A task extension method and system for autonomous aerial vehicle task executability recovery

Abstract

The application discloses a task extension method and system for autonomous aerial vehicle task executability recovery, and the method comprises the following steps: obtaining a task request and task context information of a user; analyzing the user's intention and generating structured task parameters according to the task request and the task context information; the structured task parameters comprise target weight parameters and constraint thresholds; generating a main task based on the task request, identifying the executability risk of the main task based on the current task state and the constraint thresholds; judging whether to generate an auxiliary task based on the executability risk, and generating an extended task based on the main task and the auxiliary task after the auxiliary task is generated; performing constraint instantiation processing based on the extended task, the target weight parameters and the constraint thresholds, generating a constraint optimization problem corresponding to a controller, and solving the control input sequence in a predicted time domain; encapsulating the extended task, the target weight parameters, the constraint thresholds and the control input sequence into a structured task scheme, and delivering the structured task scheme for execution.

Description

Technical Field

[0001] This invention relates to the field of autonomous aerial vehicle mission planning technology, and in particular to a mission extension method and system for restoring the executability of autonomous aerial vehicle missions. Background Technology

[0002] With the rapid development of applications such as low-altitude economy, intelligent transportation, emergency rescue, logistics distribution, infrastructure inspection, environmental monitoring, and temporary communication support, autonomous aerial vehicles (AAVs) are evolving from single-aircraft execution to multi-aircraft collaborative execution. Multi-autonomous aerial vehicle (MAV) systems can leverage the cooperation between different aircraft to complete complex tasks such as cargo transportation, area inspection, target arrival, disaster relief material delivery, and communication relay support, offering significant advantages in mission coverage, execution efficiency, and environmental adaptability. In these application scenarios, users typically submit mission requirements in the form of natural language, mission text, or high-level mission instructions, such as transporting goods from location A to location B. These mission instructions generally only describe the main mission objectives that the user expects to achieve, without fully specifying the safety constraints, communication constraints, energy constraints, payload constraints, path constraints, no-fly zone constraints, aircraft role allocation relationships, and dependencies between multiple tasks that must be satisfied during mission execution.

[0003] However, in actual low-altitude operating environments, multi-autonomous vehicle missions cannot be completed solely through path planning or single task assignment. During execution, the aircraft must simultaneously meet multiple system-level operational conditions. On one hand, the aircraft needs to avoid obstacles, no-fly zones, and other aircraft to prevent collisions, boundary violations, or entry into high-risk areas. On the other hand, the aircraft needs to maintain communication connections with base stations, edge nodes, or other aircraft under limited power, payload capacity, and computing resources. When communication blind spots exist in the mission area, the aircraft has insufficient remaining energy, the target location is far from communication coverage, the path traverses high-risk areas, or multiple missions compete for the same aircraft resources, the user's original mission instructions may not be directly executed. Even if a preliminary mission plan can be generated, problems such as communication interruptions, energy depletion, path unreachability, mission failure, or safety violations may easily occur during actual execution.

[0004] Existing systems typically require human operators to perform secondary analysis and correction of user tasks based on information such as aircraft status, environmental maps, communication coverage, remaining energy, and safety boundaries; or, assuming that the user's original task instructions are already executable, the task objectives are directly converted into aircraft allocation schemes and path plans, and local constraint processing is carried out in conjunction with safety controllers, communication rules, or energy constraints.

[0005] These methods all increase the burden of manual operation, reduce the system response speed and task execution autonomy, and most of them perform local optimization for a single constraint. It is difficult to uniformly analyze the mutual influence between multiple types of constraints in the task planning stage and generate a coordinated and consistent task structure and execution parameters accordingly.

[0006] In summary, existing technologies cannot effectively bridge the gap between high-level user mission requests and system-level executable solutions for multi-autonomous aerial vehicles. In particular, they struggle to automatically restore incomplete, fragile, or unexecutable mission instructions into structured mission solutions executable by the controller under constraints such as safety, communication, and energy.

[0007] Therefore, existing technologies still need improvement and development. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a mission extension method and system for restoring the mission executability of autonomous aerial vehicles, in view of the above-mentioned defects of the prior art. The aim is to solve the problem of mission extension that is difficult to achieve in terms of mission executability restoration and unified handling of multiple constraints in the prior art.

[0009] The technical solution adopted by this invention to solve the problem is as follows: In a first aspect, embodiments of the present invention provide a mission extension method for restoring the executability of autonomous aerial vehicle missions, the method comprising: Obtain the user's task request and task context information; Based on the task request and the task context information, the user intent is parsed and structured task parameters are generated; the structured task parameters include target weight parameters and constraint thresholds; A main task is generated based on the task request, and the executability risk of the main task is identified based on the current task status and the constraint threshold. Based on the executability risk, determine whether to generate an auxiliary task, and after generating the auxiliary task, generate an extended task based on the main task and the auxiliary task; Based on the extended task, the target weight parameters, and the constraint threshold, constraint instantiation is performed to generate the constraint optimization problem corresponding to the controller, and the control input sequence is obtained by solving it in the prediction time domain. The extended task, the target weight parameters, the constraint threshold, and the control input sequence are encapsulated into a structured task scheme and then sent out for execution.

[0010] Secondly, embodiments of the present invention also provide a mission extension system for restoring the executability of autonomous aerial vehicle missions, the system comprising: The task request acquisition module is used to acquire the user's task request and task context information; The task intent parsing module is used to parse the user intent and generate structured task parameters based on the task request and the task context information; the structured task parameters include target weight parameters and constraint thresholds; The risk identification module is used to generate a main task based on the task request, and to identify the executability risk of the main task based on the current task status and the constraint threshold. The auxiliary task extension module is used to determine whether to generate an auxiliary task based on the executability risk, and after generating the auxiliary task, to generate an extension task based on the main task and the auxiliary task; The constraint instantiation module is used to perform constraint instantiation processing based on the extended task, the target weight parameters, and the constraint threshold, generate the constraint optimization problem corresponding to the controller, and solve it in the prediction time domain to obtain the control input sequence. The structured task scheme generation and execution module is used to encapsulate the extended task, the target weight parameters, the constraint threshold, and the control input sequence into a structured task scheme and then issue it for execution.

[0011] Thirdly, embodiments of the present invention also provide a terminal, the terminal including a memory and one or more processors; the memory stores one or more programs; the programs include instructions for executing the mission extension method for mission executability recovery of autonomous aerial vehicles as described above; the processor is used to execute the programs.

[0012] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a plurality of instructions adapted to be loaded and executed by a processor to implement the steps of the mission extension method for mission executability recovery of autonomous aerial vehicles as described above.

[0013] The beneficial effects of this invention are as follows: This invention acquires the user's task request and task context information; based on the task request and task context information, it parses the user's intent and generates structured task parameters; the structured task parameters include target weight parameters and constraint thresholds; a main task is generated based on the task request, and the executability risk of the main task is identified based on the current task state and constraint thresholds; based on the executability risk, it determines whether to generate an auxiliary task, and after generating the auxiliary task, an extended task is generated based on the main task and auxiliary tasks; constraint instantiation is performed based on the extended task, target weight parameters, and constraint thresholds to generate the constraint optimization problem corresponding to the controller, and the control input sequence is solved in the prediction time domain; the extended task, target weight parameters, constraint thresholds, and control input sequence are encapsulated into a structured task scheme and issued for execution. This invention can automatically transform incomplete, fragile, or unexecutable user task instructions into a task scheme containing a main task, necessary auxiliary tasks, aircraft role allocation, task dependencies, constraint thresholds, and controller executable parameters, improving the feasibility, stability, and reliability of multi-autonomous aerial vehicle task execution. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating the mission extension method for restoring the executability of autonomous aerial vehicle missions provided in an embodiment of the present invention.

[0016] Figure 2 This is a technical architecture diagram of the mission extension method for restoring the executability of autonomous aerial vehicle missions provided in the embodiments of the present invention.

[0017] Figure 3 This is a schematic diagram of a scenario for a mission extension method for restoring the executability of autonomous aerial vehicle missions, provided in an embodiment of the present invention.

[0018] Figure 4 This is another scenario diagram of the mission extension method for restoring the executability of autonomous aerial vehicle missions provided in the embodiments of the present invention.

[0019] Figure 5 This is another scenario diagram of the mission extension method for restoring the executability of autonomous aerial vehicle missions provided in the embodiments of the present invention.

[0020] Figure 6This is another scenario illustration of the mission extension method for restoring the executability of autonomous aerial vehicle missions provided in the embodiments of the present invention.

[0021] Figure 7 This is a schematic diagram of a mission extension system for restoring the executability of autonomous aerial vehicle missions, provided in an embodiment of the present invention.

[0022] Figure 8 This is a schematic diagram of the terminal provided in the embodiment of the present invention. Detailed Implementation

[0023] This invention discloses a mission extension method and system for restoring the executability of autonomous aerial vehicle missions. To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention.

[0024] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0025] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0026] To address the aforementioned deficiencies in existing technologies, this invention provides a task extension method for recovering the executability of autonomous aerial vehicle missions. The method includes: acquiring a user's mission request and mission context information; parsing the user's intent and generating structured mission parameters based on the mission request and mission context information; the structured mission parameters including target weight parameters and constraint thresholds; generating a main task based on the mission request, and identifying the executability risk of the main task based on the current mission state and the constraint thresholds; determining whether to generate an auxiliary task based on the executability risk, and after generating the auxiliary task, generating an extended task based on the main task and the auxiliary task; performing constraint instantiation processing based on the extended task, the target weight parameters, and the constraint thresholds to generate a constraint optimization problem corresponding to the controller, and solving it in the prediction time domain to obtain a control input sequence; encapsulating the extended task, the target weight parameters, the constraint thresholds, and the control input sequence into a structured task scheme, and issuing it for execution.

[0027] This invention does not simply modify existing path planning or task allocation; instead, it introduces a proactive risk identification and task recovery mechanism. When a user's original request cannot be executed directly and safely in reality, the system does not abandon it or generate a flawed plan. Instead, while maintaining the user's primary task objective, it automatically diagnoses the problem and intelligently supplements necessary auxiliary tasks. This transforms an originally infeasible solution into a complete, robust, and directly executable structured task plan for the underlying controller. This enables multi-vehicle systems to autonomously improve task executability, reliability, robustness, and edge collaborative execution efficiency in complex and dynamic low-altitude environments.

[0028] like Figure 1 As shown, the method includes: Step S100: Obtain the user's task request and task context information.

[0029] Furthermore, the mission request includes one or more of the following: cargo transportation, target arrival, area inspection, emergency delivery, communication support, and multi-aircraft collaborative missions. The task context information includes: current state and historical task experience; the current state includes the current environment state and system state. The environmental status includes one or more of the following: environmental map, obstacle location, no-fly zone range, base station location, communication coverage radius, and mission target location; The system status includes one or more of the following: the position, speed, remaining energy, payload capacity, communication status, and availability status of each host aircraft. The historical mission experience includes one or more of the following: successful mission plans, failed mission cases, safety parameter ranges, common failure modes, communication interruption records, energy consumption records, and replanning experience.

[0030] Specifically, the method / system of this embodiment is typically deployed on a base station, edge node, or task scheduling platform. Starting with a user-input natural language task request, its purpose is to transform ambiguous high-level instructions into structured information that the system can understand and compute. In other words, the system does not passively receive instructions; the information it acquires is divided into two categories: task request and task context information.

[0031] A mission request represents the user's intent and can range from a single cargo transport or area inspection mission to a complex multi-aircraft collaborative mission involving communication support and emergency delivery. Mission context information, which forms the basis for the system to understand and assess the feasibility of mission objectives, consists of two parts: the current state and historical mission experience. The current state reflects relevant information about the physical world and system resources. Environmental state includes spatial and rule boundaries such as environmental maps, obstacles, no-fly zones, base station locations, and communication coverage radii. System state reflects available resources, i.e., the real-time dynamic attributes of each aircraft, such as position, speed, remaining energy, payload capacity, communication status, and availability. Historical mission experience introduces a time dimension, representing experiential data accumulated from past practices. It includes structured experience such as successful solutions, failure cases, safety parameter ranges, common failure modes, communication interruptions, and energy consumption records. This experience is crucial for subsequently identifying mission feasibility risks.

[0032] For example, a base station, edge node, or task scheduling platform receiving a natural language task request input by a user is denoted as: .in, This refers to the user task request received at decision time t, which includes, but is not limited to, cargo transportation, target arrival, area inspection, emergency delivery, communication support, or multi-aircraft collaborative tasks.

[0033] At the same time, the system obtains the current environment state and system state, and records them as the current state: .in, It should include at least the location, speed, remaining energy, payload capacity, communication status, availability status of each aircraft, as well as environmental maps, obstacle locations, no-fly zones, base station locations, communication coverage radius, and mission target locations.

[0034] In addition, the system retrieves historical task experience (also known as historical execution experience) related to the current task from the historical task experience database, denoted as: .in, It represents the historical mission experience information available at the current moment, including successful mission plans, failed mission cases, safety parameter ranges, common failure modes, communication interruption records, energy consumption records, and replanning experience.

[0035] Step S200: Based on the task request and the task context information, parse the user intent and generate structured task parameters; the structured task parameters include target weight parameters and constraint thresholds.

[0036] Specifically, the system combines the user's task request with the current state and historical experience to parse the user's intent, and further transforms it into structured task parameters, namely target weight parameters and constraint thresholds. For example, in emergency delivery missions, considering the disaster environment and historical cases of mission failure due to energy depletion, the system will interpret the user's intent as delivering the goods in the fastest and most reliable way while ensuring absolute safety, rather than in the most energy-efficient way.

[0037] The target weight parameters are a set of coefficients representing task priority, obtained based on the parsed user intent. In complex tasks, different objectives often conflict with each other, and the target weight parameters can effectively resolve the problem of multi-objective conflict. Furthermore, the target weight parameters include: primary task completion weight, auxiliary task execution weight, communication maintenance weight, energy constraint weight, and control smoothness weight.

[0038] Unlike weighted processing of conflicts between objectives, constraint thresholds define hard boundaries. These thresholds are fundamental conditions for task feasibility; violation of them may lead to task failure or safety incidents. Furthermore, constraint thresholds include: minimum safe distance, communication quality threshold, low energy safety margin, maximum speed limit, no-fly zone safety margin, and maximum number of replanning attempts.

[0039] For example, the system responds to task requests. Current status and historical mission experience Perform joint analysis to generate task parameters: ; in, This represents the task parsing and task parameter generation module; This represents the structured task parameters corresponding to the current task; Indicates the target weight parameter; This represents the feasibility constraint threshold.

[0040] The above formula provides a formal definition of the parameters for the natural language task and the potential task. The target weight parameter can be expressed as: ; in, Indicates the weight of the main task completion. Indicates the execution weight of auxiliary tasks. This indicates that communication maintains weight. Indicates the energy constraint weights. This indicates the weight for controlling smoothness.

[0041] Constraint threshold include: ; in, Indicates the minimum safe distance. Indicates the communication quality threshold. This indicates a low energy safety margin, and v_{\max} represents the maximum speed limit. Indicates the safety margin of the no-fly zone. This indicates the maximum number of replanning attempts. The mission structure must include at least the primary mission, auxiliary missions, aircraft roles, mission phases, mission dependencies, and mission completion conditions.

[0042] Step S300: Generate a main task based on the task request, and identify the executability risk of the main task based on the current task status and the constraint threshold.

[0043] Furthermore, the step of identifying the executability risk of the main task based on the current task status and the constraint threshold includes: Based on the current task status and the constraint threshold, identify whether the main task has one or more of the following risks: the target point is outside the communication coverage area, the path crosses obstacles or no-fly zones, the aircraft has insufficient remaining energy, the aircraft has insufficient payload capacity, the aircraft has resource conflicts, there is a risk of collision between aircraft, the task dependency relationship is incomplete, the controller input field is missing, or the task target is unreachable in the prediction time domain. If present, the identified risk will be considered as the executability risk of the main task.

[0044] Specifically, the system first generates a primary task based on the task request. The goal of this step is to ensure that the user's primary task intent remains unchanged; that is, subsequent task extensions can only supplement auxiliary tasks, but cannot delete, replace, or substantially alter the original primary task objective. For example, in a cargo transportation task, the primary task could be the fastest route plan directly connecting the pickup and delivery points. Next, the system analyzes the primary task in conjunction with the current task status and constraint thresholds, considering the current environmental state, aircraft resource status, communication coverage, safety constraints, and historical execution experience, to determine whether the original primary task carries any risk of being unexecutable or having low reliability.

[0045] In practical applications, the system sequentially checks for a series of specific risks in the main task, including: the target point being outside communication coverage, the path crossing obstacles or no-fly zones, insufficient remaining energy of the aircraft, insufficient payload capacity, resource conflicts, collision risks, incomplete task dependencies, missing controller input fields, and the task objective being completely unreachable within the prediction time domain. For example, the system compares the coordinates of each waypoint planned for the main task with the safety margin of no-fly zones; it calculates the total energy consumption of the path and determines whether it exceeds the difference between the aircraft's remaining energy and the low energy safety margin; and it checks whether the signal strength along the entire path remains above the communication quality threshold. The system can identify one or more of these risks simultaneously, and if any of the above checks fails, the identified problem is marked as an executability risk.

[0046] For example, the system extracts the main task objectives explicitly specified by the user from the user's task request and generates the main task structure: .in, This indicates the main task directly specified in the user's original task request. For example, in a cargo transportation task, the main task includes cargo identification, pickup point, delivery point, candidate aircraft, and delivery completion conditions; in an inspection task, the main task includes inspection area, inspection route, target point, and inspection completion conditions.

[0047] The system determines whether the main task poses a risk of being unexecutable or having low reliability based on the current task status, constraint thresholds, and the main task structure. For any security constraint, a security margin function can be defined: ; in, express The margin function corresponding to class-specific safety constraints. When When, it indicates that the current state satisfies the safety constraint; when When the condition is met, it indicates that the current state violates the safety constraint or poses a high risk. This type of safety margin is used to describe safety constraints such as collision avoidance, no-fly zone avoidance, and minimum separation requirements.

[0048] For example, for any two autonomous aerial vehicles and Its minimum interval safety constraint can be expressed as: ; in, and They represent aircraft and aircraft At any moment Location, Indicates the constraint threshold The minimum safe distance obtained from instantiation. When When the minimum safe separation requirement is met between the two aircraft, this indicates that the minimum safe separation requirement is met. This expression is an instantiation of the safety geometric constraint.

[0049] Step S400: Based on the executability risk, determine whether to generate an auxiliary task, and after generating the auxiliary task, generate an extended task based on the main task and the auxiliary task.

[0050] Furthermore, the step of determining whether to generate an auxiliary task based on the aforementioned executability risk includes: Based on the aforementioned executability risks, it is determined whether the main task faces a risk of failure; the risk of failure includes: unexecutability risk and / or low reliability risk; If present, a minimum necessary auxiliary task is generated based on the task request, the current task status, and the constraint threshold. The auxiliary task includes one or more of the following: communication relay task, security buffer task, energy protection task, task reallocation task, and replanning task.

[0051] Specifically, once an executability risk is identified, it is necessary to further determine whether the risk constitutes a mission failure risk by combining the aircraft status, environmental map, communication coverage, energy status, safety boundaries, and historical execution experience. Mission failure risks include two types: one is an unexecutability risk that makes the mission impossible to complete at all, such as the target point being completely in a communication blind spot or the aircraft's remaining energy not being able to cover any effective path; the other is a low reliability risk, that is, the mission may nominally have the possibility of completion, but during execution, it is highly likely to trigger a failure or interruption due to factors such as communication quality fluctuations, energy approaching critical values, or multiple aircraft paths being close together.

[0052] Once a risk of mission failure is identified, the system needs to automatically generate necessary auxiliary tasks and collaborative roles while keeping the user's primary mission objectives unchanged. These auxiliary tasks include automatically supplementing communication relay, safety buffering, route detours, energy protection, resource support, backup aircraft support, or mission reallocation, to improve mission executability.

[0053] The generated auxiliary tasks, through a task orchestration method that integrates the primary and auxiliary tasks, form an organic whole with the primary task. In this new mission structure, the aircraft group is assigned clear roles: carrier aircraft to execute the original core mission objectives, relay aircraft specifically responsible for maintaining communication links, backup aircraft on standby as resource or functional redundancy, and security auxiliary aircraft performing patrols or surveillance in critical areas. Simultaneously, the entire extended mission also has a clear mission phase orchestration, such as flying to the target point, cargo retrieval, relay maintenance, delivery, and return. This orchestration method ensures that all primary and auxiliary tasks are consistent in time, space, and logic, collectively forming a complete structured mission plan that includes aircraft roles, mission phases, mission dependencies, waypoints, mission triggering conditions, and mission completion conditions.

[0054] For example, when the system determines that the main task has a risk of being unexecutable or having low reliability, it calls the auxiliary task generation module to generate an auxiliary task based on the user's task request, the current state, and constraint thresholds. ; in, This indicates the auxiliary task generation module. This represents a set of auxiliary tasks generated to restore the executability of the task.

[0055] Auxiliary tasks include, but are not limited to: communication relay tasks, used to select an idle aircraft as a relay aircraft and set the relay hovering position when the mission target point or flight path is outside the communication coverage range of the base station; safety buffer tasks, used to increase the safety distance or generate detour tasks when the mission path is close to obstacles, no-fly zones or other aircraft; energy protection tasks, used to adjust mission allocation, set mid-course return, reduce speed or activate backup aircraft when the aircraft's remaining energy is insufficient or the mission path is too long; mission reassignment tasks, used to reselect the execution aircraft when the aircraft's payload is insufficient, resources conflict or mission fails; and replanning tasks, used to regenerate the mission structure and constraint parameters when communication interruption, controller infeasibility or external disturbance occurs during execution.

[0056] This embodiment generates auxiliary tasks based on the principle of minimum necessity, that is, it prioritizes generating the minimum necessary auxiliary tasks, thereby avoiding excessive modification to the user's main task objective: ; Simultaneously satisfy: ; in, Indicates the scale or number of auxiliary tasks. This indicates whether the task structure meets the executability verification.

[0057] The system combines the main task structure and the auxiliary task structure to obtain an extended task structure: ; in, This represents the final expanded task structure. The task structure includes a main task set, a secondary task set, a spacecraft role set, a task phase set, task dependencies, task triggering conditions, and task completion conditions.

[0058] Aircraft roles may include: ; in, Indicates the role of carrying the aircraft. Indicates the role of a communication relay aircraft. Indicates the role of a backup aircraft. Indicates the role of the inspection aircraft. This indicates the role of a safety-assisting aircraft.

[0059] Step S500: Based on the extended task, the target weight parameters, and the constraint threshold, perform constraint instantiation processing to generate the constraint optimization problem corresponding to the controller, and solve it in the prediction time domain to obtain the control input sequence.

[0060] Specifically, to enable the upper-level task planning results to directly serve downstream constraint execution, this embodiment designs a constraint instantiation mechanism oriented towards controller execution. This mechanism can transform constraints such as safety, communication, energy, load, and path into target weights, constraint thresholds, safety distances, communication thresholds, energy boundaries, and replanning strategies that can be read by the controller. This ensures that the output of task planning is no longer limited to abstract action sequences, but forms structured parameters that the controller can recognize, read, and solve.

[0061] In practical applications, the system transforms task semantics and operational constraints into executable parameters for the controller based on the aforementioned extended tasks, target weight parameters, and constraint thresholds. Specifically, target weight parameters such as main task completion weight, communication maintenance weight, and energy constraint weight can be instantiated as coefficients in the optimization objective function; while constraint thresholds such as minimum safe distance, communication quality threshold, low energy safety margin, maximum speed limit, and no-fly zone safety margin can be instantiated as hard constraints in the controller optimization problem, including safe distance, communication threshold, energy boundary, speed limit, and no-fly zone margin. Simultaneously, the system also instantiates risk-related strategies, such as the maximum number of replanning iterations, as failure handling strategies.

[0062] Furthermore, the system generates a constrained optimization problem corresponding to the controller based on the instantiation results, and solves it in the prediction time domain to obtain the control input sequence. The system utilizes the aforementioned objective weights and constraint thresholds to construct a constrained optimization problem within a finite prediction time domain. Its optimization objective is to maximize the weighted overall task performance while satisfying all the hard constraints of the instantiation. Subsequently, the controller solves this optimization problem in each control cycle, calculating an optimal control input sequence within the prediction time domain. This sequence contains the control commands for each aircraft at future moments, such as desired acceleration, attitude angular rate, or motor thrust, ultimately forming low-level commands that can be directly executed by the aircraft control system.

[0063] For example, the system constructs an optimization objective function usable by the downstream controller based on the extended task structure, target weights, and constraint thresholds. The cumulative cost function in the prediction time domain is defined as: ; For example, the system constructs a finite-time optimization problem for the downstream controller based on the extended task structure, objective weights, and constraint thresholds. Among these, Indicates the task parameters The parameterized single-step cost function and the cumulative cost function in the prediction time domain are: ; This cumulative cost is directly used as the optimization objective of the downstream controller, that is, the controller minimizes the cost under system dynamics and related constraints. Obtain the optimal control sequence in the prediction time domain.

[0064] in, Indicates the length of the prediction time domain. Represents the predicted state sequence. Indicates the control input sequence. This represents the instantaneous cost function.

[0065] The execution layer, based on structured task parameters Construct a constrained optimization problem and solve for the control input sequence in the prediction time domain. Its form can be expressed as: ; satisfy: ; ; ; in, This indicates the control strategy determined by the current task parameters. This represents a discrete-time state transition model. Represents the set of allowed control inputs. This represents the linearized safety constraint instantiated from the task parameters and constraint thresholds.

[0066] For safety constraints, a sampling barrier constraint is applied within each prediction step: ; in, This indicates the allowable shrinkage rate as a safety margin. When... When, it indicates that a non-negative safety margin is required for the next sampling time; when A smaller value indicates a more conservative limit on the decrease in safety margin.

[0067] During the controller solution process, the aforementioned nonlinear constraints can be locally linearized near the current nominal trajectory to obtain affine constraints: ; Safety constraints include collision avoidance, no-fly zone avoidance, and minimum separation.

[0068] Step S600: Encapsulate the extended task, the target weight parameter, the constraint threshold, and the control input sequence into a structured task scheme and send it out for execution.

[0069] Specifically, this embodiment can automatically generate auxiliary tasks, role assignments, task dependencies, feasibility thresholds, and controller executable parameters by combining environmental conditions, historical mission experience, and coupling constraints such as safety, communication, and energy, while keeping the user's main mission objectives unchanged. This transforms high-level mission requests into structured mission schemes suitable for collaborative execution by multiple autonomous aerial vehicles.

[0070] In practical applications, the extended task results are further transformed into structured parameters that the controller can directly use, including: task structure, role allocation, task dependencies, target weights, feasibility thresholds, safe distances, communication thresholds, energy constraints, and replanning strategies. This supports reliable collaborative execution of multiple autonomous aerial vehicles in complex low-altitude environments. Ultimately, while maintaining the user's primary task objectives, the main task, auxiliary tasks, role allocation, task dependencies, target weights, and constraint thresholds can be converted into structured task schemes. This automatically transforms incomplete, fragile, or unexecutable user task instructions into task schemes containing the main task, necessary auxiliary tasks, aircraft role allocation, task dependencies, constraint thresholds, and controller executable parameters, improving the feasibility, stability, and reliability of multi-autonomous aerial vehicle task execution.

[0071] Furthermore, the structured mission scheme includes at least: mission identifier, mission type, mission object, aircraft allocation result, aircraft role (role assignment), mission phase, mission dependencies, phase target point, waypoint, switching distance, control parameters (controller executable parameters), constraint parameters (constraint thresholds, such as safety constraints, communication constraints, and energy constraints), controller target weight, replanning strategy (maximum number of replanning attempts), and failure handling strategy, and can be directly invoked by downstream controllers. It is applicable to autonomous aerial vehicle systems that need to reliably execute missions under conditions of limited communication, limited energy, no-fly zone constraints, and dynamic environmental disturbances, such as low-altitude logistics delivery, emergency response, infrastructure inspection, and environmental monitoring.

[0072] For example, structured task solutions can be represented using JSON, XML, or other machine-readable formats, such as: .in, This represents a structured task scheme. Indicates the task identifier. This indicates an extended task structure. Represents a set of aircraft roles. Represents the set of waypoints. Represents the set of task phases. Indicates the constraint threshold. Indicates the target weight. This indicates failure handling and replanning strategies.

[0073] In one implementation, the method further includes: The system monitors the task execution status in real time and triggers replanning when the triggering conditions are met. The triggering conditions include one or more of the following: the safety margin function value is less than zero, the remaining energy of the aircraft is lower than the preset energy safety margin, the communication quality is lower than the preset communication quality threshold, and the constraint optimization problem corresponding to the controller is infeasible. When the number of replanning attempts reaches a preset threshold, the task will be downgraded or terminated.

[0074] Specifically, this embodiment also designs a dynamic replanning mechanism based on execution status. This mechanism can trigger task adjustment during execution based on security risks, communication quality, remaining energy or controller feasibility, and regenerate auxiliary tasks, role assignments, waypoints or constraint parameters.

[0075] For example, during mission execution, the system periodically acquires the aircraft's status and determines whether a replanning re-planning trigger is necessary based on the status. Replanning triggering conditions include, but are not limited to: ; ; ; ; in, Indicates the first The aircraft at all times The remaining energy, Indicates the current communication quality. This indicates whether the current control optimization problem is feasible.

[0076] When any of the multiple planning trigger conditions are met, the system updates the current state. The process involves re-executing the following steps: identifying the executability risk of the main task based on the current task status and the constraint threshold; determining whether to generate an auxiliary task based on the executability risk, and generating an extended task based on the main task and the auxiliary task after generating the auxiliary task; performing constraint instantiation processing based on the extended task, the target weight parameter, and the constraint threshold to generate the constraint optimization problem corresponding to the controller, and solving it in the prediction time domain to obtain the control input sequence; encapsulating the extended task, the target weight parameter, the constraint threshold, and the control input sequence into a structured task scheme, and issuing it for execution. This includes regenerating auxiliary tasks, role assignments, waypoints, or constraint parameters.

[0077] In one implementation, the method further includes: After the task is completed, record the task plan, execution trajectory, control parameters, success factors, reasons for failure, number of replanning attempts, safety violations, communication interruptions, energy consumption, and final task results, and update the historical task experience database.

[0078] Specifically, to reduce recurring errors and improve the executability of the initial task plan and the success rate of the final task, this embodiment also designs a task generation mechanism assisted by historical task experience. In this mechanism, the system can record the execution experience of successful and failed tasks, and retrieve and extract reusable task structures and failure avoidance strategies from subsequent similar tasks.

[0079] For example, after a task is completed, relevant information is recorded and the historical task experience database is updated: ; in, This indicates the updated historical mission experience database. This indicates the structured task plan for this task. Indicates the task execution trajectory. This indicates the task execution result. For successful tasks, the system extracts reusable task structure, role allocation pattern, safety parameter range, and stable execution conditions; for failed or near-failure tasks, the system extracts failure causes, risk patterns, and avoidance strategies. This task generation mechanism, aided by historical task experience, can be used for experience retrieval and solution generation for subsequent similar tasks.

[0080] like Figure 2 As shown in the figure, this diagram illustrates the closed-loop process in this embodiment, including user requirements, tool invocation, Mission RAG, task expansion, system parameter generation, base station-side scheduling, and execution result write-back.

[0081] like Figure 3 , Figure 4 , Figure 5 , Figure 6 As shown, the colored multi-line represents the execution trajectory of each autonomous aircraft during collaborative retrieval and delivery; gray squares represent obstacles; red dashed areas represent no-fly zones; green plus signs represent base stations; blue squares and triangles represent cargo retrieval and delivery locations; and diamonds represent relay points where autonomous aircraft hover to maintain communication when connectivity is required. These four figures illustrate the application scenarios of the method in this embodiment.

[0082] To verify the effectiveness of the method of this invention, a simulation environment for collaborative execution of multiple autonomous aerial vehicles was constructed. The experimental environment includes multiple available autonomous aerial vehicles, cargo retrieval and delivery tasks, obstacles, no-fly zones, base stations, communication coverage areas, and task target points. The system receives user task requests at the base station and generates structured task plans based on the vehicle status, environmental status, communication coverage, and historical task experience, before distributing them to the execution layer for constraint execution.

[0083] Table 1. Results of the Task Unexecutability Recovery Experiment

[0084] Table 1 is used to verify whether the present invention can restore incomplete, fragile, or initially unexecutable user task requests to an executable task solution. Evaluation metrics include the feasible-to-executable conversion rate, user intent satisfaction, initial execution feasibility, average number of replanning attempts, and final success rate.

[0085] Experimental results show that the feasible-to-executable conversion rate of the method of this invention reaches 0.81, the user intent satisfaction rate reaches 0.86, the initial execution feasibility reaches 0.88, the final success rate reaches 0.85, and the average number of replanning attempts is 0.60. Compared with fixed templates and A... Compared to waypoint planning, rule-based relay, and basic language planning schemes, the method of this invention outperforms all key indicators. This result demonstrates that, while maintaining the user's primary task objectives, this invention can automatically supplement missing elements in the original task, such as communication guarantees, safety buffers, path detours, energy protection, or resource reallocation, thereby transforming unreliable task requests into structured task solutions usable by the controller. Furthermore, the lower number of replanning iterations indicates that the initial task solution generated by this invention has higher execution quality, reducing repeated corrections and manual intervention during task execution.

[0086] Table 2. Experimental Results of Communication Relay Auxiliary Tasks

[0087] Table 2 is used to verify the ability of this invention to generate communication relay auxiliary tasks in communication-constrained scenarios. Experiments compared a relay-free scheme, a distance-rule-based relay scheme, a signal threshold-based relay scheme, a rule-based relay scheme with security control, and the method of this invention.

[0088] Experimental results show that the method of this invention achieves a task success rate of 0.87%, a main task completion rate of 0.89%, a communication interruption rate reduced to 0.05%, an invalid relay allocation rate reduced to 0.02%, and an average additional latency of 0.8. Compared with relay-free schemes, this invention significantly reduces the risk of communication interruption. Compared with rule-based relay schemes based on distance or signal thresholds, this invention can comprehensively consider the target location of the task, base station coverage, aircraft idle status, and main task execution requirements, thereby generating more reasonable relay roles and relay locations. This result indicates that even when the user's task request does not explicitly specify communication guarantee conditions, this invention can still proactively identify communication risks based on the current task environment and generate relay tasks through auxiliary task extensions, enabling the main task to maintain higher execution reliability in communication-restricted areas.

[0089] Table 3. Decomposition Experiment Results under Multiple Constraint Coupling Conditions

[0090] Table 3 is used to verify the performance of the present invention under various coupled constraints, including security, communication, and energy. The experiments compared the present invention with explicit security constraint control, the present invention with penalty control, the present invention without security barrier constraints, and the rule-based relay plus security control scheme.

[0091] Experimental results show that when combined with explicit safety constraint control, this invention achieves a task success rate of 0.86%, a safety violation rate of 0.02%, a communication interruption rate of 0.07%, a low-power exposure rate of 0.10%, a control infeasibility rate of 0.05%, and an average number of replanning attempts of 0.6. Compared with penalized control and schemes without safety barriers, this invention demonstrates superior performance in terms of task success rate, safety violation rate, communication interruption rate, low-power exposure rate, and control infeasibility rate. These results indicate that in multi-constraint coupled tasks, simply generating the task structure at the planning layer is insufficient to guarantee reliable execution. It is also necessary to transform the task extension results into constraint parameters usable by the controller and to uniformly process safety, communication, and energy constraints at the execution layer. This invention enhances the consistency between the planning and execution layers through a constraint instantiation mechanism, thereby improving execution stability in complex tasks.

[0092] Table 4. Ablation Results of Historical Experience Retrieval and Case Compilation Mechanisms

[0093] Table 4 further verifies the contribution of historical experience retrieval and case compilation mechanisms to the quality of task generation. The experiment compared the basic scheme, the basic scheme plus historical retrieval, the basic scheme plus historical retrieval and case compilation, and the complete scheme of this invention.

[0094] Experimental results show that the basic scheme has a structural format validity of 0.58, an initial execution feasibility of 0.43, an average number of replanning iterations of 1.4, and a final success rate of 0.63%. After adding historical retrieval, the structural format validity improved to 0.84, the initial execution feasibility improved to 0.79, and the average number of replanning iterations decreased to 0.9. Further improvement was achieved by incorporating case studies. The complete scheme of this invention achieved a structural format validity of 0.92, an initial execution feasibility of 0.88, an average number of replanning iterations decreased to 0.6, and a final success rate of 0.85. These results demonstrate that the historical task experience mechanism in this invention does not simply store logs, but rather improves the quality of the first-time task scheme generation by retrieving successful cases, analyzing failed cases, organizing safety parameter ranges, and extracting reusable task structures, thus providing structural priors and error avoidance information for subsequent task generation.

[0095] Table 5. Experimental results of system-level robustness under severe anomalous disturbances.

[0096] Table 5 is used to verify the system-level robustness of the present invention under severe abnormal events. The abnormality types include sudden gusts of wind, relay link failure, complete communication interruption, and mixed abnormalities.

[0097] Experimental results show that under mixed anomaly conditions, the method of this invention achieves a task success rate of 0.63%, a catastrophic failure rate of 0.11%, a collision rate of 0.15%, a communication interruption rate of 0.24%, a final stability of 0.60, and a centralized replanning count of 4.3. In comparison, the rule-based relay plus security control scheme achieves a task success rate of 0.48 under mixed anomaly conditions, and the basic language planning plus security control scheme achieves 0.39. These results demonstrate that when severe disturbances or impaired communication support occur during task execution, this invention can improve the system's continuous execution capability under abnormal conditions through task extension updates, centralized replanning, and secure recovery mechanisms. Especially in mixed anomaly scenarios, this invention maintains a high task success rate and final stability, indicating that it is applicable not only to conventional task scenarios but also to complex low-altitude task environments with sudden risks.

[0098] Table 6. Experimental Results of Abnormal Event Response and Recovery Delay

[0099] Table 6 is used to verify the response speed and recovery efficiency of the present invention after an abnormal event occurs. The evaluation indicators include event marking rate, event-to-detection latency, latency to detect security command, latency to detect replanning, event-to-recovery time, and recovery success rate.

[0100] Experimental results show that, under mixed anomaly conditions, the event labeling rate of the method of this invention is 0.92, the event-to-detection latency is 1.2, the security command detection latency is 0.9, the replanning detection latency is 1.9, the event-to-recovery time is 7.8, and the recovery success rate is 0.70%. In comparison, the event-to-recovery time for the rule relay plus security control scheme is 10.3, and for the basic language planning plus security control scheme it is 13.1. These results demonstrate that this invention not only improves task robustness before anomalies occur through task expansion and constraint instantiation, but also enables faster completion of event identification, security command issuance, centralized replanning, and stable recovery after anomalies occur, thereby shortening the anomaly response closed-loop time and improving the recovery success rate.

[0101] In summary, the beneficial effects of the present invention include, but are not limited to: (1) This invention can improve the executability of autonomous aerial vehicle mission plans. Existing mission planning methods usually generate mission plans directly based on high-level mission instructions input by the user, which easily ignores execution conditions such as communication, safety, energy, and aircraft resources. This invention can automatically determine whether the original mission has any unexecutability risks based on the analysis of the user's main mission intent, combined with the current environmental state, aircraft state, and historical execution experience, and generate a structured mission plan that includes the main mission, auxiliary missions, role allocation, and mission dependencies, thereby improving the feasibility of mission execution from user instructions to system execution.

[0102] (2) The present invention can restore the task execution capability while keeping the user's main task objective unchanged. When the original task has problems such as communication interruption, insufficient energy, unreachable path, security risk or resource conflict, the present invention does not change the user's core task objective, but automatically supplements necessary auxiliary tasks such as communication relay, security buffer, path detour, energy protection or task reallocation, so that the original incomplete or low reliability task request can be transformed into an executable task plan.

[0103] (3) This invention can improve the reliability of tasks in scenarios with multiple constraints. This invention instantiates parameters such as safety distance, communication threshold, energy boundary, target weight, and replanning strategy, and outputs them to the downstream controller for use, so that the task planning results can be effectively connected with the constraint execution process. This can reduce problems such as communication interruption, energy depletion, safety violations, or controller infeasibility caused by relying solely on single constraint optimization.

[0104] (4) This invention can reduce the need for manual intervention and repeated replanning. This invention records successful task plans and failed task cases through a historical task experience database, and retrieves and reuses relevant experience in subsequent tasks, thereby reducing problems such as incorrect task formatting, unreasonable parameter settings, missing auxiliary tasks, and repeated occurrences of known failure patterns. Experimental results show that after adopting historical task experience, the executability rate of the first task plan and the final success rate are both improved, and the number of replanning attempts is reduced.

[0105] Based on the above embodiments, the present invention also provides a mission extension system for restoring the executability of autonomous aerial vehicle missions, such as... Figure 7 As shown, the system includes: Task request acquisition module 01 is used to acquire the user's task request and task context information; The task intent parsing module 02 is used to parse the user intent and generate structured task parameters based on the task request and the task context information; the structured task parameters include target weight parameters and constraint thresholds; Risk identification module 03 is used to generate a main task based on the task request, and to identify the executability risk of the main task based on the current task status and the constraint threshold. The auxiliary task extension module 04 is used to determine whether to generate an auxiliary task based on the executability risk, and after generating the auxiliary task, to generate an extension task based on the main task and the auxiliary task; The constraint instantiation module 05 is used to perform constraint instantiation processing based on the extended task, the target weight parameters and the constraint threshold, generate the constraint optimization problem corresponding to the controller, and solve the control input sequence in the prediction time domain. The structured task scheme generation and execution module 06 is used to encapsulate the extended task, the target weight parameters, the constraint threshold, and the control input sequence into a structured task scheme and then send it out for execution.

[0106] Specifically, the system in this embodiment can be deployed on a base station, edge node, or task scheduling platform, and includes a task request acquisition module, a task intent parsing module, a risk identification module, an auxiliary task extension module, a constraint instantiation module, and a structured task scheme generation and execution module.

[0107] Furthermore, the system may also include a replanning module and a historical experience retrieval and update module.

[0108] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 8 As shown, the terminal includes a processor, memory, network interface, and display screen connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a mission extension method for restoring the executability of autonomous aerial vehicle missions. The display screen can be an LCD screen or an e-ink screen.

[0109] Those skilled in the art will understand that Figure 8 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0110] In one implementation, the terminal's memory stores one or more programs, and these programs are configured to be executed by one or more processors, including instructions for performing a mission extension method for mission executability recovery for autonomous aerial vehicles.

[0111] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0112] In summary, this invention discloses a task extension method and system for recovering the executability of autonomous aerial vehicle missions. The method includes: acquiring a user's task request and task context information; parsing the user's intent and generating structured task parameters based on the task request and task context information; the structured task parameters include target weight parameters and constraint thresholds; generating a main task based on the task request, and identifying the executability risk of the main task based on the current task state and the constraint thresholds; determining whether to generate an auxiliary task based on the executability risk, and generating an extended task based on the main task and the auxiliary task after generating the auxiliary task; performing constraint instantiation processing based on the extended task, the target weight parameters, and the constraint thresholds to generate a constraint optimization problem corresponding to the controller, and solving it in the prediction time domain to obtain a control input sequence; encapsulating the extended task, the target weight parameters, the constraint thresholds, and the control input sequence into a structured task scheme, and issuing it for execution. This invention can automatically transform incomplete, fragile, or unexecutable user task instructions into task schemes that include the main task, necessary auxiliary tasks, aircraft role allocation, task dependencies, constraint thresholds, and controller executable parameters, thereby improving the feasibility, stability, and reliability of multi-autonomous aerial vehicle task execution.

[0113] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A mission extension method for restoring the executability of autonomous aerial vehicle missions, characterized in that, The method includes: Obtain the user's task request and task context information; Based on the task request and the task context information, the user intent is parsed and structured task parameters are generated; the structured task parameters include target weight parameters and constraint thresholds; A main task is generated based on the task request, and the executability risk of the main task is identified based on the current task status and the constraint threshold. Based on the executability risk, determine whether to generate an auxiliary task, and after generating the auxiliary task, generate an extended task based on the main task and the auxiliary task; Based on the extended task, the target weight parameters, and the constraint threshold, constraint instantiation is performed to generate the constraint optimization problem corresponding to the controller, and the control input sequence is obtained by solving it in the prediction time domain. The extended task, the target weight parameters, the constraint threshold, and the control input sequence are encapsulated into a structured task scheme and then sent out for execution.

2. The mission extension method for mission executability recovery of autonomous aerial vehicles according to claim 1, characterized in that, The mission request includes one or more of the following: cargo transportation, target arrival, area inspection, emergency delivery, communication support, and multi-aircraft collaborative mission. The task context information includes: current state and historical task experience; the current state includes the current environment state and system state. The environmental status includes one or more of the following: environmental map, obstacle location, no-fly zone range, base station location, communication coverage radius, and mission target location; The system status includes one or more of the following: the position, speed, remaining energy, payload capacity, communication status, and availability status of each host aircraft. The historical mission experience includes one or more of the following: successful mission plans, failed mission cases, safety parameter ranges, common failure modes, communication interruption records, energy consumption records, and replanning experience.

3. The mission extension method for restoring mission executability of autonomous aerial vehicles according to claim 1, characterized in that, The target weight parameters include: main task completion weight, auxiliary task execution weight, communication maintenance weight, energy constraint weight, and control smoothness weight.

4. The mission extension method for mission executability recovery of autonomous aerial vehicles according to claim 1, characterized in that, The constraint thresholds include: minimum safe distance, communication quality threshold, low energy safety margin, maximum speed limit, no-fly zone safety margin, and maximum number of replanning attempts.

5. The mission extension method for mission executability recovery of autonomous aerial vehicles according to claim 1, characterized in that, The steps for identifying the executability risk of the main task based on the current task status and the constraint threshold include: Based on the current task status and the constraint threshold, identify whether the main task has one or more of the following risks: the target point is outside the communication coverage area, the path crosses obstacles or no-fly zones, the aircraft has insufficient remaining energy, the aircraft has insufficient payload capacity, the aircraft has resource conflicts, there is a risk of collision between aircraft, the task dependency relationship is incomplete, the controller input field is missing, or the task target is unreachable in the prediction time domain. If present, the identified risk will be considered as the executability risk of the main task.

6. The mission extension method for restoring mission executability of autonomous aerial vehicles according to claim 1, characterized in that, The steps for determining whether to generate an auxiliary task based on the aforementioned executability risk include: Based on the aforementioned executability risks, it is determined whether the main task faces a risk of failure; the risk of failure includes: unexecutability risk and / or low reliability risk; If present, a minimum necessary auxiliary task is generated based on the task request, the current task status, and the constraint threshold. The auxiliary task includes one or more of the following: communication relay task, security buffer task, energy protection task, task reallocation task, and replanning task.

7. The mission extension method for mission executability recovery of autonomous aerial vehicles according to claim 1, characterized in that, The method further includes: The system monitors the task execution status in real time and triggers replanning when the triggering conditions are met. The triggering conditions include one or more of the following: the safety margin function value is less than zero, the remaining energy of the aircraft is lower than the preset energy safety margin, the communication quality is lower than the preset communication quality threshold, and the constraint optimization problem corresponding to the controller is infeasible. When the number of replanning attempts reaches a preset threshold, the task will be downgraded or terminated.

8. A mission extension system for restoring the executability of autonomous aerial vehicle missions, characterized in that, The system includes: The task request acquisition module is used to acquire the user's task request and task context information; The task intent parsing module is used to parse the user intent and generate structured task parameters based on the task request and the task context information; the structured task parameters include target weight parameters and constraint thresholds; The risk identification module is used to generate a main task based on the task request, and to identify the executability risk of the main task based on the current task status and the constraint threshold. The auxiliary task extension module is used to determine whether to generate an auxiliary task based on the executability risk, and after generating the auxiliary task, to generate an extension task based on the main task and the auxiliary task; The constraint instantiation module is used to perform constraint instantiation processing based on the extended task, the target weight parameters, and the constraint threshold, generate the constraint optimization problem corresponding to the controller, and solve it in the prediction time domain to obtain the control input sequence. The structured task scheme generation and execution module is used to encapsulate the extended task, the target weight parameters, the constraint threshold, and the control input sequence into a structured task scheme and then issue it for execution.

9. A terminal, characterized in that, The terminal includes a memory and one or more processors; the memory stores one or more programs; the programs contain instructions for executing the mission extension method for mission executability recovery of autonomous aerial vehicles as described in any one of claims 1 to 7; the processors are used to execute the programs.

10. A computer-readable storage medium storing a plurality of instructions thereon, characterized in that, The instructions are applicable to being loaded and executed by a processor to implement the steps of the mission extension method for mission executability recovery of autonomous aerial vehicles as described in any one of claims 1 to 7.

Images