A multi-low-altitude intelligent agent cooperative path planning method and system, electronic equipment and storage medium

Abstract

A path planning method and system of multi-low-altitude intelligent agent cooperation, electronic equipment and storage medium, relate to the field of flight management, and alleviate the problems of insufficient utilization of vertical airspace resources and low airspace carrying capacity in the prior art. A path planning method obtains a three-dimensional airspace model for a target low-altitude airspace, and then obtains a low-altitude airspace constraint model; based on the task type information of each low-altitude intelligent agent, a flight height constraint model is obtained; a path search method is used to obtain a phased flight path of each low-altitude intelligent agent, and then obtain spatiotemporal occupancy information; the spatiotemporal occupancy information is subjected to conflict detection to plan the phased flight path; the phased flight path is subjected to smoothing, optimization processing and executability verification; and the phased flight path that passes the verification is used as an executable flight path. The present application is suitable for multi-intelligent agent cooperative path planning under conditions such as complex urban low-altitude environment and low-altitude task density improvement.

Description

Technical Field

[0001] This invention relates to the field of flight management, and more specifically to the field of path planning for low-altitude flight. Background Technology

[0002] With the development of technologies related to the low-altitude economy, low-altitude intelligent agents, as flight intelligent agents with autonomous perception, decision-making, and execution capabilities, have been gradually applied in fields such as urban logistics distribution, emergency rescue, urban inspection, and low-altitude transportation connections. In these applications, multiple low-altitude intelligent agents typically need to operate simultaneously within the same urban low-altitude airspace, performing various flight operations according to different mission requirements. At the same time, the urban low-altitude environment is generally characterized by dense buildings, limited available airspace, and frequent dynamic changes in flight missions, placing higher demands on the flight management and path planning of low-altitude intelligent agents.

[0003] With the continuous expansion of low-altitude intelligent agent applications, existing low-altitude flight management and path planning technologies have gradually revealed certain limitations in actual operation. For example, they are unable to fully meet the needs of complex urban low-altitude operation scenarios in terms of airspace utilization efficiency, dynamic environment adaptability, multi-aircraft collaborative safety, and mission differentiation management.

[0004] Regarding airspace utilization, existing solutions limit the flight of low-altitude agents to fixed altitudes or a few discrete altitude layers to reduce the risk of vertical conflicts between different flying objects. While this method reduces the complexity of spatial search, it is relatively inefficient in utilizing vertical airspace resources. As the number of low-altitude agents increases, congestion can easily occur within the limited altitude range, making it difficult to effectively improve the overall airspace carrying capacity.

[0005] In response to dynamic changes in the flight environment, existing technologies often employ static modeling or low-frequency updates for no-fly zones, temporary control areas, and areas with sudden risks. When airspace conditions change, it is often necessary to completely replan existing flight paths, or rely solely on simple local detour rules, which makes it difficult to balance planning efficiency and flight safety, thus affecting the system's real-time response capabilities.

[0006] In multi-low-altitude agent cooperative flight scenarios, most existing conflict detection and avoidance methods are based on spatial distance, focusing primarily on the relative positions of low-altitude agents at a given moment, lacking systematic modeling of the time dimension of the flight process. This approach struggles to accurately predict potential conflicts in the future, easily leading to delayed or overly conservative conflict assessments, thus impacting overall operational efficiency.

[0007] In the case of multiple tasks running in parallel, existing technologies usually adopt a uniform path planning and scheduling strategy for task management. This fails to fully consider the differences between different tasks in terms of flight altitude, timeliness, and safety level, and lacks a refined management mechanism for task type and priority. This can easily lead to tasks with high timeliness or high safety requirements being affected by other tasks.

[0008] In path planning, since path planning is mostly based on discrete space models, the flight paths obtained by existing technologies are usually in the form of broken lines, containing a lot of redundant waypoints. There is a lack of systematic connection between the path planning results and the actual flight control, and the engineering feasibility needs to be further improved.

[0009] In summary, existing multi-low-altitude intelligent agent collaborative path planning technologies suffer from several problems, including insufficient utilization of vertical airspace resources, low airspace carrying capacity, poor real-time response capability under dynamic changes in the flight environment, delayed or overly conservative conflict judgment, high-timeliness or high-safety-requirement tasks being affected by other tasks, and a lack of systematic connection between path planning results and actual flight control. Summary of the Invention

[0010] This invention alleviates the problems of existing multi-low-altitude intelligent agent collaborative path planning technologies, such as insufficient utilization of vertical airspace resources, low airspace carrying capacity, poor real-time response capability under dynamic changes in the flight environment, delayed or overly conservative conflict judgment, influence of other tasks on high-time-efficiency or high-safety-requirement tasks, and lack of systematic connection between path planning results and actual flight control. This invention provides the following solution: Option 1: A path planning method for multi-low-altitude intelligent agents in cooperation, wherein the method involves path planning for several low-altitude intelligent agents; the method includes the following steps: Step S01: Perform three-dimensional discrete modeling on the target low-altitude airspace to obtain a three-dimensional airspace model; the three-dimensional airspace model includes impassable areas; based on the impassable areas in the three-dimensional airspace model, obtain a low-altitude airspace constraint model. Step S02: Based on the task type information of each low-altitude agent, obtain the flight altitude constraint model; Step S03: Based on the three-dimensional airspace model, the low-altitude airspace constraint model, and the flight altitude constraint model, a path search method is used to obtain the phased flight path of each low-altitude intelligent agent; based on the phased flight path of each low-altitude intelligent agent, the spatiotemporal occupancy information in the target low-altitude airspace is obtained. The spatiotemporal occupancy information includes several spatial units and corresponding occupancy time intervals; Step S04: If the impassable area, the task type information, the phased flight path, and the number of low-altitude intelligent agents change, then perform conflict detection on the spatiotemporal occupancy information. Based on the result of the conflict detection, and in conjunction with the task priority, perform parameter adjustment and / or path replanning on the relevant paths; obtain the phased flight path for each low-altitude intelligent agent; otherwise, proceed to step S05. Step S05: Smooth and optimize the phased flight path, and verify the executability of the processed phased flight path. The phased flight path that passes the verification is taken as an executable flight path.

[0011] Furthermore, in one embodiment of the present invention, step S02 is to obtain the corresponding flight altitude layer based on the task type information of each low-altitude intelligent agent; and to establish a mapping relationship between the task type and the flight altitude layer based on the flight altitude layer of each low-altitude intelligent agent to obtain a flight altitude constraint model.

[0012] Furthermore, in one embodiment of the present invention, the establishment of the mapping relationship between mission type and flight altitude layer can be done in any of the following ways: Hierarchical mapping based on task priority: Assigning tasks with high timeliness and high security level to specific height layers, thus creating spatial isolation between them and regular tasks in the vertical direction; Hierarchical mapping based on aircraft performance: Based on the flight performance parameters of different types of low-altitude intelligent agents, they are assigned to altitude layers suitable for their operational characteristics; Fixed altitude mapping based on mission type: For a specific mission type, the target altitude value for its cruise flight is directly specified.

[0013] Furthermore, in one embodiment of the present invention, the path search method described in step S03 is implemented using the A* algorithm. The A* algorithm determines a feasible flight path from the starting position to the target position by evaluating the cost of spatial units in the three-dimensional airspace and performing heuristic search.

[0014] Furthermore, in one embodiment of the present invention, the A* algorithm is an improved A* algorithm. The improved A* algorithm introduces a spatiotemporal occupancy detection mechanism during the expansion of neighboring spatial units in path search. When checking the accessibility of neighboring spatial units, the algorithm determines the static state of the spatial unit, calculates the estimated time for the low-altitude agent to reach the spatial unit, and checks whether the spatial unit and its neighboring spatial units have been occupied by other spatial agents during the time period. If there is a spacetime occupancy conflict, the neighboring spatial unit is not passable at the current moment and it is necessary to search for other paths or adjust the flight time to avoid the conflict. The static state refers to the presence of obstacles in the unit or the unit being a no-fly zone.

[0015] Furthermore, in one embodiment of the present invention, the phased flight path in step S03 includes a takeoff phase flight path, a cruise phase flight path, and a landing phase flight path.

[0016] Furthermore, in one embodiment of the present invention, the conflict detection in step S04 is to detect the time interval of different low-altitude intelligent agents in the space unit in the spatiotemporal occupancy information, and determine whether there is time overlap or potential overlap in the same or adjacent space units.

[0017] Option 2: A path planning system for multi-low-altitude intelligent agents, wherein the system performs path planning for several low-altitude intelligent agents, and includes the following modules: Module 1 is used to perform three-dimensional discrete modeling of the target low-altitude airspace to obtain a three-dimensional airspace model; the three-dimensional airspace model includes impassable areas; based on the impassable areas in the three-dimensional airspace model, a low-altitude airspace constraint model is obtained. Module 2 is used to obtain a flight altitude constraint model based on the task type information of each low-altitude agent; Module 3 is used to obtain the phased flight path of each low-altitude intelligent agent based on the three-dimensional airspace model, the low-altitude airspace constraint model, and the flight altitude constraint model, using a path search method; and to obtain the spatiotemporal occupancy information in the target low-altitude airspace based on the phased flight path of each low-altitude intelligent agent. The spatiotemporal occupancy information includes several spatial units and corresponding occupancy time intervals; Module 4 is used to perform conflict detection on the spatiotemporal occupancy information if the impassable area, the task type information, the phased flight path, and the number of low-altitude intelligent agents change; based on the result of the conflict detection, and in combination with the task priority, to perform parameter adjustment and / or path replanning on the relevant paths; to obtain the phased flight path for each low-altitude intelligent agent; otherwise, proceed to Module 5; Module 5 is used to smooth and optimize the phased flight path, and to perform executability verification on the processed phased flight path. The phased flight path that passes the verification is used as an executable flight path.

[0018] Option 3: An electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in memory, it implements the method described in Scheme 1.

[0019] Option 4: A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method described in Option 1.

[0020] The present invention discloses a multi-low-altitude intelligent agent collaborative path planning method, system, electronic device, and storage medium, which effectively alleviates the problems of insufficient utilization of vertical airspace resources, low airspace carrying capacity, poor real-time response capability under dynamic changes in the flight environment, delayed or overly conservative conflict judgment, high-time-efficiency or high-safety-requirement tasks being affected by other tasks, and lack of systematic connection between path planning results and actual flight control. Specific beneficial effects include: 1. The path planning method described in this invention achieves a complete technical closed loop from path generation and conflict resolution to path optimization and verification through mechanisms such as three-dimensional airspace discrete modeling and spatial indexing, mission type-based altitude hierarchical management, phased path generation, dynamic no-fly zone updates and affected path identification, spatiotemporal conflict detection and avoidance strategies, and path smoothing and safety verification. This improves the executability and engineering applicability of low-altitude agent path planning results. Compared with existing technologies, this invention can improve airspace utilization efficiency and collaborative operation capabilities under conditions of increased low-altitude mission density and dynamic environmental changes, and enhance the effectiveness of response to dynamic risk areas and multi-agent conflict resolution.

[0021] 2. The path planning method described in this invention introduces a spatiotemporal occupancy modeling approach for flight paths to perform conflict detection, thereby enabling collaborative avoidance among multiple low-altitude agents. Furthermore, it can make real-time adjustments based on dynamic changes in the flight environment, improving real-time response capabilities and preventing delayed or overly conservative conflict judgments. This is beneficial in operating environments with multiple tasks running in parallel and high flight density, effectively reducing the risk of flight conflicts and enhancing the safety and orderliness of collaborative operation among multiple low-altitude agents. Compared to existing technologies that only employ simple avoidance strategies, it has stronger adaptability and engineering practical value.

[0022] 3. The path planning method described in this invention introduces a path smoothing optimization mechanism after completing path planning and collaborative processing. By continuously optimizing the planned path, the path is made more in line with the flight control characteristics of low-altitude intelligent agents, solving the problem of lack of systematic connection between path planning results and actual flight control, thereby reducing flight control complexity and improving the stability of path execution.

[0023] The method described in this invention is applicable to the field of collaborative path planning among multiple intelligent agents under complex urban low-altitude environments, increased low-altitude mission density, and dynamic environmental changes. Attached Figure Description

[0024] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of the path planning method described in Implementation Method 1. Detailed Implementation

[0025] Various embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. The embodiments described with reference to the drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0026] Implementation Method 1: This implementation method describes a path planning method for multi-low-altitude intelligent agents, wherein the method involves path planning for several low-altitude intelligent agents; the method includes the following steps: Step S01: Perform three-dimensional discrete modeling on the target low-altitude airspace to obtain a three-dimensional airspace model; the three-dimensional airspace model includes impassable areas; based on the impassable areas in the three-dimensional airspace model, obtain a low-altitude airspace constraint model. Step S02: Based on the task type information of each low-altitude agent, obtain the flight altitude constraint model; Step S03: Based on the three-dimensional airspace model, the low-altitude airspace constraint model, and the flight altitude constraint model, a path search method is used to obtain the phased flight path of each low-altitude intelligent agent; based on the phased flight path of each low-altitude intelligent agent, the spatiotemporal occupancy information in the target low-altitude airspace is obtained. The spatiotemporal occupancy information includes several spatial units and corresponding occupancy time intervals; Step S04: If the impassable area, the task type information, the phased flight path, and the number of low-altitude intelligent agents change, then perform conflict detection on the spatiotemporal occupancy information. Based on the result of the conflict detection, and in conjunction with the task priority, perform parameter adjustment and / or path replanning on the relevant paths; obtain the phased flight path for each low-altitude intelligent agent; otherwise, proceed to step S05. Step S05: Smooth and optimize the phased flight path, and verify the executability of the processed phased flight path. The phased flight path that passes the verification is taken as an executable flight path.

[0027] In this embodiment, in step S03, preferably after generating the phased flight path, the flight path is evaluated and selected to determine the flight path result that meets the preset optimization objective. The optimization objective is used to characterize the comprehensive requirements of the flight path in terms of flight efficiency, operational safety, or feasibility.

[0028] In this embodiment, the spatiotemporal occupancy information mentioned in step S03 is to calculate the time and distance attributes of the phased flight path to determine the duration of each path segment and the time information of arriving at each path position when the low-altitude intelligent agent flies along the flight path.

[0029] In this embodiment, after generating the phased flight path in step S03, time and distance calculations are performed on the flight path to determine the flight sequence of the low-altitude agent on each path segment. Based on this flight sequence, the occupancy relationship of the flight path in the three-dimensional airspace is mapped to the corresponding spatial unit, and the occupancy relationship is registered as spatiotemporal occupancy information to describe the time interval of the low-altitude agent's occupancy within a specific spatial unit. This provides a time dimension basis for path identification, conflict detection, and collaborative handling during the collaborative operation of multiple low-altitude agents, enabling the analysis of collaborative operation of multiple low-altitude agents.

[0030] In this embodiment, step S04, which involves adjusting the parameters of the relevant paths and / or replanning the paths based on the results of the conflict detection and in conjunction with task priorities, is as follows: Based on the updated airspace model and operating conditions, correlation analysis is performed on existing flight paths to identify flight paths that are related to the changes in spatial location and / or flight time, forming a set of paths that require further analysis.

[0031] In this embodiment, the related path mentioned in step S04 refers to processing only the flight paths that may be affected, avoiding repeated analysis of unaffected paths and improving system operating efficiency.

[0032] In this embodiment, the related path mentioned in step S04 refers to processing only the flight paths that may be affected, avoiding repeated analysis of unaffected paths and improving system operating efficiency.

[0033] In this embodiment, step S04, based on the result of the conflict detection, performs coordinated processing on the relevant flight paths, taking into account the priority information of the flight paths involved and the tasks performed by the low-altitude intelligent agents.

[0034] The coordinated handling includes adjusting the flight path's operating parameters and / or replanning the flight path, and can select appropriate handling methods according to different conflict situations, so as to reduce or eliminate the risk of conflict between multiple low-altitude intelligent agents while ensuring the operation requirements of high-priority tasks, and realize the orderly coordinated operation of multiple low-altitude intelligent agents.

[0035] Implementation Method 2: This implementation method further defines the path planning method described in Implementation Method 1. In this implementation method, step S01 involves performing three-dimensional discrete modeling of the target low-altitude airspace to obtain a three-dimensional airspace model; the three-dimensional airspace model includes impassable areas; and based on the impassable areas in the three-dimensional airspace model, a low-altitude airspace constraint model is obtained.

[0036] In this embodiment, the impassable area includes no-fly zone information, obstacle information, terrain restriction information, and other flight restriction conditions.

[0037] In this embodiment, the three-dimensional discrete modeling includes the following steps: Step S011: Obtain the spatial boundary information of the target low-altitude airspace. The spatial boundary information includes at least the horizontal area and the permitted flight altitude range.

[0038] Step S012: The spatial boundary information is converted into a unified three-dimensional spatial coordinate system, and the continuous low-altitude airspace is divided into regularized units under this coordinate system.

[0039] The above-mentioned transformation of spatial boundary information is to facilitate the unified processing of spatial locations.

[0040] The regularization partitioning involves dividing the low-altitude airspace into segments in the horizontal and vertical directions, discretizing the continuous low-altitude airspace into multiple spatial units with defined spatial locations and volume ranges. Each spatial unit represents a minimum management unit in the low-altitude airspace and is associated with a corresponding spatial location index to support rapid location and status query of the spatial unit.

[0041] Step S013: After completing the spatial unit division, initialize the operating status of each spatial unit.

[0042] Based on the pre-obtained information, the inaccessible areas are mapped to the corresponding spatial units in the spatial coordinate system as the coverage of the constraints. Spatial units that overlap with the constraints in terms of spatial location or height are marked as inaccessible areas; the remaining spatial units are marked as accessible areas.

[0043] This embodiment further defines step S01 and provides an example of it. The method discretizes the low-altitude operating area to form a set of spatial units for path planning and state management, and marks no-fly zones, obstacles, and other impassable areas with status. In this way, the spatial structure information, coordinate position relationships, and impassable areas of the low-altitude airspace are uniformly mapped to the set of spatial units, constructing a low-altitude airspace operating model that can be used for path planning and collaborative management, and forming a low-altitude airspace operating constraint model that can be used for path planning, state updates, and multi-low-altitude intelligent agent collaborative analysis. This provides basic spatial support for subsequent altitude-based path generation, dynamic path adjustment, and conflict detection and handling, facilitating subsequent path planning and collaborative management.

[0044] This implementation method is applicable to more refined and efficient modeling and management of low-altitude airspace in complex urban low-altitude environments, so as to make full use of vertical airspace resources and improve the concurrent operation capability of low-altitude intelligent agents while ensuring flight safety.

[0045] Implementation Method 3: This implementation method further defines the path planning method described in Implementation Method 1. In this implementation method, step S02 is to obtain the corresponding flight altitude layer based on the task type information of each low-altitude agent; and to establish a mapping relationship between the task type and the flight altitude layer based on the flight altitude layer of each low-altitude agent to obtain a flight altitude constraint model.

[0046] Implementation Method Four: This implementation method further defines the path planning method described in Implementation Method Two. In this implementation method, the establishment of the mapping relationship between mission type and flight altitude layer can be done in any of the following ways: Hierarchical mapping based on task priority: Assigning tasks with high timeliness and high security level to specific height layers, thus creating spatial isolation between them and regular tasks in the vertical direction; Hierarchical mapping based on aircraft performance: Based on the flight performance parameters of different types of low-altitude intelligent agents, they are assigned to altitude layers suitable for their operational characteristics; Fixed altitude mapping based on mission type: For a specific mission type, the target altitude value for its cruise flight is directly specified.

[0047] In this embodiment, the high timeliness and high safety level are constraints set by the user of low-altitude logistics delivery or emergency rescue, for example, emergency rescue tasks require high timeliness. The high safety level can be set according to the object being transported.

[0048] The specific altitude layer can be configured by manually setting the mission type of the aircraft for each flight altitude layer.

[0049] Application Example: In a scenario where urban low-altitude logistics delivery and emergency rescue are combined, the following mapping relationship is established: emergency medical supply delivery tasks are assigned to the 500-meter altitude layer to reduce conflicts with conventional traffic and ensure rapid passage for emergency missions; regular express delivery tasks are assigned to the 300-meter altitude layer, serving as the primary operating airspace for daily delivery tasks; and short-distance instant delivery tasks are assigned to the 150-meter altitude layer, suitable for small delivery aircraft operating at low altitudes. Through this hierarchical mapping, low-altitude agents with different operational characteristics are vertically distributed in an orderly manner, reducing the probability of conflict and improving airspace utilization efficiency.

[0050] In this embodiment, different flight altitude layers correspond to different altitude ranges. The flight demand range represents the altitude interval in the low-altitude airspace that is available for low-altitude intelligent agents to cruise and fly. The association between adjacent altitude layers can be configured according to operational needs to provide alternative altitude options when the main altitude layer is unavailable.

[0051] In this embodiment, the available flight altitude range of the low-altitude agent is determined from the corresponding flight altitude layer according to the mission type of the low-altitude agent, and the flight altitude is allocated in combination with the current usage of the altitude layer.

[0052] When the flight altitude layer meets the availability conditions, the cruising flight altitude of the low-altitude intelligent agent is restricted to the altitude range corresponding to the altitude layer; when the flight altitude layer does not meet the availability conditions, an alternative altitude range can be selected from the adjacent altitude layers associated with it for allocation.

[0053] This embodiment further defines step S02 and provides an example of it. After completing the three-dimensional discrete modeling and operational constraint construction of the low-altitude airspace, the method performs hierarchical management and allocation of the flight altitude of the low-altitude agents, so that low-altitude agents of different mission types form relatively independent altitude distributions in space. This reduces the risk of conflict between multiple low-altitude agents in the vertical direction and provides altitude constraints for subsequent phased flight path generation and collaborative operation. It effectively reduces the risk of conflict between multiple agents in the vertical direction, realizes the refined utilization of low-altitude airspace resources, and provides basic support for the concurrent operation and path planning of multiple low-altitude agents.

[0054] This implementation achieves priority-driven intelligent scheduling, which combines different task types and their operational characteristics to reasonably distinguish and schedule the flight altitude and operational strategies of low-altitude intelligent agents, thereby reducing mutual interference between different tasks.

[0055] Implementation Method 5: This implementation method further defines the path planning method described in Implementation Method 1. In this implementation method, the path search method in step S03 is implemented using the A* algorithm. The A* algorithm determines a feasible flight path from the starting position to the target position by evaluating the cost of spatial units in the three-dimensional airspace and performing a heuristic search.

[0056] Implementation Method Six: This implementation method further defines the path planning method described in Implementation Method Five. In this implementation method, the A* algorithm is an improved A* algorithm. The improved A* algorithm introduces a spatiotemporal occupancy detection mechanism during the expansion of neighboring spatial units in the path search process. When checking the traversability of neighboring spatial units, it determines the static state of the spatial unit, calculates the estimated time for the low-altitude agent to reach the spatial unit, and checks whether the spatial unit and its neighboring spatial units have been occupied by other aerial agents within that time period. If a spatiotemporal occupancy conflict exists, the neighboring spatial unit is not traversable at the current moment, and it is necessary to search for other paths or adjust the flight time to avoid the conflict. The static state refers to the presence of obstacles in the unit or the unit being a no-fly zone.

[0057] This implementation further defines the A* algorithm and provides an example. Existing standard A* algorithms primarily consider obstacle avoidance in the spatial dimension when performing path search, searching for the optimal path from the starting point to the destination by evaluating the drivability and movement cost of spatial units. However, in multi-low-altitude agent cooperative flight scenarios, in addition to avoiding static obstacles and no-fly zones, it is also necessary to avoid conflicts between different low-altitude agents during flight. This requires the path search algorithm to consider not only the spatial dimension but also the temporal dimension.

[0058] When using the aforementioned Standard A algorithm to determine the accessibility of a spatial cell, it only checks whether the cell is an obstacle or a no-fly zone, without considering whether the cell will be occupied by other low-altitude agents at some future time. When multiple low-altitude agents perform path planning simultaneously, if the search is based solely on the spatial dimension, it can easily lead to conflicts in time and space between the paths planned by different agents.

[0059] Therefore, this embodiment improves the standard A* algorithm to enable the A* algorithm to consider the spatiotemporal coordination relationship between multiple low-altitude agents during the path generation stage, thereby achieving early prediction and avoidance of conflicts and generating flight paths that meet spatiotemporal safety constraints.

[0060] Implementation Method Seven: This implementation method further defines the path planning method described in Implementation Method One. In this implementation method, the phased flight path in step S03 includes the takeoff phase flight path, the cruise phase flight path, and the landing phase flight path.

[0061] In this embodiment, the takeoff phase flight path describes the flight process of the low-altitude agent ascending from its starting position to its assigned flight altitude; the cruise phase flight path describes the flight process of the low-altitude agent flying from the end position of the takeoff phase to the start position of the landing phase within the assigned flight altitude range, based on a path search algorithm; and the landing phase flight path describes the flight process of the low-altitude agent descending from its cruise altitude to its target position.

[0062] In this embodiment, a phased constraint strategy is adopted: During takeoff, the low-altitude agent ascends vertically from its starting position to the assigned cruising altitude; the path during this phase is a vertical straight line without horizontal movement. During cruising, the low-altitude agent flies horizontally within the cruising altitude range, using a path search algorithm to plan the optimal path to avoid obstacles and no-fly zones. During landing, the low-altitude agent descends vertically from the cruising altitude to the target position; the path during this phase is also a vertical straight line. By applying differentiated motion constraints and path planning methods to different phases, the generated flight path conforms to the flight control characteristics of the low-altitude agent and is easy to interface with and execute with the flight control system.

[0063] Implementation Method 8: This implementation method is a further limitation of the path planning method described in Implementation Method 1. In this implementation method, the conflict detection in step S04 is to detect the time interval of different low-altitude intelligent agents in the space unit in the spatiotemporal occupancy information, and to determine whether there is time overlap or potential overlap in the same or adjacent space units.

[0064] In this embodiment, the conflict detection results include the time and location of the conflict, the distance between the conflicting parties, and the identifiers of the low-altitude agents involved. Task priorities are divided into high, medium, and low levels based on timeliness and importance. High-priority tasks (such as emergency rescue) maintain their original paths during conflicts, while low-priority tasks (such as routine delivery) require adjustment or replanning to avoid high-priority tasks. Parameter adjustment methods include takeoff time delay and flight speed adjustment. Takeoff time delay refers to postponing the takeoff time of a new task by a fixed interval (e.g., 60 seconds) for multiple attempts. Flight speed adjustment refers to adjusting the flight speed within a configured speed range according to the severity of the conflict. If parameter adjustments cannot resolve the conflict, the low-priority path is replanned.

[0065] This embodiment further defines step S04 and provides an example of conflict detection. The conflict detection analyzes the time intervals occupied by different low-altitude intelligent agents in a space cell to determine whether there is time overlap or potential overlap in the same or adjacent space cells, thereby enabling early identification of potential conflict risks during future flights.

[0066] Existing low-altitude agent path planning schemes, in scenarios involving multiple aircraft operating in parallel, typically handle flight conflicts based solely on spatial distance or simple avoidance rules, lacking systematic modeling of the flight path's temporal dimension. This makes it difficult to identify and coordinately handle potential conflicts in advance. This new method introduces a spatiotemporal occupancy modeling approach for flight paths, combining the flight paths of low-altitude agents in the three-dimensional airspace with temporal information. It detects conflicts in the flight paths of multiple low-altitude agents based on the temporal overlap within spatial units. Upon detecting potential conflicts, it adjusts parameters and / or replans the relevant flight paths based on task priorities to achieve collaborative avoidance.

[0067] This implementation enables the rapid identification of affected flight paths and efficient adjustment of relevant paths when the low-altitude environment changes dynamically during flight, thereby improving the low-altitude flight management system's response capability to dynamic no-fly zones and sudden risks.

[0068] This implementation method achieves conflict detection and avoidance. In the scenario of multi-low-altitude intelligent agent cooperative flight, a conflict detection and avoidance mechanism that considers both spatial position and flight time is established to enable early prediction and reasonable handling of potential conflicts.

[0069] Implementation Method Nine: This implementation method further defines the path planning method described in Implementation Method One. In this implementation method, the smoothing and optimization processing, no-fly zone verification, and executability verification described in step S05 are all common path processing methods in the prior art. Path smoothing and optimization processing has been widely used in the fields of robot path planning and UAV navigation. For example, the Douglas-Peucker algorithm is used to remove redundant waypoints, or spline interpolation methods are used to make the path smoother and more continuous to adapt to the kinematic constraints of the aircraft. In the fields of autonomous driving and mobile robots, path smoothing processing is used to convert discrete planned paths into executable continuous trajectories.

[0070] No-fly zone verification and feasibility verification are standard procedures in aviation flight management systems. Flight plans must undergo airspace restriction checks before execution to ensure compliance with no-fly zones, obstacles, and other operational constraints. In UAV flight management systems, pre-flight verification is used to verify whether the planned path meets airspace management requirements and safety standards.

[0071] This embodiment applies the above-mentioned existing technical methods to a multi-low-altitude intelligent agent collaborative path planning scenario. After completing path generation and conflict handling based on spatiotemporal occupancy, the path is smoothed and optimized to improve feasibility. A unified verification is performed before the path is output to ensure that the path meets all operational constraints.

[0072] This embodiment further defines step S05 and provides an example of it. Existing low-altitude agent path planning schemes typically focus on generating flight paths under basic airspace constraints, and their output results are mostly discrete path points or polyline paths. This method smooths and optimizes the flight path to reduce redundant waypoints and improve the continuity between adjacent path segments, making the flight path more consistent with the actual flight control characteristics of low-altitude agents.

[0073] Implementation Method 10: This implementation method further defines the path planning method described in Implementation Method 1. In this implementation method, the executability verification in step S05 includes verification of no-fly zones, obstacle areas, and other operational restrictions.

[0074] If the flight path fails the feasibility verification, the process can be returned to the aforementioned path adjustment or planning steps for reprocessing until a flight path that meets the operational requirements is obtained.

[0075] This embodiment further defines step S05 and provides an example of executability verification. In the prior art, the path planning result is directly used for flight execution after constraint judgment is completed in the generation stage. It lacks a mechanism for unified verification of the flight path before path output, which easily leads to no-fly zone conflicts or violations of operational constraints during actual operation. The executability verification of this method is used to determine whether the flight path meets low-altitude airspace operational constraints and flight safety requirements throughout the entire flight process. It performs unified verification of no-fly zones, obstacles, and other flight restrictions, avoiding safety conflicts from being triggered again during the execution stage of the planned path. This enhances the reliability of the path planning result and reduces the need for temporary corrections or interruptions during flight. Therefore, it ensures that the flight path is spatially executable.

Claims

1. A path planning method for multi-low-altitude intelligent agents working together, characterized in that, The method involves path planning for several low-altitude intelligent agents; the method includes the following steps: Step S01: Perform three-dimensional discrete modeling on the target low-altitude airspace to obtain a three-dimensional airspace model; the three-dimensional airspace model includes impassable areas; based on the impassable areas in the three-dimensional airspace model, obtain a low-altitude airspace constraint model. Step S02: Based on the task type information of each low-altitude agent, obtain the flight altitude constraint model; Step S03: Based on the three-dimensional airspace model, the low-altitude airspace constraint model, and the flight altitude constraint model, a path search method is used to obtain the phased flight path of each low-altitude intelligent agent; based on the phased flight path of each low-altitude intelligent agent, the spatiotemporal occupancy information in the target low-altitude airspace is obtained. The spatiotemporal occupancy information includes several spatial units and corresponding occupancy time intervals; Step S04: If the impassable area, the task type information, the phased flight path, and the number of low-altitude intelligent agents change, then perform conflict detection on the spatiotemporal occupancy information. Based on the result of the conflict detection, and in conjunction with the task priority, perform parameter adjustment and / or path replanning on the relevant paths; obtain the phased flight path for each low-altitude intelligent agent; otherwise, proceed to step S05. Step S05: Smooth and optimize the phased flight path, and verify the executability of the processed phased flight path. The phased flight path that passes the verification is taken as an executable flight path.

2. The path planning method according to claim 1, characterized in that, Step S02 involves obtaining the corresponding flight altitude layer based on the task type information of each low-altitude agent; and establishing a mapping relationship between the task type and the flight altitude layer based on the flight altitude layer of each low-altitude agent to obtain a flight altitude constraint model.

3. The path planning method according to claim 2, characterized in that, The mapping relationship between mission types and flight altitude layers can be established in any of the following ways: Hierarchical mapping based on task priority: Assigning tasks with high timeliness and high security level to specific height layers, thus creating spatial isolation between them and regular tasks in the vertical direction; Hierarchical mapping based on aircraft performance: Based on the flight performance parameters of different types of low-altitude intelligent agents, they are assigned to altitude layers suitable for their operational characteristics; Fixed altitude mapping based on mission type: For a specific mission type, the target altitude value for its cruise flight is directly specified.

4. The path planning method according to claim 1, characterized in that, The path search method described in step S03 is implemented using the A* algorithm. The A* algorithm determines a feasible flight path from the starting position to the target position by evaluating the cost of spatial units in the three-dimensional airspace and performing heuristic search.

5. The path planning method according to claim 4, characterized in that, The A* algorithm is an improved A* algorithm. In the process of expanding the neighbor space cell in the path search, the improved A* algorithm introduces a spatiotemporal occupancy detection mechanism. When checking the accessibility of the neighbor space cell, it judges the static state of the space cell, calculates the estimated time for the low-altitude agent to reach the space cell, and checks whether the space cell and its neighboring space cells have been occupied by other space agents during the time period. If there is a spacetime occupancy conflict, the neighboring spatial unit is not passable at the current moment and it is necessary to search for other paths or adjust the flight time to avoid the conflict. The static state refers to the presence of obstacles in the unit or the unit being a no-fly zone.

6. The path planning method according to claim 1, characterized in that, The phased flight path described in step S03 includes the takeoff phase flight path, the cruise phase flight path, and the landing phase flight path.

7. The path planning method according to claim 1, characterized in that, The conflict detection in step S04 involves detecting the time intervals of different low-altitude agents within a spatial unit in the spatiotemporal occupancy information, and determining whether there is any time overlap or potential overlap within the same or adjacent spatial units.

8. A path planning system for multi-low-altitude intelligent agents working together, characterized in that, The system performs path planning for several low-altitude intelligent agents and includes the following modules: Module 1 is used to perform three-dimensional discrete modeling of the target low-altitude airspace to obtain a three-dimensional airspace model; the three-dimensional airspace model includes impassable areas; Based on the impassable areas in the three-dimensional airspace model, a low-altitude airspace constraint model is obtained. Module 2 is used to obtain a flight altitude constraint model based on the task type information of each low-altitude agent; Module 3 is used to obtain the phased flight path of each low-altitude intelligent agent based on the three-dimensional airspace model, the low-altitude airspace constraint model, and the flight altitude constraint model, using a path search method; and to obtain the spatiotemporal occupancy information in the target low-altitude airspace based on the phased flight path of each low-altitude intelligent agent. The spatiotemporal occupancy information includes several spatial units and corresponding occupancy time intervals; Module 4 is used to perform conflict detection on the spatiotemporal occupancy information if the impassable area, the task type information, the phased flight path, and the number of low-altitude intelligent agents change; based on the result of the conflict detection, and in combination with the task priority, to perform parameter adjustment and / or path replanning on the relevant paths; to obtain the phased flight path for each low-altitude intelligent agent; otherwise, proceed to Module 5; Module 5 is used to smooth and optimize the phased flight path, and to perform executability verification on the processed phased flight path. The phased flight path that passes the verification is used as an executable flight path.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images