Heavy-load unmanned aerial vehicle collaborative operation and construction method for iron tower foundation pouring

Abstract

The heavy-load unmanned aerial vehicle collaborative operation and construction method for iron tower foundation pouring is characterized by comprising the following steps that S1, a collaborative operation system is constructed, the system is composed of three core modules including a heterogeneous unmanned aerial vehicle formation module, a ground control station and task intelligent core module and an intelligent load interface system module, and all the modules are complementary in function and cooperatively linked. According to the construction method, based on the collaborative operation system, the construction method follows a closed-loop digital process of planning-execution-fault tolerance, it is ensured that each step meets the process requirement of time-limited synchronous pouring of concrete, and the construction method is specifically divided into three stages including the first stage of task planning and feasibility analysis, the second stage of time-limited synchronous pouring, the third stage of time-limited synchronous pouring and the fourth stage of time-limited synchronous pouring and the fourth stage of time-limited synchronous pouring. Stage 2: time synchronization scheduling and autonomous execution; and stage 3: dynamic fault tolerance and seamless replacement. The method has the characteristics of intelligence, high efficiency and high safety, and is suitable for remote areas.

Description

Technical Field

[0001] This invention belongs to the field of transportation and pouring technology of tower foundation concrete for iron towers, and specifically relates to a heavy-duty UAV collaborative operation and construction method for pouring iron tower foundations. Background Technology

[0002] When constructing power transmission towers in remote or complex terrains (such as mountains and forests), the transportation and pouring of the tower foundation concrete presents a critical and challenging task. Existing technologies have the following main limitations: 1. Traditional ground transportation has significant limitations: Relying on heavy trucks, cement mixers, etc., requires the construction of temporary roads, which is not only costly and time-consuming, but also causes enormous damage to the natural environment. In protected areas or areas with harsh geological conditions, this method is almost impossible.

[0003] 2. Traditional air transport solutions are expensive and dangerous: While using helicopters to transport concrete is an alternative, its operating costs are extremely high, the capacity per trip is limited, and it is highly susceptible to weather conditions (such as wind, rain, and fog). In addition, helicopters pose extremely high safety risks when operating in narrow valleys or near obstacles.

[0004] 3. Bottlenecks in Single-UAV Technology: Although heavy-duty drones with single-unit payload capacities of tens to hundreds of kilograms have emerged on the market, the efficiency of single-unit round-trip transportation is low for the large and continuous supply of concrete required for foundation pouring, making it difficult to meet the stringent time requirement that the concrete must be poured before initial setting. Simply increasing the size of a single drone would bring a series of problems such as stability, energy consumption, and cost.

[0005] There is a mismatch in current drone applications: Currently, drones in the power industry are mainly used for lightweight inspections and mapping, designed for information collection rather than the physical transport and handling of heavy materials. Existing multi-drone collaborative technologies are mostly applied to lightweight package delivery or academic research; their control logic and hardware systems cannot meet the needs of industrial-grade heavy-duty transportation, especially for scheduling continuous material flows (such as concrete) with strict time windows.

[0006] In summary, existing technologies lack an integrated solution that can provide a safe, efficient, low-cost, and environmentally friendly continuous and timely supply of concrete for tower foundations in remote areas. Summary of the Invention

[0007] The purpose of this invention is to provide a heavy-duty UAV collaborative operation and construction method for pouring iron tower foundations. This method is characterized by high efficiency and high safety and is suitable for remote areas.

[0008] The core idea of ​​this invention is to organize multiple medium-sized payload drones into a time-synchronized "virtual aerial conveyor belt" to achieve continuous and uninterrupted transportation of concrete from the ground mixing plant to the tower foundation pouring point.

[0009] The core of the technical solution of this invention includes two parts: a "collaborative operation system" and a "construction method based on the system".

[0010] To achieve the above objectives, the technical solution adopted by the present invention is a heavy-duty UAV collaborative operation and construction method for pouring iron tower foundations, characterized by the following steps: S1. Construct a collaborative work system: The collaborative operation system is the core support for achieving continuous pouring. It consists of three core modules: "heterogeneous UAV formation," "ground control station and mission intelligence core," and "intelligent payload interface system." Each module complements the other's functions and works in synergy. (1) Heterogeneous UAV Squadron: Composed of at least two types of UAVs with different functions. The squadron includes at least two types of UAVs with different functions, covering the entire process of "environmental perception-material transportation" through division of labor and cooperation: "Mapping UAV (Type-S)": Adopting a lightweight and highly mobile design, equipped with LiDAR and high-definition cameras, its core function is to perform centimeter-level precision three-dimensional digital modeling of the pouring point and surrounding environment before operation, providing accurate environmental data support for subsequent path planning and safe operation. "Material Transportation UAV (Type-M)": Designed specifically for medium-load transportation of concrete and other materials (e.g., single-unit load up to 40 kg), with "high reliability and high transportation efficiency" as its core objectives, it has functions such as automated navigation, precise take-off and landing, and remote unloading function controlled by ground station pilots; at the same time, the squadron is equipped with a sufficient number of transportation UAVs (N_available) to meet the material needs of continuous operation and provide redundancy support for fault tolerance backup. When facing heavy and uneven materials, multiple UAVs can work together to maintain force balance and ensure safe material transportation.

[0011] (2) Ground Control Station (GCS) and Mission Intelligence Core: Ground Control Station: The command center of the entire system, providing a human-machine interface for mission definition, process monitoring, emergency intervention, and providing pilots with telemetry data and real-time video feedback for precise unloading. Mission Intelligence Core: The core software system deployed within GCS, serving as the system's "brain." It integrates the core algorithm of this invention—Time Synchronized Continuous Material Flow Scheduling Algorithm (TSS-CMFSA)—with specific functions including: ① 3D Visualization and Planning: Load and display the digital twin model, allowing operators to specify concrete mixing areas and pouring points in a virtual environment.

[0012] ② Task allocation and scheduling: Run a dynamic task allocation framework based on multi-agent deep reinforcement learning (MARL).

[0013] ③ Trajectory planning: Run a constraint-aware collaborative trajectory planning algorithm to generate safe and efficient flight paths for the fleet.

[0014] (3) Intelligent load interface system: This system is a key connecting module for realizing "unmanned operation". The drone is equipped with an automatic loading / unloading module, which can accurately dock with the ground loading device to realize the automated acceptance of materials, improve the operation efficiency and reduce human operation error.

[0015] S2 Construction Method: Based on the aforementioned collaborative operation system, the construction method of this invention follows a closed-loop digital process of "planning-execution-fault tolerance," ensuring that each step meets the technological requirement of "time-limited synchronous pouring" of concrete, specifically divided into three stages: Phase 1: Task Planning and Feasibility Analysis; Before the operation begins, the operator inputs the core parameters of the pouring task at the ground control station, including: total material requirement (V_total), material critical time window (T_critical_min, i.e., initial setting time of concrete), single-unit load (C_drone), total number of available drones (N_available), flight path distance (D_path_m), drone flight speed, loading / unloading time, etc. The task intelligence core conducts feasibility analysis through the TSS-CMFSA algorithm, the process of which is as follows: 1) Calculate the complete cycle time (T_cycle_s) for a single UAV to complete the "loading-flight-pouring-return" process. 2) Verify whether the delivery time of a single material delivery is much shorter than the critical time window to ensure that the concrete remains active during transportation; 3) Optimize the calculation scheduling interval (T_dispatch_interval_s) and set it to be consistent with the single-machine pouring time (t_pour_s) to achieve seamless connection of pouring; 4) Based on the cycle time and scheduling interval, calculate the minimum number of active drones (N_active_req) required to maintain a continuous flow. 5) Verify that the total number of available drones (N_available) meets the total requirement of "number of active drones + number of backup drones (N_standby_req)". The task is deemed "feasible" only if all verification items pass, and can proceed to the next stage.

[0016] Phase Two: Time Synchronization Scheduling and Autonomous Execution; After the task is initiated, the system first divides the UAVs into queues: "Ready Queue," "Standby Queue," and "ActiveList," and then executes the task according to the following logic: Using the "dispatch interval (T_dispatch_interval_s)" calculated in Phase 1 as the precise beat, drones are dispatched sequentially from the standby queue to take off and perform transportation tasks; 2) Each transport drone flies autonomously along a preset route, sequentially going through the state cycle of "loading" → "autonomously flying to the work site" → "waiting for manual unloading instructions" → "remotely unloading" → "autonomously returning to the base".

[0017] 3) Through precise sequential scheduling, multiple drones form an aerial virtual conveyor belt that connects end to end, ensuring that there are always drones continuously pouring above the pouring point, thus avoiding interruption of material supply.

[0018] Phase Three: Dynamic Fault Tolerance and Seamless Replacement; During operation, the system continuously monitors the operational status of all active drones, and immediately activates the fault tolerance mechanism in case of any anomaly. 1) If a drone malfunctions (such as power failure) or deviates significantly from the scheduled time, the system will immediately remove it from the activity list and mark it as "abnormal status". 2) Simultaneously activate one standby drone from the backup queue and place it at the front of the standby queue; 3) At the next scheduling time, the backup drone will be dispatched first to seamlessly fill the operational gap left by the faulty drone, ensuring that the continuity and stability of the concrete material flow are not affected.

[0019] Furthermore, the ground control station (GCS) in the collaborative operation system includes: a task intelligence core for executing scheduling algorithms; a human-machine interface configured to receive manual unloading commands from a human pilot; at least two material transport drones, each drone including: an onboard controller; a payload interface; and a remote unloading actuator; wherein the task intelligence core is configured to: calculate a scheduling time interval based on task parameters including single-drone payload; issue task commands sequentially to the drones based on the scheduling time interval; and plan spatiotemporally conflict-free autonomous flight routes for the drones; wherein the onboard controller of each drone is configured to: control the drone to autonomously fly along the autonomous flight route to the unloading area; wherein the ground control station is also configured to: upon receiving the manual unloading command, send an unloading signal to the drone located in the unloading area to trigger the remote unloading actuator in response to the manual command.

[0020] Furthermore, in the aforementioned collaborative operation system, a ground scheduling server operates as a ground control station (GCS). The server includes a processor and a memory, the memory storing computer-executable instructions. When these instructions are executed, the server: calculates a scheduling time interval based on mission parameters including a 40 kg single-unit payload; sequentially issues autonomous flight commands to multiple UAVs to form an "aerial virtual conveyor belt" based on the scheduling time interval; receives unloading commands from a human pilot via a human-machine interface; and responds to the unloading commands by sending an unloading signal to the target UAV.

[0021] Furthermore, in the aforementioned collaborative operation system, a computer-readable storage medium is configured to be stored in the UAV's onboard controller. The medium stores computer-executable instructions, which, when executed, cause the UAV to: (a) receive an autonomous flight path instruction from a ground control station; (b) autonomously control the UAV to fly along the flight path to the unloading area; (c) after arriving at the unloading area, suspend autonomous operation and enter a waiting instruction state; (d) receive a remote unloading signal originating from the ground control station; and (e) in response to the remote unloading signal, drive an unloading actuator to unload.

[0022] Further, the construction method in step S2 includes the following steps: (a) the ground control station assigns a task, which is generated based on task parameters including single-machine load; (b) the UAV takes off and hovers to the location where the material is loaded, and loads the material; (c) the UAV autonomously flies along a set trajectory to the unloading area; (d) when the UAV arrives at the unloading area, it receives an unloading control command from the pilot at the ground control station; (e) in response to the unloading control command, the UAV is controlled to unload the material; (f) the UAV autonomously returns to its home location.

[0023] Furthermore, the heterogeneous UAV formation includes: at least one mapping UAV for on-site 3D mapping, and a transport formation consisting of multiple medium-sized payload UAVs responsible for transporting concrete. All UAVs are uniformly commanded and dispatched by a ground control station that integrates a mission intelligence core.

[0024] Furthermore, the construction method in step S2 includes path planning, and in the path cost calculation, the time-sensitive task window is used as a key constraint to ensure that materials are delivered within the specified time limit.

[0025] Furthermore, in the construction method of step S2, the system always keeps at least one hot backup drone in standby mode. When the drone in transit malfunctions or is delayed, the backup drone can be automatically activated and seamlessly inserted into the transportation sequence to maintain the continuity of material flow.

[0026] The beneficial effects of this invention are: (1) Ensure construction quality and guarantee the performance of the foundation structure. By using the Time Synchronous Continuous Material Flow Scheduling Algorithm (TSS-CMFSA) of the ground control station, a continuous concrete supply is achieved through multiple UAVs acting as an "aerial virtual conveyor belt". This fundamentally avoids the problem of "cold joints" between new and old concrete caused by material interruption, ensuring the integrity and density of the concrete pouring for the tower foundation, thereby guaranteeing the structural strength, durability and long-term bearing capacity of the foundation, and meeting the requirements of the engineering quality standards.

[0027] (2) Improve operational efficiency and shorten the total construction period. This invention transforms dispersed drone formations into a highly efficient and collaborative "aerial conveyor belt," maximizing fleet utilization through precise time synchronization scheduling (schedule intervals matched with individual drone unloading times) and avoiding waiting gaps caused by disordered operations. Compared to traditional ground transportation (relying on temporary roads and manual transfer) and single-drone operations (insufficient payload and poor continuity), this invention can significantly reduce the total time spent on concrete pouring, and is especially suitable for large-scale, time-limited tower foundation pouring tasks.

[0028] (3) Enhance operational safety and reduce construction risks. On the one hand, the entire process adopts an automated operation mode (automatic loading, autonomous flight, and precise unloading), which greatly reduces the high-risk operations of personnel in complex terrains such as mountainous and forest areas (such as cableway transportation and operations around the foundation pit), and reduces personnel safety hazards. On the other hand, the system has a built-in dynamic fault-tolerant management and hot backup mechanism. Even if a single drone fails, a backup drone can seamlessly fill the gap, ensuring that the material flow is not interrupted and avoiding construction accidents caused by pouring stagnation (such as foundation scrapping due to initial concrete setting).

[0029] (4) Reduce overall costs and minimize environmental impact. In terms of cost, there is no need to build temporary roads for equipment access or rent expensive heavy helicopters. Operations can be completed simply by using drone formations, which significantly reduces the initial infrastructure costs and mid-term operating costs of the project. In terms of environment, drone operations do not require the destruction of surface vegetation (no temporary road excavation), and the flight noise is much lower than that of helicopters, which has minimal impact on the ecological environment of the operation area. This meets the development requirements of green infrastructure and ecological protection and is an environmentally friendly construction solution.

[0030] (5) Enhanced scenario adaptability and good scalability. Based on algorithm-driven task planning logic, this invention allows operators to automatically adapt to pouring tasks of different scales (small / large foundations), distances (short / medium distance transportation), and timeframes (short / long initial setting periods) by adjusting core input parameters such as total pouring volume, initial setting time, and UAV performance parameters, without requiring large-scale hardware system modifications. Furthermore, the system's operational capabilities can be further expanded by increasing the number of UAVs or upgrading the payload module, demonstrating broad versatility and scalability, making it suitable for remote areas. Attached Figure Description

[0031] Figure 1 This is a technical roadmap for the collaborative operation system of the present invention.

[0032] Figure 2 This is a diagram of the first type of drone of the present invention.

[0033] Figure 3 This is a diagram of the second type of drone of the present invention.

[0034] Figure 4 This is a diagram illustrating the collaborative operation of multiple unmanned aerial vehicles (UAVs) according to the present invention.

[0035] Figure 5 This invention forms an aerial drone formation diagram. Detailed Implementation

[0036] A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations includes the following steps: S1. Construct a collaborative work system: The technical roadmap for collaborative work systems is as follows: Figure 1 As shown. The collaborative operation system is the core support for achieving continuous pouring, consisting of three core modules: "heterogeneous UAV formation," "ground control station and mission intelligence core," and "intelligent payload interface system." These modules complement each other and work in synergy. (1) Heterogeneous UAV Squadron: Composed of at least two types of UAVs with different functions. The squadron includes at least two types of UAVs with different functions, which cover the entire process of "environmental perception-material transportation" through division of labor and cooperation: "Mapping UAV (Type-S)": Adopting a lightweight and highly mobile design, equipped with LiDAR and high-definition cameras, its core function is to perform centimeter-level precision three-dimensional digital modeling of the pouring point and surrounding environment before operation, providing accurate environmental data support for subsequent path planning and safe operation. "Material Transportation UAV (Type-M)": Designed specifically for medium-load transportation of concrete and other materials (e.g., single-unit load can reach 40 kg), with "high reliability and high transportation efficiency" as the core goal, it has functions such as automated navigation, precise take-off and landing, and remote unloading function controlled by ground station pilots; at the same time, the squadron is equipped with a sufficient number of transportation UAVs (N_available) to meet the material needs of continuous operation and provide redundancy support for fault tolerance backup. Figure 2 and Figure 3 These are two types of drones with different functions. When dealing with heavy, uneven materials, multiple drones can work together to maintain force balance and safely transport the materials, such as... Figure 4 As shown.

[0037] (2) Ground Control Station (GCS) and Mission Intelligence Core: Ground Control Station: The command center of the entire system, providing a human-machine interface for mission definition, process monitoring, emergency intervention, and providing pilots with telemetry data and real-time video feedback for precise unloading. Mission Intelligence Core: The core software system deployed within GCS, serving as the system's "brain." It integrates the core algorithm of this invention—Time Synchronized Continuous Material Flow Scheduling Algorithm (TSS-CMFSA)—with specific functions including: ① 3D Visualization and Planning: Load and display the digital twin model, allowing operators to specify concrete mixing areas and pouring points in a virtual environment.

[0038] ② Task allocation and scheduling: Run a dynamic task allocation framework based on multi-agent deep reinforcement learning (MARL).

[0039] ③ Trajectory planning: Run a constraint-aware collaborative trajectory planning algorithm to generate safe and efficient flight paths for the fleet.

[0040] (3) Intelligent load interface system: This system is a key connecting module for realizing "unmanned operation". The drone is equipped with an automatic loading / unloading module, which can accurately dock with the ground loading device to realize the automated acceptance of materials, improve the operation efficiency and reduce human operation error.

[0041] S2 Construction Method: Based on the aforementioned collaborative operation system, the construction method of this invention follows a closed-loop digital process of "planning-execution-fault tolerance," ensuring that each step meets the technological requirement of "time-limited synchronous pouring" of concrete, specifically divided into three stages: Phase 1: Task Planning and Feasibility Analysis; Before the operation begins, the operator inputs the core parameters of the pouring task at the ground control station, including: total material requirement (V_total), material critical time window (T_critical_min, i.e., initial setting time of concrete), single-unit load (C_drone), total number of available drones (N_available), flight path distance (D_path_m), drone flight speed, loading / unloading time, etc. The task intelligence core conducts feasibility analysis through the TSS-CMFSA algorithm, the process of which is as follows: 1) Calculate the complete cycle time (T_cycle_s) for a single UAV to complete the "loading-flight-pouring-return" process. 2) Verify whether the delivery time of a single material delivery is much shorter than the critical time window to ensure that the concrete remains active during transportation; 3) Optimize the calculation scheduling interval (T_dispatch_interval_s) and set it to be consistent with the single-machine pouring time (t_pour_s) to achieve seamless connection of pouring; 4) Based on the cycle time and scheduling interval, calculate the minimum number of active drones (N_active_req) required to maintain a continuous flow. 5) Verify that the total number of available drones (N_available) meets the total requirement of "number of active drones + number of backup drones (N_standby_req)". The task is deemed "feasible" only if all verification items pass, and can proceed to the next stage.

[0042] Phase Two: Time Synchronization Scheduling and Autonomous Execution; After the task is initiated, the system first divides the UAVs into queues: "Ready Queue," "Standby Queue," and "ActiveList," and then executes the task according to the following logic: Using the "dispatch interval (T_dispatch_interval_s)" calculated in Phase 1 as the precise beat, drones are dispatched sequentially from the standby queue to take off and perform transportation tasks; 2) Each transport drone flies autonomously along a preset route, sequentially going through the state cycle of "loading" → "autonomously flying to the work site" → "waiting for manual unloading instructions" → "remotely unloading" → "autonomously returning to the base".

[0043] 3) Through precise sequential scheduling, multiple drones form an aerial virtual conveyor belt that connects end to end, ensuring that there are always drones continuously pouring above the pouring point, thus avoiding interruption of material supply.

[0044] Phase Three: Dynamic Fault Tolerance and Seamless Replacement; During operation, the system continuously monitors the operational status of all active drones, and immediately activates the fault tolerance mechanism in case of any anomaly. 1) If a drone malfunctions (such as power failure) or deviates significantly from the scheduled time, the system will immediately remove it from the activity list and mark it as "abnormal status". 2) Simultaneously activate one standby drone from the backup queue and place it at the front of the standby queue; 3) At the next scheduling time, the backup drone will be dispatched first to seamlessly fill the operational gap left by the faulty drone, ensuring that the continuity and stability of the concrete material flow are not affected.

[0045] Further (core innovation), the ground control station (GCS) in the collaborative operation system includes: a task intelligence core for executing scheduling algorithms; a human-machine interface configured to receive manual unloading commands from human pilots; at least two material transport drones, each drone including: an onboard controller; a payload interface; and a remote unloading actuator; wherein the task intelligence core is configured to: calculate a scheduling time interval based on task parameters including single-drone payload; issue task commands sequentially to the drones based on the scheduling time interval; and plan spatiotemporally conflict-free autonomous flight routes for the drones; wherein the onboard controller of each drone is configured to: control the drone to autonomously fly along the autonomous flight route to the unloading area; wherein the ground control station is also configured to: upon receiving the manual unloading command, send an unloading signal to the drone located in the unloading area to trigger the remote unloading actuator in response to the manual command.

[0046] Furthermore (the core innovation), in the aforementioned collaborative operation system, a ground scheduling server operates as a ground control station (GCS). The server includes a processor and a memory, in which computer-executable instructions are stored. When these instructions are executed, the server: calculates a scheduling time interval based on mission parameters, including a single-unit payload of 40 kg; sequentially issues autonomous flight commands to multiple UAVs to form an "aerial virtual conveyor belt" based on the scheduling time interval; receives unloading commands from a human pilot via a human-machine interface; and responds to the unloading commands by sending an unloading signal to the target UAV.

[0047] Further (core innovation), in the collaborative operation system, a computer-readable storage medium is configured to be stored in the UAV's onboard controller. The medium stores computer-executable instructions, which, when executed, cause the UAV to: (a) receive autonomous flight path instructions from the ground control station; (b) autonomously control the UAV to fly along the flight path to the unloading area; (c) after arriving at the unloading area, pause autonomous operation and enter a waiting instruction state; (d) receive a remote unloading signal from the ground control station; and (e) respond to the remote unloading signal by driving an unloading actuator to unload.

[0048] Further (core innovation), the construction method in step S2 includes the following steps: (a) the ground control station assigns a task, which is generated based on task parameters including single-machine load; (b) the UAV takes off and hovers to the location where the material is loaded, and loads the material; (c) the UAV autonomously flies along a set trajectory to the unloading area; (d) when the UAV arrives at the unloading area, it receives an unloading control command from the pilot at the ground control station; (e) in response to the unloading control command, the UAV is controlled to unload the material; (f) the UAV autonomously returns to its home location.

[0049] Furthermore, the heterogeneous UAV formation includes: at least one mapping UAV for on-site 3D mapping, and a transport formation consisting of multiple medium-sized payload UAVs responsible for transporting concrete. All UAVs are uniformly commanded and dispatched by a ground control station that integrates a mission intelligence core.

[0050] Furthermore, the construction method in step S2 includes path planning, and in the path cost calculation, the time-sensitive task window is used as a key constraint to ensure that materials are delivered within the specified time limit.

[0051] Furthermore, in the construction method of step S2, the system always keeps at least one hot backup drone in standby mode. When the drone in transit malfunctions or is delayed, the backup drone can be automatically activated and seamlessly inserted into the transportation sequence to maintain the continuity of material flow.

[0052] Working Principle: The working principle of this invention is based on an event-driven multi-agent collaborative framework. Its core objective is to generate and execute a time-synchronized, conflict-free, and highly robust material delivery scheme for the tower foundation pouring task. The entire process (construction method) can be decomposed into three interrelated mathematical models: a dynamic task allocation model, a spatiotemporal trajectory planning model, and a serialized material flow control model.

[0053] 1. Dynamic Task Allocation Model: When a drone is idle, the decision engine will assign it to a task from the pool of pending tasks. Select the optimal task This choice is based on a multi-objective optimization function, the core of which is the computational task. t i For drones d j Overall score S ( d j , t i ): Among them, the comprehensive score function S ( d j , t i ) is defined as: In the formula: Weighting coefficients representing priority Represents task t i The inherent priority score, This represents the timeliness weighting coefficient. Indicates task t i Compared to drones d j Timeliness score This represents the efficiency weighting coefficient. Indicates drone d j With the task t i Model matching score, Indicates drone d j Execute the task t i Distance efficiency score; the specific meanings of each component are as follows: Priority score Refers to the task t i Its inherent priority. Timeliness score: in, It is a task t i The remaining time, Indicates the latest deadline for the task. This represents the current time in the system. This formula ensures that tasks with tighter time constraints receive higher scores. It is a very small positive number ( Typically 0.001-0.01), used to avoid division by zero errors. Model matching score. in, Indicates task t i Required drone type C match , C mismatch These are preset maximum positive and negative numbers used to enforce that only the correct type of drone can perform specific tasks. Efficiency Score: In the formula: Indicates drone d j The current three-dimensional coordinate position, Indicates task t i The starting point coordinates; this score is inversely proportional to the Euclidean distance from the drone's current position to the mission starting point, encouraging the selection of missions closer to the target location to improve efficiency. Weighting coefficients w p , w t , w e The weights corresponding to priority, timeliness, and efficiency are respectively ( w p , w t , w e Specifically, these are 0.1-0.2, 0.5-0.6, and 0.2-0.3, respectively, with the sum of the three being 1. The system possesses adaptive capabilities; if the task... t i If a timeout is detected after execution, the weight will be adjusted. w t This will increase proportionally, making the system prioritize task timeliness in the future. Furthermore, task allocation must satisfy dependency constraints; if tasks... t i There is a set of dependent tasks. D i A task can only be assigned when all dependent tasks have been completed. In the formula: Indicates all dependent tasks t k , Indicates task t i The set of prerequisite dependent tasks, Indicates dependent tasks t k The current execution status, This indicates that the task has been completed.

[0054] 2. Spatiotemporal trajectory planning model: To ensure the safety of multi-aircraft flight, the system adopts a spatiotemporal trajectory planning model. The (SpatiotemporalA*) algorithm is used for trajectory planning. The planning problem is modeled as a four-dimensional spatiotemporal grid. The search problem in which For discrete time dimension, It has three spatial dimensions. The planner's goal is to find a spacetime path starting from the originating point. To the final spacetime point Feasible path This minimizes the path cost while satisfying physical obstacle constraints, i.e., any point on the path... It must be located in free space; among which, These represent the four-dimensional spacetime coordinates of the starting point. These represent the four-dimensional spacetime coordinates of the endpoint. These represent the discrete spatiotemporal node sequences included in the planned path. These represent the first and second parts of the path, respectively. k The four-dimensional spacetime coordinates of each node. Where world is a three-dimensional binary occupancy raster map. Indicates for path P any node in p k P represents the planned, conflict-free flight trajectory. Spatiotemporal conflict constraints refer to any point on the path... The scheduled time slot cannot be booked by other drones. in This is the current spacetime reservation table, a dynamic collection that records all spacetime points occupied by planned paths. Once path P is successfully planned, all its spacetime points will be immediately added to the reservation table. This ensures that subsequent planning will not conflict with it, thus achieving the safety of multi-aircraft coordinated flight.

[0055] 3. Serialized material flow control model For continuous tasks such as pouring concrete for steel tower foundations, the system organizes multiple medium-sized payload drones (Type-M) into an "aerial virtual conveyor belt." Its core is establishing a time-synchronized material flow. Assume the total pouring task is... Q (Unit: kg), the initial setting time of concrete isT deadline (Unit: s), then the required minimum average material flow rate R min for Assume the payload of a single Type-M UAV is q (Unit: kg) In a preferred embodiment of the present invention q The weight is set at 40kg. The time required for a single complete transport cycle (loading-flight-unloading-return) is... C (Unit: s), then the number of drones the system needs to deploy. N At least for This configuration (40kg) is more economical and feasible compared to technically more complex and costly heavy-payload drones (such as 80kg). This also means that to maintain a constant material flow rate, the system will need to manage a larger fleet of drones and execute more frequent cycles. This further highlights the core value of the mission intelligence core and TSS-CMFSA algorithm in managing large-scale, high-density, time-sensitive fleets in this invention.

[0056] To create a continuous material flow, the system employs equally spaced sequential scheduling. The first drone takes off at time t0, and subsequent drones take off at time t... k Determined by the following formula in, This indicates the sequence number of the first drone in the operational formation, and the scheduling interval. for This represents the total number of drones participating in the cyclical operation. Using this scheduling method, the time interval between material arrival at the pouring point remains constant. This ensures the continuity and uniformity of the pouring process, effectively preventing cold joints in the concrete caused by material supply interruptions and ensuring the overall quality of the foundation structure. The system is also equipped with a hot backup drone; if a deviating drone from its predetermined trajectory is detected, the backup drone will be immediately activated and seamlessly inserted into the sequence to maintain the stability of the material flow. Figure 5 As shown, an aerial drone formation is formed to ensure the continuity and uniformity of the construction process.

[0057] Through the synergistic effect of the three mathematical models mentioned above, this invention transforms the complex task of pouring iron tower foundations into a calculable, plannable, and executable automated process, realizing intelligent control of the entire chain from task allocation and safe flight to continuous operation.

Claims

1. A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations, characterized in that... Includes the following steps: S1. Construct a collaborative work system: The collaborative operation system is the core support for achieving continuous pouring. It consists of three core modules: "heterogeneous UAV formation," "ground control station and mission intelligence core," and "intelligent payload interface system." Each module complements the other's functions and works in synergy. (1) Heterogeneous UAV swarm: It consists of at least two types of UAVs with different functions; the swarm includes at least two types of UAVs with different functions, which cover the entire process of "environmental perception-material transportation" through division of labor and cooperation: "Mapping UAV": It adopts a lightweight and highly mobile design, equipped with lidar and high-definition camera. Its core function is to perform centimeter-level three-dimensional digital modeling of the pouring point and the surrounding environment before operation, so as to provide accurate environmental data support for subsequent path planning and safe operation; "Material transportation UAV": It is designed for medium-load transportation of concrete, etc., with "high reliability and high transportation efficiency" as the core goal. It has functions such as automated navigation, precise take-off and landing, and remote unloading function controlled by the ground station pilot; At the same time, the swarm is equipped with a sufficient number of transportation UAVs to meet the material needs of continuous operation and provide redundancy support for fault tolerance backup; (2) Ground control station and mission intelligence core: Ground Control Station: The command center of the entire system, providing a human-machine interface for mission definition, process monitoring, and emergency intervention, as well as providing pilots with telemetry data and real-time video feedback for precise unloading; Mission Intelligence Core: The core software system deployed within GCS, serving as the system's "brain"; it integrates the core algorithm of this invention—the time-synchronized continuous material flow scheduling algorithm, with specific functions including: ① 3D Visualization and Planning: Load and display the digital twin model, allowing operators to specify concrete mixing areas and pouring points in a virtual environment; ② Task allocation and scheduling: Run a dynamic task allocation framework based on multi-agent deep reinforcement learning; ③ Trajectory planning: Run a constraint-aware collaborative trajectory planning algorithm to generate safe and efficient flight paths for the fleet; (3) Intelligent load interface system: This system is a key connecting module for realizing "unmanned operation". The drone is equipped with an automatic loading / unloading module, which can accurately dock with the ground loading device to realize the automated acceptance of materials, improve the operation efficiency, and reduce human operation error. S2 Construction Method: Based on the aforementioned collaborative operation system, the construction method follows a closed-loop digital process of "planning-execution-fault tolerance," ensuring that each step meets the technological requirement of "time-limited synchronous pouring" of concrete. This process is specifically divided into three stages: Phase 1: Task Planning and Feasibility Analysis; Before the operation begins, the operator inputs the core parameters of the pouring task at the ground control station, including: total material requirements, material critical time window, single-machine load, total number of available drones, flight path distance, drone flight speed, loading / unloading time, etc.; The task intelligence core conducts feasibility analysis through the TSS-CMFSA algorithm, the process of which is as follows: 1) Calculate the complete cycle time for a single UAV to complete "loading-flight-pouring-return"; 2) Verify whether the delivery time of a single material delivery is much shorter than the critical time window to ensure that the concrete remains active during transportation; 3) Optimize the calculation and scheduling time interval, setting it to be consistent with the single-machine tilting time to achieve seamless connection of pouring; 4) Based on the cycle time and scheduling interval, calculate the minimum number of active drones required to maintain a continuous flow; 5) Verify that the total number of available drones meets the total requirement of "number of active drones + number of backup drones"; the task is deemed "feasible" only when all verification items pass, and can proceed to the next stage. Phase Two: Time Synchronization Scheduling and Autonomous Execution; After the task is initiated, the system first divides the drones into queues: "Standby Queue," "Backup Queue," and "Activity List," and then executes the task according to the following logic: Using the "scheduling time interval" calculated in Phase 1 as the precise beat, drones are dispatched sequentially from the standby queue to take off and perform transportation tasks; 2) Each transport drone flies autonomously along a preset route, sequentially experiencing the status cycles of "loading", "autonomously flying to the work site", "waiting for manual unloading instructions", "remotely unloading", and "autonomously returning to the base". 3) Through precise sequential scheduling, multiple drones form an aerial virtual conveyor belt that connects end to end, ensuring that there are always drones continuously pouring above the pouring point, thus avoiding interruption of material supply; Phase Three: Dynamic Fault Tolerance and Seamless Replacement; During operation, the system continuously monitors the operational status of all active drones, and immediately activates the fault tolerance mechanism in case of any anomaly. 1) If a drone malfunctions or deviates significantly from the scheduled time, the system will immediately remove it from the activity list and mark it as "abnormal status"; 2) Simultaneously activate one standby drone from the backup queue and place it at the front of the standby queue; 3) At the next scheduling time, the backup drone will be dispatched first to seamlessly fill the operational gap left by the faulty drone, ensuring that the continuity and stability of the concrete material flow are not affected.

2. The method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... The ground control station in the collaborative operation system includes: a task intelligence core for executing scheduling algorithms; a human-machine interface configured to receive manual unloading commands from human pilots; and at least two material transport drones, each drone including: an onboard controller; a payload interface; and a remote unloading actuator. The task intelligence core is configured to: calculate a scheduling time interval based on task parameters, including single-drone payload; sequentially issue task commands to the drones based on the scheduling time interval; and plan spatiotemporally conflict-free autonomous flight routes for the drones. The onboard controller of each drone is configured to: control the drone to autonomously fly along the autonomous flight route to the unloading area. The ground control station is also configured to: upon receiving the manual unloading command, send an unloading signal to the drone located in the unloading area to trigger the remote unloading actuator in response to the manual command.

3. The method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... In the aforementioned collaborative operation system, a ground dispatch server operates as a ground control station. The server includes a processor and a memory. The memory stores computer-executable instructions. When the instructions are executed, the server calculates a dispatch time interval based on task parameters, including a single-unit payload of 40 kg. The system sequentially issues autonomous flight commands to multiple drones to form an "aerial virtual conveyor belt" based on the scheduling time interval; it receives unloading commands from human pilots via the human-machine interface; and in response to the unloading commands, it sends unloading signals to the target drone.

4. The method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... In the aforementioned collaborative operation system, a computer-readable storage medium is configured to be stored in an onboard controller of a UAV, the medium storing computer-executable instructions, which, when executed, cause the UAV to: (a) receive an autonomous flight path instruction from a ground control station; (b) autonomously control the UAV to fly along the flight path to the unloading area; (c) after arriving at the unloading area, suspend autonomous operation and enter a waiting instruction state; (d) receive a remote unloading signal originating from the ground control station; and (e) in response to the remote unloading signal, drive an unloading actuator to unload material.

5. A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... The construction method in step S2 includes the following steps: (a) the ground control station assigns a task, which is generated based on task parameters including single-machine load; (b) the UAV takes off and hovers at the location where the material is loaded, and loads the material; (c) the UAV autonomously flies along a set trajectory to the unloading area; (d) when the UAV arrives at the unloading area, it receives an unloading control command from the pilot at the ground control station; (e) in response to the unloading control command, the UAV is controlled to unload the material; (f) the UAV autonomously returns to its home base.

6. A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... The heterogeneous UAV formation includes: at least one surveying UAV for on-site 3D mapping, and a transport formation consisting of multiple medium-sized payload UAVs responsible for transporting concrete. All UAVs are uniformly commanded and dispatched by a ground control station that integrates a mission intelligence core.

7. A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... The construction method in step S2 includes path planning. In the path cost calculation, the time-sensitive task window is used as a key constraint to ensure that materials are delivered within the specified time limit.

8. A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... In the construction method of step S2, the system always keeps at least one hot backup drone in standby mode. When the drone in transit malfunctions or is delayed, the backup drone can be automatically activated and seamlessly inserted into the transportation sequence to maintain the continuity of material flow.

9. A method for collaborative operation and construction of heavy-duty UAVs for pouring iron tower foundations according to claim 1, characterized in that... It is decomposed into three interrelated mathematical models: dynamic task allocation model, spatiotemporal trajectory planning model, and serialized material flow control model; 1) Dynamic task allocation model: When a drone is idle, the decision engine will assign it from the pool of pending tasks. Select the optimal task ; This choice is based on a multi-objective optimization function, the core of which is the computational task. t i For drones d j Overall score S ( d j , t i ): Among them, the comprehensive score function S ( d j , t i ) is defined as: In the formula: Weighting coefficients representing priority Represents task t i The inherent priority score, This represents the timeliness weighting coefficient. Indicates task t i Compared to drones d j Timeliness score This represents the efficiency weighting coefficient. Indicates drone d j With the task t i Model matching score, Indicates drone d j Execute the task t i Distance efficiency score; the specific meanings of each component are as follows; priority score Refers to the task t i The inherent priority; the timeliness score is in, It is a task t i The remaining time, Indicates the latest deadline for the task. This indicates the current time in the system; this formula ensures that tasks with more pressing time requirements receive higher scores. It is a very small positive number. Typically 0.001-0.01, used to avoid division by zero errors; Model matching score: in, Indicates task t i Required drone type C match , C mismatch These are preset maximum positive and negative numbers used to enforce that only the correct type of drone can perform specific tasks; Efficiency Score: In the formula: Indicates drone d j The current three-dimensional coordinate position, Indicates task t i The starting point coordinates; this score is inversely proportional to the Euclidean distance from the drone's current position to the mission starting point, encouraging the selection of missions closer to the target location to improve efficiency; weighting coefficients. w p , w t , w e These correspond to the weights of priority, timeliness, and efficiency, respectively. w p , w t , w e Specifically, these are 0.1-0.2, 0.5-0.6, and 0.2-0.3, respectively, and the sum of these three items is 1; the system has adaptive capabilities, if the task... t i If a timeout is detected after execution, the weight will be adjusted. w t The proportional growth will make the system place greater emphasis on task timeliness in the future; furthermore, task allocation must satisfy dependency constraints, if tasks... t i There is a set of dependent tasks. D i A task can only be assigned when all dependent tasks have been completed. In the formula: Indicates all dependent tasks t k , Indicates task t i The set of prerequisite dependent tasks, Indicates dependent tasks t k The current execution status, This indicates that the task has been completed. 2) Spatiotemporal trajectory planning model: To ensure the safety of multi-aircraft flight, the system adopts a spatiotemporal trajectory planning model. The algorithm performs trajectory planning; the planning problem is modeled in a four-dimensional spacetime grid. The search problem in which For discrete time dimension, It has three spatial dimensions; the planner's goal is to find a spatiotemporal path starting from the starting point. To the final spacetime point Feasible path This minimizes the path cost while satisfying physical obstacle constraints, i.e., any point on the path... It must be located in free space; among which, These represent the four-dimensional spacetime coordinates of the starting point. These represent the four-dimensional spacetime coordinates of the endpoint. These represent the discrete spatiotemporal node sequences included in the planned path. These represent the first and second parts of the path, respectively. k The four-dimensional spatiotemporal coordinates of each node; Where world is a three-dimensional binary occupancy raster map. Indicates for path P any node in p k P represents the planned, conflict-free flight trajectory; spatiotemporal conflict constraints refer to any point on the path. The scheduled time slot cannot be booked by other drones; in This is the current spatiotemporal reservation table, a dynamic set that records all spatiotemporal points occupied by planned paths. Once path P is successfully planned, all its spatiotemporal points will be immediately added to the reservation table. This ensures that subsequent planning will not conflict with it, thus achieving the safety of multi-aircraft cooperative flight; 3) Serialized material flow control model For continuous tasks such as pouring concrete for tower foundations, the system organizes multiple medium-sized payload drones into an "aerial virtual conveyor belt"; its core is to establish a time-synchronized material flow; assuming the total pouring task is... Q Unit: kg; Initial setting time of concrete is T deadline If the unit is seconds, then the required minimum average material flow rate is... R min for Assume the payload of a single Type-M UAV is q Unit: kg. In a preferred embodiment of the present invention, q The weight is set at 40kg; the time required for a single complete transport cycle is C The unit is seconds (s), which indicates the number of drones the system needs to deploy. N At least: This 40kg configuration is more economical and feasible compared to heavy-load drones, which are more technically complex and costly. This also means that in order to maintain a constant material flow rate, the system will need to schedule a larger fleet of drones and execute more frequent cycles. This further highlights the core value of the task intelligence core and TSS-CMFSA algorithm in managing large-scale, high-density, and time-sensitive fleets in this invention. To create a continuous material flow, the system employs equally spaced sequential scheduling; the first drone takes off at time t0, and subsequent drones take off at time t... k Determined by the following formula in, This indicates the sequence number of the first drone in the operational formation, and the scheduling interval. for This represents the total number of drones participating in the cyclical operation. Using this scheduling method, the time interval between material arrival at the pouring point remains constant. This ensures the continuity and uniformity of the pouring process, effectively avoids cold joints in concrete caused by material supply interruptions, and ensures the overall quality of the foundation structure.

Images