A three-dimensional path planning method for unmanned aerial vehicle wireless charging

Abstract

The application provides a three-dimensional path planning method for unmanned aerial vehicle wireless charging, and belongs to the technical field of unmanned aerial vehicle control. The method comprises the following steps: S1, constructing a system model; S2, problem modeling, establishing a remaining energy maximization problem; S3, problem optimization; and adjusting a sampling strategy according to the energy state of the current unmanned aerial vehicle, that is, applying an energy-aware EA-RRT algorithm to path planning. The application combines SWIPT wireless energy transmission and fixed charging stations to construct a hybrid charging system, which significantly improves the endurance of the unmanned aerial vehicle in a complex urban environment. The energy-aware EA-RRT algorithm can dynamically plan a path, preferentially guide to a charging area when the power is insufficient, and efficiently fly to a target when the power is sufficient, thereby realizing autonomous energy management. Meanwhile, the algorithm takes into account the turning angle, obstacle avoidance and task coverage efficiency, and ensures flight safety and task quality.

Description

Technical Field

[0001] This invention provides a three-dimensional path planning method for wireless charging of unmanned aerial vehicles (UAVs), belonging to the field of UAV control technology. Background Technology

[0002] In recent years, drone technology has developed rapidly, and its high mobility and convenient deployment have led to its widespread application in logistics, monitoring, and communication. However, the limitations of drones in dynamic obstacle avoidance and their limited endurance restrict their ability to operate continuously in complex scenarios.

[0003] To improve the dynamic obstacle avoidance performance of UAVs, various improved path planning algorithms have been proposed, such as the Fast Search Random Tree algorithm and the Ant Colony algorithm. By comparing the performance of algorithms like EA-RRT, Astar, and ACO in 3D multi-obstacle environments, the performance of these methods in terms of path quality and planning efficiency has been explored. However, these methods primarily focus on enhancing path planning capabilities and have not effectively addressed the energy consumption problem of UAVs.

[0004] With the introduction of Wireless Powered Communication (SWIPT) technology, unmanned aerial vehicles (UAVs) have gained new potential in the parallel processing of energy harvesting and information transmission. Related research has improved the energy efficiency of the system by optimizing SWIPT power allocation and UAV trajectory, or by combining non-orthogonal multiple access technology and smart reflectors. However, the energy harvested solely by SWIPT technology is still insufficient to meet the continuous operational needs of UAVs. Summary of the Invention

[0005] This invention provides a three-dimensional path planning method for wireless charging of drones, focusing on the energy-aware EA-RRT algorithm to integrate SWIPT technology with fixed wireless charging stations, thereby improving the stability of energy supply. An optimization model is constructed to maximize the remaining energy of the drone.

[0006] A three-dimensional path planning method for wireless charging of drones includes the following steps:

[0007] S1. Constructing the System Model

[0008] The system model consists of two parts: a wireless charging model and a trajectory model. In a three-dimensional (3D) urban environment, the UAV navigates through tasks involving building obstacles and numerous wireless charging stations. The UAV communicates with base stations (BS) while simultaneously harvesting energy during this interaction using SWIPT technology. The location of the base stations is... When a UAV is located near a Wireless Charging Station (WCS), it acquires energy via wireless power transfer. A collection of Wireless Charging Stations (WCS) is called a The total mission duration is divided into T time slots, with each time slot lasting for a duration of... In time slot t, the position coordinates of the UAV are defined as follows: The drone's three consecutive waypoints are:

[0009]

[0010] Their horizontal projection vectors are as follows:

[0011]

[0012] Maximum turning angle constraint in horizontal plane for:

[0013]

[0014] in It is the maximum turning angle;

[0015] Command track segment length Must meet:

[0016]

[0017] in Indicates the shortest distance;

[0018] Define a reconnaissance efficiency metric ,Right now:

[0019]

[0020] in It is the power received by the unmanned aerial vehicle from the base station in time slot t. This refers to the energy collected by the drone in the current time slot;

[0021] The constraint for priority coverage is:

[0022]

[0023] in, and These are the initial reconnaissance efficiency and the minimum reconnaissance efficiency, respectively. It is in time The standard deviation of the signal strength, and It is the gain coefficient;

[0024] Channel power gain from base station to drone Represented as:

[0025]

[0026] in, This is the channel power gain at a reference distance of 1 meter;

[0027] When the unmanned aerial vehicle (UAV) communicates with the base station, the UAV is in a time slot. The power received from the base station is:

[0028]

[0029] in This indicates the base station's transmission power;

[0030] Using a nonlinear energy harvesting circuit, the harvested energy is expressed as:

[0031] in, This is the upper limit of the saturation value of the energy harvesting circuit. and These represent the circuit sensitivity parameter and the circuit offset parameter, respectively.

[0032] In each time slot Proportional time is used for information transmission. If time is used for energy harvesting, then the energy harvested by the drone in a certain time slot is denoted as:

[0033]

[0034] Among them, the information transmission duration needs to meet the following requirements. ;

[0035] The speed of the drone is V, and its flight power consumption is expressed as:

[0036]

[0037] in, These are all constant parameters, which are related to the weight of the drone, wing area, and air density.

[0038] The drone's transmission power is In two cycles Between consecutive time slots, the flight energy consumption and transmission energy consumption of the UAV are respectively Employing a hybrid energy management architecture, it integrates dynamic energy harvesting from fixed WCS and SWIPT technologies; when the drone hovers above the wireless charging station, the energy used for charging is:

[0039]

[0040] in, It is a binary variable; This indicates that the drone is charging from WCS k, otherwise ; For WCS charging efficiency, This refers to the transmit power of WCS. The proportion of time slots spent by the drone at the charging station;

[0041] By applying a hybrid wireless power charging model, drones can perform time-slot operations. The total energy collected during the period was The remaining energy of the UAV in the time slot Updated to:

[0042]

[0043] in, It refers to the remaining energy of the drone in two consecutive time slots. This refers to the energy consumed by the drone during time slot t.

[0044] S2. Problem Modeling

[0045] Establish the problem of maximizing residual energy:

[0046]

[0047] Constraint C3 requires that the length of the UAV's trajectory in each time slot should be within a minimum length. and maximum length Between; constraint C4 indicates that the UAV should reach its destination. Constraint C5 means that the UAV can connect to at most one WCS per time slot, and C6 means... It is a binary variable, where C7 indicates that the drone's remaining energy should be greater than or equal to its minimum energy. And not greater than the maximum energy C8 represents the time allocation factor.

[0048] S3. Problem Optimization

[0049] The sampling strategy is adjusted based on the current energy state of the UAV, i.e., the Energy Awareness-Responsive Tracking (EA-RRT) algorithm is applied for path planning. Its core objective function is:

[0050]

[0051] in, These represent the weights of energy gain, goal orientation, turning angle constraint, and task coverage efficiency, respectively. The path cost is determined by the distance from the current node to the goal, and the constraint penalty term is used to regulate turning angle over-limit behavior. This involves incorporating energy gains into the node selection strategy and defining a bimodal energy gain function:

[0052]

[0053] Energy gain function It consists of charging station revenue and SWIPT revenue. Charging station revenue is determined by an indicator function. Combining charging efficiency, transmit power, and time, the SWIPT benefit is determined by the energy harvesting model, the channel gain from the base station to the UAV, and the base station transmit power.

[0054] Time allocation factor These are dynamic parameters for real-time adaptive adjustment of the energy state based on the UAV, defined as follows:

[0055]

[0056] in, Ensure that the drone always reserves a minimum amount of time for energy harvesting, while maintaining sufficient time for communication tasks;

[0057] Combined with a probabilistic model of the remaining energy of the drone:

[0058]

[0059] in, Indicates the nearest charging station A Gaussian distribution centered on this region is favorable for energy replenishment during low-energy states, while Maintain global exploration with sufficient energy;

[0060] For the current position node New nodes are generated through the guiding rules. :

[0061]

[0062] in It dynamically adjusts the step size based on the local obstacle density. The nodes are randomly generated;

[0063] The optimal location node was selected as:

[0064]

[0065] At the same time, in order to bypass obstacles in the 3D urban environment, a collision-free approach is chosen. Then calculate the remaining energy.

[0066] This invention constructs a hybrid charging system by combining SWIPT wireless power transfer with fixed charging stations, significantly improving the endurance of UAVs in complex urban environments. The proposed Energy Awareness-Based Relay (EA-RRT) algorithm dynamically plans paths, prioritizing navigation to charging areas when battery is low and efficiently flying to the target when battery is sufficient, achieving autonomous energy management. Simultaneously, the algorithm considers turning angle, obstacle avoidance, and mission coverage efficiency, ensuring flight safety and mission quality. Simulations show that this scheme outperforms traditional algorithms in both path length and net energy consumption, providing a more reliable and efficient autonomous flight solution for long-duration tasks such as logistics and inspection. Attached Figure Description

[0067] Figure 1 This is a schematic diagram of the system model of the present invention;

[0068] Figure 2 The simulation map is for the example implementation;

[0069] Figure 3 A planar comparison of the trajectories of drones deployed at two charging stations, as shown in the example.

[0070] Figure 4 A 3D comparison of the drone trajectories for deploying two charging stations, as an example. Detailed Implementation

[0071] The specific technical solutions of the present invention will be described with reference to the embodiments.

[0072] A three-dimensional path planning method for wireless charging of drones includes the following steps:

[0073] S1. Constructing the System Model

[0074] First, this invention describes the system model under consideration, which includes two parts: a wireless charging model and a trajectory model. For example... Figure 1 As shown, in a three-dimensional (3D) urban environment, UAV navigation traverses tasks involving building obstacles and numerous wireless charging stations. The UAV communicates with base stations (BS) while simultaneously harvesting energy during this interaction using SWIPT technology. The location of the base stations is... When a UAV is located near a Wireless Charging Station (WCS), it acquires energy via wireless power transfer. A collection of Wireless Charging Stations (WCS) is called a The total mission duration is divided into T time slots, with each time slot lasting for a duration of... In time slot t, the position coordinates of the UAV are defined as follows: The drone's three consecutive waypoints are:

[0075]

[0076] Their horizontal projection vectors are as follows:

[0077]

[0078] Maximum turning angle constraint in horizontal plane for:

[0079]

[0080] in It is the maximum turning angle;

[0081] The minimum flight path length constraint requires that before changing flight attitude or direction, the UAV must continuously fly along the current flight path for at least a specified short distance to prevent energy loss and control instability caused by frequent maneuvers. length Must meet:

[0082]

[0083] in Indicates the shortest distance;

[0084] Meanwhile, in order to evaluate the effective mission area covered or the key information acquired per unit path length, a reconnaissance efficiency metric is defined. ,Right now:

[0085]

[0086] in It is the power received by the unmanned aerial vehicle from the base station in time slot t. This refers to the energy collected by the drone in the current time slot;

[0087] When planning their trajectories, drones must balance path detours with mission benefits and avoid ineffective patrols. For example, prioritizing areas with high signal strength within obstacles is necessary. Therefore, the constraint of priority coverage is:

[0088]

[0089] in, and These are the initial reconnaissance efficiency and the minimum reconnaissance efficiency, respectively. It is in time The standard deviation of the signal strength, and It is the gain coefficient;

[0090] Channel power gain from base station to drone Represented as:

[0091]

[0092] in, This is the channel power gain at a reference distance of 1 meter;

[0093] When the unmanned aerial vehicle (UAV) communicates with the base station, the UAV is in a time slot. The power received from the base station is:

[0094]

[0095] in This indicates the base station's transmission power;

[0096] Using a nonlinear energy harvesting circuit, the harvested energy is expressed as:

[0097] in, This is the upper limit of the saturation value of the energy harvesting circuit. and These represent the circuit sensitivity parameter and the circuit offset parameter, respectively.

[0098] In each time slot Proportional time is used for information transmission. If time is used for energy harvesting, then the energy harvested by the drone in a certain time slot is denoted as:

[0099]

[0100] Among them, the information transmission duration needs to meet the following requirements. ;

[0101] The speed of the drone is V, and its flight power consumption is expressed as:

[0102]

[0103] in, These are all constant parameters, which are related to the weight of the drone, wing area, and air density.

[0104] The drone's transmission power is In two cycles Between consecutive time slots, the flight energy consumption and transmission energy consumption of the UAV are respectively Employing a hybrid energy management architecture, it integrates dynamic energy harvesting from fixed WCS and SWIPT technologies; when the drone hovers above the wireless charging station, the energy used for charging is:

[0105]

[0106] in, It is a binary variable; This indicates that the drone is charging from WCS k, otherwise ; For WCS charging efficiency, This refers to the transmit power of WCS. The proportion of time slots spent by the drone at the charging station;

[0107] By applying a hybrid wireless power charging model, drones can perform time-slot operations. The total energy collected during the period was The remaining energy of the UAV in the time slot Updated to:

[0108]

[0109] in, It refers to the remaining energy of the drone in two consecutive time slots. This refers to the energy consumed by the drone during time slot t.

[0110] S2. Problem Modeling

[0111] Next, this invention patent establishes the problem of maximizing remaining energy. To maximize the remaining energy of the UAV in the final time slot using the energy-sensing trajectory, the remaining energy maximization problem is established:

[0112]

[0113] Constraint C3 requires that the length of the UAV's trajectory in each time slot should be within a minimum length. and maximum length Between; constraint C4 indicates that the UAV should reach its destination. Constraint C5 means that the UAV can connect to at most one WCS per time slot, and C6 means... It is a binary variable, where C7 indicates that the drone's remaining energy should be greater than or equal to its minimum energy. And not greater than the maximum energy C8 represents the time allocation factor.

[0114] S3. Problem Optimization

[0115] In optimization problems that maximize remaining energy, complex constraints make it difficult to directly and efficiently solve for path point coordinates, time allocation variables, turning angles, segment lengths, and non-convexity using traditional optimization methods. This invention adjusts the sampling strategy based on the current energy state of the UAV, applying the Energy Awareness-Responsive Tracking (EA-RRT) algorithm for path planning. Its core objective function is:

[0116]

[0117] in, These represent the weights of energy gain, goal orientation, turning angle constraint, and task coverage efficiency, respectively. The path cost is determined by the distance from the current node to the goal, and the constraint penalty term is used to regulate turning angle over-limit behavior. This involves incorporating energy gains into the node selection strategy and defining a bimodal energy gain function:

[0118]

[0119] Energy gain function It consists of charging station revenue and SWIPT revenue. Charging station revenue is determined by an indicator function. Combining charging efficiency, transmit power, and time, the SWIPT benefit is determined by the energy harvesting model, the channel gain from the base station to the UAV, and the base station transmit power.

[0120] Time allocation factor These are dynamic parameters for real-time adaptive adjustment of the energy state based on the UAV, defined as follows:

[0121]

[0122] in, Ensure that the drone always reserves a minimum amount of time for energy harvesting, while maintaining sufficient time for communication tasks;

[0123] The core of EA-RRT lies in its energy-dependent sampling strategy. Unlike traditional RRT, which relies entirely on random sampling, this method incorporates a probabilistic model that adapts to the UAV's remaining energy.

[0124]

[0125] in, Indicates the nearest charging station A Gaussian distribution centered on this region is favorable for energy replenishment during low-energy states, while Maintain global exploration with sufficient energy;

[0126] For the current position node New nodes are generated through the guiding rules. :

[0127]

[0128] in It dynamically adjusts the step size based on the local obstacle density. The nodes are randomly generated;

[0129] The optimal location node was selected as:

[0130]

[0131] At the same time, in order to bypass obstacles in the 3D urban environment, a collision-free approach is chosen. Then calculate the remaining energy.

[0132] Finally, this embodiment verifies through simulation analysis that the proposed trajectory optimization algorithm achieves the optimal trajectory in the energy harvesting process compared to other algorithms.

[0133] A 3D map is generated using a campus setting. To evaluate the performance of the proposed EA-RRT algorithm, simulations were performed in a complex 3D urban environment. The drone's task is to navigate from a starting point to a destination, while avoiding buildings and utilizing wireless charging stations. This invention considers two charging stations. , , , , , , , , .

[0134] The proposed EA-RRT is evaluated against two reference algorithms: the Astar algorithm, known for its heuristic approach to shortest path planning, and the Ant Colony Optimization (ACO) algorithm, which simulates ants foraging to solve path planning challenges. Figure 2 As shown, two wireless charging stations are deployed based on a 3D map. The drone needs to traverse the building from the starting point to the destination, passing through the charging stations along the way.

[0135] The path curve of the drone is obtained through simulation, such as Figure 3 and Figure 4 As shown in the figure, comparing the simulation data, the EA-RRT algorithm exhibits the best performance, with the highest search efficiency (0.0162 seconds) and the shortest path length (1185.5 meters). In terms of energy consumption, EA-RRT (12262.83 J) is significantly lower than ACO (12614.80 J) and Astar (37278.34 J); its harvested energy (3548.41 J) is on par with ACO and far superior to Astar. EA-RRT excels in real-time performance, path quality, and energy efficiency, making it suitable for energy-sensitive scenarios; ACO is the second-best choice; and Astar is not recommended due to its high energy consumption.

Claims

1. A three-dimensional path planning method for wireless charging of unmanned aerial vehicles, characterized in that, Includes the following steps: S1. Constructing the System Model The system model consists of two parts: a wireless charging model and a trajectory model; UAV navigation traverses an environment involving building obstacles and multiple wireless charging stations; the UAV communicates with the base station (BS) while simultaneously harvesting energy during communication using SWIPT technology; By applying a hybrid wireless power charging model, drones can perform time-slot operations. The total energy collected during the period was The remaining energy of the UAV in the time slot Updated to: ， in, and It refers to the remaining energy of the drone in two consecutive time slots. This refers to the energy consumed by the drone during time slot t. The specific steps for constructing the system model are as follows: The location of the base station is When a UAV is located near a Wireless Charging Station (WCS), it acquires energy through wireless power transfer; the collection of Wireless Charging Stations (WCS) is called... ; The total mission duration is divided into T time slots, with each time slot lasting for a certain duration. In the time slot Define the drone's position coordinates as The drone's three consecutive waypoints are: ， Their horizontal projection vectors are as follows: ， Maximum turning angle constraint in horizontal plane for: ， in It is the maximum turning angle; Command track segment length Must meet: ， in Indicates the shortest distance; Define a reconnaissance efficiency metric ,Right now: ， in It is the power received by the unmanned aerial vehicle from the base station in time slot t. This refers to the energy collected by the drone in the current time slot; The constraint for priority coverage is: ， in, and These are the initial reconnaissance efficiency and the minimum reconnaissance efficiency, respectively. It is in time The standard deviation of the signal strength, and It is the gain coefficient; Channel power gain from base station to drone Represented as: ， in, This is the channel power gain at a reference distance of 1 meter; When the unmanned aerial vehicle (UAV) communicates with the base station, the UAV is in a time slot. The power received from the base station is: ， in This indicates the base station's transmission power; Using a nonlinear energy harvesting circuit, the harvested energy is expressed as: ， in, This is the upper limit of the saturation value of the energy harvesting circuit. and These represent the circuit sensitivity parameter and the circuit offset parameter, respectively. In each time slot Proportional time is used for information transmission. If time is used for energy harvesting, then the energy harvested by the drone in a certain time slot is denoted as: ， Among them, the information transmission duration needs to meet the following requirements. ; The speed of the drone is V, and its flight power consumption is expressed as: ， in, These are all constant parameters, which are related to the weight of the drone, wing area, and air density. The drone's transmission power is In two cycles Between consecutive time slots, the flight energy consumption and transmission energy consumption of the UAV are respectively Employing a hybrid energy management architecture, it integrates dynamic energy harvesting from fixed WCS and SWIPT technologies; when the drone hovers above the wireless charging station, the energy used for charging is: ， in, It is a binary variable; This indicates that the drone is charging from WCS k, otherwise ; For WCS charging efficiency, This refers to the transmit power of WCS. The proportion of time slots spent by the drone at the charging station; By applying a hybrid wireless power charging model, drones can perform time-slot operations. The total energy collected during the period was The remaining energy of the UAV in the time slot Updated to: ， in, It refers to the remaining energy of the drone in two consecutive time slots. This refers to the energy consumed by the drone during time slot t. S2. Problem Modeling Establish a model for the problem of maximizing residual energy; S3. Problem Optimization Adjust the sampling strategy based on the current energy state of the UAV, that is, apply the Energy Awareness-Responsive Tracking (EA-RRT) algorithm for path planning; The core objective function is: ， in, These represent the weights of energy gain, goal orientation, turning angle constraint, and task coverage efficiency, respectively. The path cost is determined by the distance from the current node to the goal, and the constraint penalty term is used to regulate turning angle over-limit behavior. This involves incorporating energy gains into the node selection strategy and defining a bimodal energy gain function: ， Energy gain function It consists of charging station revenue and SWIPT revenue. Charging station revenue is determined by an indicator function. Combining charging efficiency, transmit power, and time, the SWIPT benefit is determined by the energy harvesting model, the channel gain from the base station to the UAV, and the base station transmit power. Time allocation factor These are dynamic parameters for real-time adaptive adjustment of the energy state based on the UAV, defined as follows: ， in, Ensure that the drone always reserves a minimum amount of time for energy harvesting, while maintaining sufficient time for communication tasks; Combined with a probabilistic model of the remaining energy of the drone: ， in, Indicates the nearest charging station A Gaussian distribution centered on this region is favorable for energy replenishment during low-energy states, while Maintain global exploration with sufficient energy; For the current position node New nodes are generated through the guiding rules. : ， in It dynamically adjusts the step size based on the local obstacle density. The nodes are randomly generated; The optimal location node was selected as: ， At the same time, in order to bypass obstacles in the 3D urban environment, a collision-free approach is chosen. Then calculate the remaining energy.

2. The three-dimensional path planning method for wireless charging of unmanned aerial vehicles according to claim 1, characterized in that, S2. The model for maximizing residual energy is established as follows: Constraint C3 requires that the length of the UAV's trajectory in each time slot should be within a minimum length. and maximum length Between; constraint C4 indicates that the UAV should reach its destination. ; Constraint C5 means that the UAV can connect to at most one WCS per time slot, and C6 means... It is a binary variable, where C7 indicates that the drone's remaining energy should be greater than or equal to its minimum energy. And not greater than the maximum energy C8 represents the time allocation factor.

Images