlapping with the time and space ranges of the enemy activities; the main purpose of its evaluation is to find the boundaries of the time and space regions that should be avoided in time and space, which can form a "surface-like" understanding of the task performance. Compared with the shortest time evaluation, the cognitive level is higher, and a certain overall understanding of the task can be obtained.

[0140] Under the influence of the environment, capabilities, and situation, the maximum time and space range that we may appear under the condition of being able to complete the off-road maneuver task also belongs to the evaluation of the boundaries of off-road activities. The reachable time and space domain is used to limit the time and space range of the off-road plan, and can reflect the comprehensive adaptability of the task to the environment to a certain extent. The premise of this analysis is that the time and space regions to be avoided formed by the environment and situation have been predicted. On this basis, first analyze according to the maximum activity range of our side under the influence of the environment and capabilities, and then exclude the time and space regions that should be avoided due to environmental events or situations. This evaluation is based on the evaluation of the boundaries of the time and space regions to be avoided, and can also form a "surface-like" understanding of the task performance, but it belongs to a more complete "surface-like" cognition and can form a complete understanding of the overall performance of the task.

[0141] Under the comprehensive influence of the environment, the optimal route that combines time, cost, and risk is an assessment of the optimal performance of a task. Calculating and evaluating the optimal performance of a task under the influence of the relevant environment and situation is the ultimate goal of the relevant assessment. The prerequisite for this assessment is to complete the assessment of the reachable spatio-temporal domain under the influence of the environment. On this basis, it is necessary to evaluate the risk and cost of the alternative routes, and search for the continuous route with the minimum cost and risk on the basis of meeting the maximum speed. This assessment is essentially to establish a "point-like" understanding of the task. However, different from the "point-like" understanding of extreme situations, etc., this assessment is based on the understanding of all aspects of the task and is the highest-level understanding of the task performance.

[0142] The solution of this case can fully reflect the continuous changes of the environment and situation over time by constructing a comprehensive environmental impact model, and can make full use of the time difference of space activities to solve space conflict problems. The calculation results of the task efficiency reflect the unity of time and space; it emphasizes that the battlefield situation, the capabilities and living habits of all parties involved, etc. are all regarded as factors equally important as the environmental impact, and are included in the comprehensive impact law of the task to achieve a more combat-like environmental impact efficiency assessment; it emphasizes multi-level, multi-angle, and all-round assessment. The efficiency assessment is not an evaluation of the performance of a certain aspect of the task, but a multi-angle and all-round evaluation of the task under the influence of the environment. It is a repeated process and a process from specific low-level understanding to gradually rising to abstract high-level understanding. The solution of this case uses multiple angles to evaluate the task efficiency, realizes the all-round evaluation of the task performance ability under the influence of the environment, including evaluating the extreme performance of the task in some aspects, evaluating the spatio-temporal range of the task activities, and evaluating the best comprehensive performance of the task, reflecting the all-round understanding of all levels of the task, so as to achieve the purpose of scientifically and effectively formulating strategies during task execution, reducing the operation load and complexity of the computing device, improving the model evaluation efficiency, and further verifying in the face of cross-country mobility tasks. The solution of this case can more scientifically and reasonably achieve the efficiency assessment and has good application prospects.

[0143] Unless otherwise specifically stated, the relative steps, numerical expressions, and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0144] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present invention, which are used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; 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 all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A task-oriented multi-dimensional effectiveness evaluation method, characterized in that, it includes the following content: Construct an effectiveness evaluation framework for multi-dimensional indicators of a task under the influence of the environment. This effectiveness evaluation framework is implemented based on a task-oriented spatio-temporal comprehensive environmental impact model. Among them, the comprehensive environmental impact model uses a function model that records the continuous change of the comprehensive environment with spatial position and time in a spatio-temporal coordinate system to quantify the impact of the environment on the task. The multi-dimensional indicators include: evaluating the extreme performance ability of the task with the shortest time to complete the task as an indicator based on the comprehensive environmental impact model, predicting and evaluating the spatio-temporal region boundaries that should be avoided for task activities formed by environmental and situation impacts, evaluating the spatio-temporal range that task activities can reach under the influence of environmental capabilities and situations, and the best performance of task activities under the comprehensive environmental impact; The spatio-temporal comprehensive environmental impact model is expressed as a mathematical model F = M(E(T), S(T), P(T)) of environmental variable E(T), situation variable S(T), and ability variable P(T) that changes continuously with time T. Among them, by determining a point on the ground surface as the origin, with the eastward direction as the X-axis, the northward direction as the Y-axis, and time T as the vertically upward Z-axis, a three-dimensional rectangular coordinate system is established, and this three-dimensional rectangular coordinate system is used as the spatio-temporal coordinate system; Perform rasterization processing on the spatio-temporal comprehensive environmental impact model. Divide the spatio-temporal into spatio-temporal unit cubes along the axes of the spatio-temporal coordinate system at the same spatial and time intervals, and use the spatio-temporal unit cubes as spatio-temporal voxels; Take the starting point and target point of the spatio-temporal coordinates where the task execution plan has been determined as anchor points, and use the spatio-temporal trajectory tracking algorithm to predict the shortest time cost of the straight-line segment of the task execution object moving along a straight line in space-time; Based on the shortest time cost, solve the spatio-temporal prism through path search, and use the solved spatio-temporal prism to obtain the spatio-temporal range that task activities can reach. Based on the principle that spatial trajectories can overlap or cross, search for the optimal path of task activities in the spatio-temporal environment with the goal of minimizing the comprehensive threat cost in the spatio-temporal range sections that task activities can reach in different time periods; Perform multi-dimensional evaluation on the effectiveness of the task's spatio-temporal environment impact based on the spatio-temporal range that task activities can reach, the degree of adaptation of the task to the environment corresponding to the shortest time cost, and the optimal execution plan within the spatio-temporal range of the feasible plan corresponding to the minimum threat degree cost.

2. The task-oriented multi-dimensional effectiveness evaluation method according to claim 1, characterized in that, Solve the spatio-temporal prism through path search based on the shortest time cost. The spatio-temporal trajectory of the movement between the centers of adjacent spatial grids is used as a continuous spatio-temporal trajectory connecting the current spatio-temporal point and the center of the target spatial grid, and is composed of multiple spatio-temporal straight-line segments that are connected end to end. Each spatio-temporal straight-line segment is in a different spatio-temporal voxel. Except for the current spatio-temporal point and the center of the target spatial grid, the spatio-temporal start point and end point of each segment are the spatio-temporal coordinates at which the spatio-temporal trajectory enters and exits the spatio-temporal voxel. The spatio-temporal trajectory tracking sequentially uses the spatio-temporal coordinates of the last time it exits the spatio-temporal voxel as the spatio-temporal start point of the movement of the next spatio-temporal straight-line segment, and combines the spatial movement direction, the spatio-temporal range of the spatio-temporal voxel, and the maximum speed of traveling in a specific direction within the spatio-temporal voxel to calculate the spatio-temporal coordinates when exiting the spatio-temporal voxel, until it reaches the center of the target spatial grid in space. Use the solved spatio-temporal prism to obtain the spatio-temporal range that the task activity can reach. Among them, the spatio-temporal prism includes: a dangerous escape spatio-temporal prism for describing the spatio-temporal area that the task activity should avoid and a reachable domain spatio-temporal prism for describing the reachable spatio-temporal area. Use the difference set of the spatio-temporal area ranges represented by the reachable domain spatio-temporal prism and the dangerous escape spatio-temporal prism to evaluate and predict the spatio-temporal range that the task activity can reach.

3. The task-oriented multi-dimensional effectiveness evaluation method according to claim 2, characterized in that, In an environment that changes continuously over time, assume that each possible spatial point that may be passed during off-road travel is composed of an earliest arrival time and a latest stay time to form the earliest arrival spatio-temporal point and the latest stay spatio-temporal point of this passing point in the spatio-temporal coordinate system; the earliest arrival spatio-temporal points and the latest stay spatio-temporal points of all possible passing points respectively form continuous lower and upper pyramidal spatio-temporal surfaces for describing future and past spatio-temporal surfaces; use the starting spatio-temporal point as the lower anchor point and the arrival spatio-temporal point as the upper anchor point, and use the lower and upper pyramidal spatio-temporal surfaces as the boundary surfaces. The spatio-temporal area between the two boundary surfaces is used as the reachable domain spatio-temporal prism area for representing the spatio-temporal range in which off-roaders who can complete the task on time may appear.

4. The task-oriented multi-dimensional effectiveness evaluation method according to claim 3, characterized in that, For the prediction and evaluation of the reachable area, use the starting spatio-temporal point as the lower anchor point and the arrival spatio-temporal point as the upper anchor point. Starting from the starting spatio-temporal point, calculate the earliest arrival time of each possible spatial point passed during task execution, and use the spatio-temporal points corresponding to these earliest arrival times to form the lower pyramidal boundary surface of the reachable domain spatio-temporal prism; Starting from the arrival spatio-temporal point, calculate in reverse the latest time that each possible spatial point passed during task execution can stay on the premise of being able to reach the end point on time, and use the spatio-temporal points corresponding to these latest times to form the upper pyramidal boundary surface of the reachable domain spatio-temporal prism; for the obtained upper and lower pyramidal boundary surfaces, perform an intersection operation to obtain the target spatio-temporal range represented by the reachable domain spatio-temporal prism.

5. The task-oriented multi-dimensional effectiveness evaluation method according to claim 2, characterized in that, The dangerous escape spatio-temporal prism region is divided into an occurrence area for describing the spatio-temporal range where a dangerous event occurs, a affected area for describing the spatio-temporal range that cannot be escaped due to limited mobility when the dangerous event spreads, and a vacuum area for describing the spatio-temporal range that cannot be reached without experiencing the dangerous event due to limited mobility. Among them, the lower spatio-temporal boundary of the affected area, the spatio-temporal boundaries of the occurrence area except for the upper and lower boundaries, and the upper spatio-temporal boundary of the vacuum area together divide a closed spatio-temporal region. Using this closed spatio-temporal region as the dangerous escape spatio-temporal prism for representing the spatio-temporal region range that must be avoided to complete the task or cannot be reached due to the dangerous event, the lower anchor point of the dangerous escape spatio-temporal prism is the spatio-temporal edge at the start time of the dangerous event, and the upper anchor point is the spatio-temporal edge at the end time of the dangerous event.

6. The task-oriented multi-dimensional effectiveness evaluation method according to claim 5, characterized in that, in the solution of the dangerous escape spatio-temporal prism, both the upper and lower boundary surfaces are within the spatial range where the dangerous event occurs. Among them, when solving the lower spatio-temporal boundary surface, starting from the moment when the dangerous event occurs, with the maximum speed allowed by the environment and the optimal path and using the spatial edge of the dangerous event as the destination, calculate backward in time the latest departure time for all spatial points within the dangerous event range to escape danger. The corresponding spatio-temporal points form a continuous conical spatio-temporal surface, and this conical spatio-temporal surface is the lower spatio-temporal edge of the dangerous escape spatio-temporal prism for indicating that when the spatio-temporal where the off-road vehicle is located is below this spatio-temporal surface, it can escape the upcoming dangerous event; when solving the upper spatio-temporal boundary surface, starting from the moment when the dangerous event ends, drive inward from the spatial edge of the dangerous event with the maximum speed allowed by the environment and the optimal path, and calculate forward in time the earliest time that all spatial points within the dangerous event range can reach after the dangerous event ends. The corresponding spatio-temporal points form a continuous conical spatio-temporal surface, and this continuous conical spatio-temporal surface is the upper spatio-temporal edge of the dangerous escape spatio-temporal prism for indicating that if the current off-road vehicle has not been affected by the dangerous event, it is necessarily impossible to appear below this spatio-temporal surface.

7. The task-oriented multi-dimensional effectiveness evaluation method according to claim 4 or 6, characterized in that, when calculating the spatio-temporal boundary surface forward or backward in the spatio-temporal voxel, establish an open list and a determined list. Directly add the point-like anchor points, hashed multi-point-like anchor points, line-like anchor points, and spatio-temporal points located at the center of the spatial grid near the edge of the surface-like anchor point to the determined list. Also add the spatio-temporal points located inside the surface-like anchor point at the center of the spatial grid and with a time near the surface-like anchor point to the determined list. Evaluate the shortest time cost for the center points of the spatial grids adjacent to the existing determined list, and add the formed spatio-temporal points to the open list; Find the space-time point with the earliest time in the open list and add it to the determined list. Use the path search algorithm to search in eight directions adjacent to the space based on this point, update the earliest arrival time of adjacent space-time points in the open list, or evaluate the center of the unevaluated space grid and add it to the open list. Then, find the space-time point with the shortest time in the open list again and add it to the determined list. Repeat this process until the earliest possible arrival time of each space grid center is calculated; when calculating the latest staying space-time point at the center position of the space grid, start from the space-time point with the latest staying time in the open list and search backward for the latest departure time of each space grid center.

8. The task-oriented multi-dimensional effectiveness evaluation method according to claim 1, characterized in that, in the voxelized space-time environment, set the space-time coordinates at the start and arrival of the task execution plan, and construct a straight line segment based on the space-time points entering and exiting the space-time voxel to obtain the space-time trajectory of the task execution plan; use the space-time coordinates of the space-time points where the task execution enters or exits the space-time voxel in the model to obtain the travel length and staying time within each space-time voxel passed by the space-time trajectory, and obtain the comprehensive environmental cost of the task execution plan through the cost growth rate of each space-time voxel passed by the space-time trajectory, the staying time and travel length within the corresponding space-time voxel; and perform hierarchical path search using time periods, starting from the first calculated time period within the reachable domain, using each node on the lower boundary of the reachable domain or the center point of the target voxel in the previous time period as the starting point, calculate the alternative space-time paths from the node to the center point of the target voxel in the current time period that meet the current speed condition; evaluate the multiple alternative paths connected to the center point of each target voxel in the current calculated time period, only retain the path with the smallest cumulative comprehensive threat cost among them, discard other alternative paths, and combine the minimum comprehensive threat cost of the corresponding starting point to evaluate the minimum cumulative comprehensive threat cost to the center point of the current target voxel; if in the current time period calculation, the reachable space-time domain includes the upper boundary of the space-time prism, use each node on the upper boundary as the center point of the target voxel to participate in the calculation, and evaluate the overall cumulative comprehensive threat cost from the corresponding node on the upper boundary to the final target point; Enter the next calculation time period until all calculations within 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 optimal path, and obtain the optimal execution plan within the feasible solution space-time range corresponding to the minimum threat degree cost based on this optimal path.

9. The task-oriented multi-dimensional effectiveness evaluation method according to claim 1, characterized in that, in the voxelized space-time environment, based on the time consumed by the linear movement between any two adjacent space-time nodes, the space-time trajectory between the start point and the target point of the task execution plan is obtained through path search to obtain the space-time trajectory with the least time consumption for the linear movement between the start point and the target point, and the adaptability of the task to the environment corresponding to the shortest time cost is obtained based on this space-time trajectory.

Images