Unmanned aerial vehicle low-altitude air route dynamic planning method and device, equipment and medium

Abstract

The application relates to a low-altitude air route dynamic planning method, device, equipment and medium for a UAV. The method comprises the following steps: acquiring real-time dynamic environment change data and obstacle distribution information of a low-altitude flight area, constructing a preliminary data set, and generating an initial flight path; according to the initial flight path, energy consumption data is predicted, if the energy consumption data exceeds a threshold value, optimization is performed to obtain an optimized path; risk assessment quantification is performed, a safety margin adjustment mechanism is combined to obtain a safety index, if the safety index is lower than a threshold value, a safety distance is adjusted to generate a safety path; flight conditions are monitored to generate a condition monitoring update report, new obstacle position information is acquired, and a dynamic obstacle avoidance path is generated; and according to the dynamic obstacle avoidance path, an execution instruction set of the UAV is generated. The method can improve the real-time adaptability of the dynamic environment of the UAV, and can plan a route that meets the task requirements in time when facing sudden obstacles or environmental changes.

Description

Technical Field

[0001] This invention belongs to the field of low-altitude flight path planning for unmanned aerial vehicles (UAVs), and in particular relates to methods, devices, equipment and media for dynamic planning of low-altitude flight paths for UAVs. Background Technology

[0002] With the rapid expansion of urban air traffic and drone applications, dynamic planning technology for low-altitude drone routes has emerged. This not only affects the implementation effectiveness in practical scenarios such as logistics delivery and emergency rescue, but also has profound significance for the construction of future air traffic management systems.

[0003] Traditional methods for low-altitude flight path planning are mainly based on static environment assumptions and use a single-dimensional optimization method to plan the low-altitude flight path of UAVs. The optimization objective is to minimize the flight distance, and the planned flight path is the shortest path connecting the starting point and the destination under static conditions.

[0004] However, traditional methods have the following problems: they are difficult to adapt to the dynamic conditions that change in real time during low-altitude flight, and they ignore the combined effects of multiple factors such as time, energy and risk. As a result, when faced with sudden obstacles or environmental changes, the planning results lack flexibility and safety, and are difficult to meet the actual mission requirements. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, device, equipment, and medium for dynamic planning of low-altitude flight paths for unmanned aerial vehicles (UAVs) to address the aforementioned technical problems.

[0006] Firstly, this application provides a dynamic planning method for low-altitude flight paths of unmanned aerial vehicles (UAVs), including:

[0007] Acquire real-time dynamic environmental change data and obstacle distribution information of the low-altitude flight area, construct a preliminary dataset, and generate an initial flight path based on the preliminary dataset in combination with time constraints;

[0008] Based on the initial flight path, predict energy consumption data. If the energy consumption data exceeds the threshold, optimize the initial flight path to obtain an optimized path with reduced energy consumption.

[0009] The optimized path is risk-quantified and combined with a safety margin adjustment mechanism to obtain a safety index. If the safety index is lower than the threshold, the optimized path is adjusted for a safe distance to generate a safe path.

[0010] Monitor flight conditions for safe paths, generate condition monitoring update reports, and obtain new obstacle location information based on condition monitoring update reports to generate dynamic obstacle avoidance paths;

[0011] Based on the dynamic obstacle avoidance path, an execution command set for the UAV is generated; the execution command set is used to instruct the UAV to fly in accordance with the dynamic obstacle avoidance path.

[0012] In one embodiment, after generating the execution instruction set, the method further includes:

[0013] When the drone is flying on a dynamic obstacle avoidance path, it acquires real-time environmental feedback data;

[0014] Key parameters are extracted from the dynamic obstacle avoidance path, and these key parameters are processed to obtain a structured path parameter set;

[0015] Real-time environmental feedback data and structured path parameter sets are input into a preset model for matching analysis to obtain a model deviation report;

[0016] Based on the model deviation report and the preset optimization level rules, a local adjustment path is generated.

[0017] In one embodiment, real-time environmental feedback data and a structured path parameter set are input into a preset model for matching analysis to obtain a model deviation report, including:

[0018] Dynamic environmental parameters are extracted from real-time environmental feedback data and input into a preset UAV dynamics model to obtain theoretical linear acceleration and theoretical angular velocity.

[0019] The theoretical linear acceleration and theoretical angular velocity are compared with the actual linear acceleration and actual angular velocity in the real-time environmental feedback data, and the residual values ​​of the actual linear acceleration and actual angular velocity are calculated.

[0020] The total energy consumption of the dynamic obstacle avoidance path is predicted based on the preset energy consumption model, and compared with the real-time battery consumption data in the structured path parameter set to calculate the energy consumption deviation rate.

[0021] Based on the location and speed of new obstacles in the structured path parameter set, a risk score is calculated and compared with a preset risk level table to generate a risk increment.

[0022] A model deviation report is generated based on the residual values ​​of actual linear acceleration, actual angular velocity, energy consumption deviation rate, and risk increment.

[0023] In one embodiment, an initial flight path is generated based on a preliminary dataset, including:

[0024] Based on the preliminary dataset, the dynamic flight path parameters are generated using the following formula:

[0025]

[0026]

[0027] in, For dynamic flight path parameters, For the initial dataset, Let be the total cost function. For obstacle avoidance weights, The current three-dimensional position coordinates of the drone. For the first The three-dimensional position coordinates of the obstacle As energy consumption weight, Let be the instantaneous energy consumption rate function of the UAV;

[0028] Based on dynamic flight path parameters, using A * The algorithm generates multiple candidate paths;

[0029] Based on the candidate paths, the corresponding total path cost is generated using the following formula:

[0030]

[0031]

[0032]

[0033] in, The total cost of the path. For the actual cost, For the remaining cost, This is the timeout penalty coefficient. The current node time For portions exceeding the deadline, This represents the Euclidean distance between adjacent nodes. Distance weights The time increment to reach the current node. As time weight, Distance to the nearest obstacle Risk weighting Weighted by minimum remaining time. The current three-dimensional position coordinates, The target's three-dimensional position coordinates, For maximum speed, Weighted by remaining time;

[0034] By comparing the total path costs, the candidate path with the minimum total path cost is determined as the initial flight path.

[0035] In one embodiment, based on the initial flight path, energy consumption data is predicted. If the energy consumption data exceeds a threshold, the initial flight path is optimized to obtain an optimized path with reduced energy consumption, including:

[0036] Based on the initial flight path, predict energy consumption data using the following formula:

[0037]

[0038]

[0039] in, This is a forecast of energy consumption. For total power, Let velocity be the function. For acceleration function, For height function, Reserve power for emergencies. Based on power consumption, To assist in power consumption, For motor efficiency, air density, The drag coefficient, For windward area, For flight speed, For the quality of drones, It is the acceleration due to gravity. For the climb angle, For acceleration;

[0040] If the energy consumption data exceeds the threshold, construct an energy consumption heat map of the initial flight path, and determine the high-energy-consuming flight segments based on the energy consumption heat map;

[0041] The high-energy-consuming flight segments are analyzed to obtain energy consumption analysis results, and optimized path parameters are generated based on the energy consumption analysis results;

[0042] Based on the optimized path parameters, the initial flight path is adjusted to generate an optimized path with reduced energy consumption.

[0043] In one embodiment, the optimized path undergoes a risk assessment and quantification. Combined with a safety margin adjustment mechanism, a safety index is obtained. If the safety index is below a threshold, the optimized path is adjusted for a safe distance to generate a safe path, including:

[0044] Extract key node data from the optimized path and obtain multi-source threat data;

[0045] By correlating and analyzing key node data and multi-source threat data, node risk parameters are obtained.

[0046] Based on the node risk parameters and a preset risk level table, the node risk value is generated using the following formula:

[0047]

[0048] in, The node risk value. Let i be the weight of the i-th type of threat. For the risk level of the i-th type of threat, For environmental risk coefficient, For environmental risks;

[0049] Based on the node risk value and the safety margin adjustment mechanism, the safety index is calculated.

[0050] If the security index is below the threshold, the safe distance of the optimized path is optimized based on the node risk value to generate a safe path.

[0051] In one embodiment, the formula for calculating the security index is:

[0052]

[0053]

[0054]

[0055]

[0056] in, For safety indicators, Minimum obstacle distance The adjusted safe distance, To adjust the remaining energy, To adjust the time tolerance, Energy consumption threshold The deadline for the task. The maximum node risk value. The original safe distance. For expansion coefficient, The safety indicators from the previous moment. As the energy buffer coefficient, This represents the time relaxation factor.

[0057] Secondly, this application also provides a device for dynamic planning of low-altitude flight paths for unmanned aerial vehicles, comprising:

[0058] The initial flight path generation module is used to acquire real-time dynamic environmental change data and obstacle distribution information of the low-altitude flight area, construct a preliminary dataset, and generate an initial flight path based on the preliminary dataset in combination with time constraints.

[0059] The energy consumption data prediction module is used to predict energy consumption data based on the initial flight path. If the energy consumption data exceeds the threshold, the initial flight path is optimized to obtain an optimized path that reduces energy consumption.

[0060] The safety index generation module is used to quantify the risk assessment of the optimized path and, combined with the safety margin adjustment mechanism, obtain the safety index. If the safety index is lower than the threshold, the safety distance of the optimized path is adjusted to generate a safe path.

[0061] The dynamic obstacle avoidance path generation module is used to monitor the flight conditions of a safe path, generate a condition monitoring update report, and obtain new obstacle location information based on the condition monitoring update report to generate a dynamic obstacle avoidance path.

[0062] The execution instruction set generation module is used to generate an execution instruction set for the UAV based on the dynamic obstacle avoidance path; the execution instruction set is used to instruct the UAV to fly in accordance with the dynamic obstacle avoidance path.

[0063] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the UAV low-altitude flight path dynamic planning method of the first aspect of this application.

[0064] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any step of the UAV low-altitude flight path dynamic planning method of the first aspect of this application.

[0065] The technical solution provided in this application includes at least the following beneficial effects:

[0066] A preliminary dataset is constructed by real-time collection of dynamic environmental change data and obstacle distribution information in low-altitude flight areas. An initial flight path is generated based on time constraints. Energy consumption data is then predicted based on the initial flight path. If energy consumption exceeds a threshold, the initial flight path is optimized to obtain an energy-efficient path. A multi-dimensional risk assessment is then performed on the optimized path, and a safety margin adjustment mechanism is used to calculate a safety index. If the safety index is below the threshold, a safety distance compensation correction is applied to the optimized path to generate a safe path. The safe path is then monitored under different flight conditions, generating a condition monitoring update report, acquiring the location information of new obstacles, and generating a dynamic obstacle avoidance path. Finally, the execution command set for the UAV is generated based on the generated dynamic obstacle avoidance path. This method improves the real-time adaptability of the UAV to the dynamic environment, enabling timely planning of flight routes that meet mission requirements when facing sudden obstacles or environmental changes, thereby improving the UAV's flight efficiency. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 A schematic diagram of an implementation environment provided for an exemplary embodiment of this application;

[0069] Figure 2 A flowchart illustrating a dynamic planning method for low-altitude flight paths of an unmanned aerial vehicle (UAV) provided as an exemplary embodiment of this application;

[0070] Figure 3 This is a schematic diagram of the structure of a dynamic planning device for low-altitude flight paths of an unmanned aerial vehicle (UAV) provided as an exemplary embodiment of this application. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0072] The UAV low-altitude flight path dynamic planning method provided in this application embodiment can be applied to, for example... Figure 1 In the implementation environment shown, terminal 101 communicates with drone 102 via a network. Terminal 101 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Drone 102 can be, but is not limited to, various types of drones.

[0073] The application scenarios of the embodiments of this application will be described in conjunction with the above implementation environment.

[0074] As an illustration, the UAV low-altitude flight path dynamic planning method provided in this application embodiment includes, but is not limited to, at least one of the following scenarios.

[0075] First, the dynamic planning method for low-altitude flight paths of UAVs can be applied to agricultural plant protection. For example, when UAVs are spraying pesticides, they can generate an initial path based on farmland terrain and crop distribution, optimize the path by combining the energy consumption model of pesticide spraying volume and flight speed, and monitor the position of field personnel and wind direction changes in real time to dynamically adjust the safety distance and generate obstacle avoidance paths, thereby improving the safety and accuracy of UAV operations.

[0076] Secondly, this dynamic planning method for low-altitude flight paths of UAVs can also be applied to the inspection of power and communication infrastructure. For example, when UAVs fly along power transmission lines, they can monitor bird activity and weather conditions in real time to prevent threats to the flight safety of UAVs, while dynamically adjusting the inspection route to reduce the risk of damage to UAVs.

[0077] Third, this dynamic planning method for low-altitude flight paths of UAVs can also be applied to emergency rescue and disaster site reconnaissance. For example, by using real-time remote sensing data of earthquake-stricken areas to construct a dynamic obstacle distribution model, an initial path to the disaster area in the shortest time can be generated. At the same time, the risk of obstacle movement caused by aftershocks and the interference of dense smoke areas on UAV sensors can be dynamically assessed, and the safety margin of the path can be adjusted to ensure the completion of the flight mission.

[0078] In one exemplary embodiment, such as Figure 2 As shown, a dynamic planning method for low-altitude flight paths of unmanned aerial vehicles (UAVs) is provided, which is then applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps:

[0079] S201: Obtain real-time dynamic environmental change data and obstacle distribution information of the low-altitude flight area, construct a preliminary dataset, and generate an initial flight path based on the preliminary dataset in combination with time constraints.

[0080] Real-time dynamic environmental change data refers to multi-dimensional environmental parameters collected in real time during UAV flight, including meteorological data, terrain changes, moving obstacle information, and airspace control instructions. Obstacle distribution information refers to the spatial distribution data of objects with relatively fixed positions within the low-altitude flight area. The preliminary dataset refers to a structured data set after spatiotemporal alignment and gridding of the real-time dynamic environmental change data and obstacle distribution information.

[0081] For example, the terminal obtains real-time dynamic environmental change data and obstacle distribution information of the low-altitude flight area where the UAV is located from the UAV's lidar, Doppler weather instrument, and UWB (Ultra-Wideband Positioning) positioning. It then performs spatiotemporal alignment and dynamic meshing processing on the real-time dynamic environmental change data and obstacle distribution information to construct a preliminary dataset. Finally, based on the time constraints of the flight mission, it improves the A... * The algorithm generates an initial flight path that takes into account the shortest path, optimal timeliness, and lowest risk value.

[0082] S202: Based on the initial flight path, predict energy consumption data. If the energy consumption data exceeds a threshold, optimize the initial flight path to obtain an optimized path with reduced energy consumption data.

[0083] For example, the terminal predicts the energy consumption data of the UAV flying along the initial flight path based on the segment parameters (speed, acceleration, altitude) of the initial flight path using a high-precision energy consumption model. If the energy consumption data exceeds the threshold, gradient descent is used to optimize and adjust the path parameters of the initial flight path to generate an optimized path that reduces the UAV's energy consumption data.

[0084] S203. The optimized path is subjected to risk assessment and quantification. Combined with the safety margin adjustment mechanism, a safety index is obtained. If the safety index is lower than the threshold, the safety distance of the optimized path is adjusted to generate a safe path.

[0085] Among them, the safety margin adjustment mechanism refers to an algorithm that dynamically expands the safety protection boundary based on real-time risks. The security index refers to a numerical value that quantifies the overall security of a path.

[0086] For example, the terminal performs node risk assessment on the optimized path to obtain node risk value, and then calculates the security index by combining the security margin adjustment mechanism. If the security index is lower than the threshold (0.75), the safe distance of the optimized path is adjusted and expanded, and the adjusted optimized path is processed by B-spline curve to generate a safe path.

[0087] S204 monitors flight conditions for a safe path, generates a condition monitoring update report, and obtains new obstacle location information based on the condition monitoring update report to generate a dynamic obstacle avoidance path.

[0088] Among them, the Condition Monitoring Update Report refers to the structured data of the flight environment of the safe path collected in real time by the drone's sensors, including environmental dynamic parameters, drone status, and information on new obstacles.

[0089] For example, the terminal monitors flight conditions in the safe path in real time using the drone's lidar, Doppler weather instrument, and UWB positioning, including obstacle dynamics, meteorological parameters, and drone status. It integrates the monitored flight conditions to generate a condition monitoring update report, then obtains the location information of new obstacles from the report, and finally, based on this information, improves the A... * The algorithm and B-spline curve smoothing process generate dynamic obstacle avoidance paths.

[0090] S205, Generate an execution instruction set for the UAV based on the dynamic obstacle avoidance path; the execution instruction set is used to instruct the UAV to fly according to the dynamic obstacle avoidance path.

[0091] For example, the terminal discretizes the dynamic obstacle avoidance path into a series of waypoints and generates an execution command set based on the UAV's state. This execution command set includes the target position, desired speed, heading angle, altitude, etc., and is updated at regular intervals. For example, the execution command set is used to instruct the UAV to fly according to the generated dynamic obstacle avoidance path.

[0092] This application provides a method for dynamic planning of low-altitude flight paths for unmanned aerial vehicles (UAVs). It constructs a preliminary dataset by real-time collection of dynamic environmental change data and obstacle distribution information in the low-altitude flight area. An initial flight path is generated based on time constraints. Energy consumption data is then predicted based on the initial flight path. If the energy consumption exceeds a threshold, the initial flight path is optimized to obtain an optimized path with reduced energy consumption. A multi-dimensional risk assessment is then performed on the optimized path, and a safety margin adjustment mechanism is used to calculate a safety index. If the safety index is below the threshold, a safety distance compensation correction is applied to the optimized path to generate a safe path. The method then monitors the flight conditions along the safe path, generates a condition monitoring update report, acquires the location information of new obstacles, and generates a dynamic obstacle avoidance path. Finally, the execution command set for the UAV is generated based on the generated dynamic obstacle avoidance path. This method improves the real-time adaptability of the UAV to the dynamic environment, enabling timely planning of flight paths that meet mission requirements when facing sudden obstacles or environmental changes, thereby improving the UAV's flight efficiency.

[0093] In one embodiment, after generating the execution instruction set, the method further includes:

[0094] S301: When the drone is flying on a dynamic obstacle avoidance path, it acquires real-time environmental feedback data.

[0095] S302, extract key parameters from the dynamic obstacle avoidance path, and process the key parameters to obtain a structured path parameter set;

[0096] S303: Input real-time environmental feedback data and structured path parameter set into the preset model for matching analysis to obtain a model deviation report;

[0097] S304. Based on the model deviation report and the preset optimization level rules, a local adjustment path is generated.

[0098] Real-time environmental feedback data refers to the dynamic environmental and status parameters collected in real time by the UAV during its flight along a dynamic obstacle avoidance path, including actual motion status, changes in the external environment (location of new obstacles, wind speed, etc.), and energy consumption. Key parameters refer to the core path features extracted from the dynamic obstacle avoidance path, including key waypoint coordinates, node safety distances, and risk values. The model deviation report refers to the quantitative result of the difference between the real-time environmental feedback data and the structured path parameter set and the predicted values ​​in the preset model.

[0099] For example, when a UAV flies along a dynamic obstacle avoidance path, the terminal acquires real-time environmental feedback data from the UAV's sensors and extracts key parameters such as key waypoint coordinates, node safety distances, and risk values ​​from the dynamic obstacle avoidance path. These key parameters are then processed through spatiotemporal coding to generate a structured path parameter set. The real-time environmental feedback data and the structured path parameter set are then input into a preset model for matching analysis, generating a model deviation report. Finally, based on preset optimization level rules and the model deviation report, gradient descent is used to adjust local path parameters, generating a locally adjusted path. This embodiment dynamically optimizes the local path through real-time environmental feedback and model deviation analysis, improving the obstacle avoidance efficiency of the UAV in dynamic obstacle environments.

[0100] In one embodiment, real-time environmental feedback data and a structured path parameter set are input into a preset model for matching analysis to obtain a model deviation report, including:

[0101] S401 extracts dynamic environmental parameters from real-time environmental feedback data and inputs them into a preset UAV dynamics model to obtain theoretical linear acceleration and theoretical angular velocity;

[0102] S402 compares the theoretical linear acceleration, theoretical angular velocity, and actual linear acceleration and actual angular velocity in the real-time environmental feedback data, and calculates the residual values ​​of the actual linear acceleration and actual angular velocity.

[0103] S403 predicts the total energy consumption of the dynamic obstacle avoidance path based on the preset energy consumption model, and compares it with the real-time battery consumption data in the structured path parameter set to calculate the energy consumption deviation rate.

[0104] S404: Calculate the risk score based on the location and speed of new obstacles in the structured path parameter set, compare it with the preset risk level table, and generate risk increments.

[0105] S405 generates a model deviation report based on the residual values ​​of actual linear acceleration, actual angular velocity, energy consumption deviation rate, and risk increment.

[0106] Dynamic environmental parameters refer to external disturbances that change in real time during UAV flight and affect motion control, including wind speed and relative speed to obstacles. The UAV dynamics model is a mathematical model that predicts the ideal motion state of the UAV. It can be used to calculate the theoretical linear acceleration and theoretical angular velocity of the UAV during flight, and also to calculate the linear acceleration residuals and angular velocity residuals of the actual linear acceleration and actual angular velocity from the theoretical linear acceleration, theoretical angular velocity, and real-time environmental feedback data.

[0107] For example, the terminal extracts dynamic environmental parameters from real-time environmental feedback data, inputs them into a preset UAV dynamics model for processing and calculation, and outputs the theoretical linear acceleration and theoretical angular velocity of the UAV. Then, it compares the theoretical linear acceleration and theoretical angular velocity with the actual linear acceleration and actual angular velocity from the real-time environmental feedback data to calculate the linear acceleration residual and angular velocity residual. Next, it predicts the total energy consumption required for the UAV to fly along a dynamic obstacle avoidance path using a preset energy consumption model, and compares the total energy consumption with real-time battery consumption data to calculate the energy consumption deviation rate. Subsequently, it extracts the position and velocity of new obstacles from the structured path parameter set, calculates a real-time risk score, and generates a risk increment by comparing it with a preset risk level table. Finally, it fuses the residual values ​​of actual linear acceleration, actual angular velocity, energy consumption deviation rate, and risk increment to generate a model deviation report containing four-dimensional quantitative indicators. This embodiment achieves real-time monitoring of the UAV's dynamic state, energy efficiency, and environmental risk through joint quantitative analysis of multi-dimensional motion residuals and risk increments, enhancing the accuracy of obstacle avoidance decisions in complex dynamic environments.

[0108] In one embodiment, an initial flight path is generated based on a preliminary dataset, including:

[0109] S501, based on the preliminary dataset, uses the following formula to generate dynamic flight path parameters:

[0110]

[0111]

[0112] in, For dynamic flight path parameters, For the initial dataset, Let be the total cost function. For obstacle avoidance weights, The current three-dimensional position coordinates of the drone. For the first The three-dimensional position coordinates of the obstacle As energy consumption weight, Let be the instantaneous energy consumption rate function of the UAV;

[0113] S502, based on dynamic flight path parameters, utilizes A * The algorithm generates multiple candidate paths;

[0114] S503, based on the candidate paths, use the following formula to generate the corresponding total path cost:

[0115]

[0116]

[0117]

[0118] in, The total cost of the path. For the actual cost, For the remaining cost, This is the timeout penalty coefficient. The current node time For portions exceeding the deadline, This represents the Euclidean distance between adjacent nodes. Distance weights The time increment to reach the current node. As time weight, Distance to the nearest obstacle Risk weighting Weighted by minimum remaining time. The current three-dimensional position coordinates, The target's three-dimensional position coordinates, For maximum speed, Weighted by remaining time;

[0119] S504. By comparing the total path cost, the candidate path with the minimum total path cost is determined as the initial flight path.

[0120] Among them, dynamic flight path parameters refer to those obtained through the cost function. Generate path feature quantization vectors to guide the generation of UAV flight paths. Candidate paths refer to those generated through A. * The algorithm searches for a set of feasible paths in a three-dimensional space (50m×50m×10m grid). The total path cost is a comprehensive quantitative indicator for evaluating the quality of a path.

[0121] For example, based on the initial dataset, the terminal generates dynamic flight path parameters through a cost function. These parameters include a four-dimensional vector of velocity, acceleration, heading angle, and obstacle distance. This is then used in a three-dimensional A... * The algorithm performs an extended search within a 50m×50m×10m network space (divided according to the low-altitude flight area), retains 5 candidate paths with the lowest cost, calculates the total cost of these candidate paths, compares and ranks the calculated total costs, and selects the candidate path with the lowest total cost as the initial flight path. This embodiment uses dynamic flight path parameter calculation and A... * The algorithm enables efficient, safe, and mission-compliant flight path planning through path search and multi-objective cost optimization.

[0122] In one embodiment, based on the initial flight path, energy consumption data is predicted. If the energy consumption data exceeds a threshold, the initial flight path is optimized to obtain an optimized path with reduced energy consumption, including:

[0123] S601, based on the initial flight path, uses the following formula to predict energy consumption data:

[0124]

[0125]

[0126] in, This is a forecast of energy consumption. For total power, Let velocity be the function. For acceleration function, For height function, Reserve power for emergencies. Based on power consumption, To assist in power consumption, For motor efficiency, air density, The drag coefficient, For windward area, For flight speed, For the quality of drones, It is the acceleration due to gravity. For the climb angle, For acceleration;

[0127] S602, if the energy consumption data exceeds the threshold, construct an energy consumption heat map of the initial flight path, and determine the high-energy-consumption flight segment based on the energy consumption heat map;

[0128] S603 analyzes high-energy-consuming flight segments, obtains energy consumption analysis results, and generates optimized path parameters based on the energy consumption analysis results;

[0129] S604, Based on the optimized path parameters, the initial flight path is adjusted to generate an optimized path with reduced energy consumption data.

[0130] Among them, the energy consumption heat map is a data visualization model that uses color gradients to intuitively display the energy consumption intensity of a flight segment in a three-dimensional grid space.

[0131] For example, the terminal calculates the total power segmented by formula based on the generated initial flight path, and then predicts the energy consumption data using an energy consumption prediction model. If the energy consumption data exceeds a threshold, an energy consumption heatmap of the initial flight path is constructed in a three-dimensional grid space, and high-energy-consuming segments are marked in red. The causes of high energy consumption in these segments are analyzed to obtain energy consumption analysis results. Gradient descent optimization is then applied to adjust the climb angle and reduce the flight speed, generating optimized path parameters. Combined with time constraints, the initial flight path is then adjusted and planned based on these optimized path parameters to generate an optimized path with reduced energy consumption. This embodiment accurately locates high-energy-consuming segments using an energy consumption heatmap, and adjusts the initial flight path using gradient descent optimization and time constraints to reduce path energy consumption while ensuring mission timeliness.

[0132] In one embodiment, the optimized path undergoes a risk assessment and quantification. Combined with a safety margin adjustment mechanism, a safety index is obtained. If the safety index is below a threshold, the optimized path is adjusted for a safe distance to generate a safe path, including:

[0133] S701 extracts key node data from the optimized path and obtains multi-source threat data;

[0134] S702 performs correlation analysis on key node data and multi-source threat data to obtain node risk parameters;

[0135] S703, based on the node risk parameters and a preset risk level table, the node risk value is generated using the following formula:

[0136]

[0137] in, The node risk value. Let i be the weight of the i-th type of threat. For the risk level of the i-th type of threat, For environmental risk coefficient, For environmental risks;

[0138] S704 calculates the security index based on the node risk value and the safety margin adjustment mechanism;

[0139] S705: If the security index is lower than the threshold, the optimized path will be optimized for a safe distance based on the node risk value to generate a safe path.

[0140] For example, the formula for calculating the security index is as follows:

[0141]

[0142]

[0143]

[0144]

[0145] in, For safety indicators, Minimum obstacle distance The adjusted safe distance, To adjust the remaining energy, To adjust the time tolerance, Energy consumption threshold The deadline for the task. The maximum node risk value. The original safe distance. For expansion coefficient, The safety indicators from the previous moment. As the energy buffer coefficient, This represents the time relaxation factor.

[0146] Key node data refers to the coordinates of turning points and high-risk areas in the optimized path, including path curvature and altitude. Multi-source threat data refers to the diverse dynamic risk sources faced by the UAV during flight, including weather turbulence and moving obstacles. Node risk parameters are intermediate variables that quantify the intensity of a single type of threat at a node.

[0147] For example, the terminal extracts key node data from the optimized path and collects multi-source threat data at the key nodes in real time through the drone's sensors. By performing correlation analysis on the key node data and multi-source threat data, node risk parameters are generated. Then, combined with a preset risk level table, a node risk value is calculated using a formula. If the node risk value is greater than 5, the corresponding node is marked as high-risk. A safety index is calculated using a safety margin adjustment mechanism. If the safety index is lower than a threshold, the safety distance for high-risk nodes is extended based on the node risk value. Smoothing is then performed using a B-spline curve to generate a safe path. This embodiment improves the obstacle avoidance capability of drones in complex environments by quantitatively assessing path risk using key node data and multi-source threat data, and optimizing the path based on the node risk value.

[0148] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0149] Based on the same inventive concept, this application also provides a device for dynamic planning of low-altitude flight paths for unmanned aerial vehicles (UAVs) to implement the methods described above. The solution provided by this device is similar to the solution described in the methods above. Therefore, the specific limitations of one or more embodiments of the dynamic planning device for low-altitude flight paths for UAVs provided below can be found in the limitations of the dynamic planning method for low-altitude flight paths for UAVs described above, and will not be repeated here.

[0150] In one exemplary embodiment, such as Figure 3 As shown, a dynamic planning device 800 for low-altitude flight paths of unmanned aerial vehicles (UAVs) is provided, comprising:

[0151] The initial flight path generation module 801 is used to acquire real-time dynamic environmental change data and obstacle distribution information of the low-altitude flight area, construct a preliminary dataset, and generate an initial flight path based on the preliminary dataset in combination with time constraints.

[0152] The energy consumption data prediction module 802 is used to predict energy consumption data based on the initial flight path. If the energy consumption data exceeds the threshold, the initial flight path is optimized to obtain an optimized path with reduced energy consumption data.

[0153] The safety index generation module 803 is used to quantify the risk assessment of the optimized path and obtain the safety index by combining the safety margin adjustment mechanism. If the safety index is lower than the threshold, the safety distance of the optimized path is adjusted to generate a safe path.

[0154] The dynamic obstacle avoidance path generation module 804 is used to monitor the flight conditions of the safe path, generate a condition monitoring update report, and obtain new obstacle location information based on the condition monitoring update report to generate a dynamic obstacle avoidance path.

[0155] The execution instruction set generation module 805 is used to generate an execution instruction set for the UAV based on the dynamic obstacle avoidance path; the execution instruction set is used to instruct the UAV to fly in accordance with the dynamic obstacle avoidance path.

[0156] In one embodiment, the device further includes:

[0157] The real-time environmental feedback data acquisition module is used to acquire real-time environmental feedback data when the UAV is flying on a dynamic obstacle avoidance path.

[0158] The key parameter extraction module is used to extract key parameters from the dynamic obstacle avoidance path and process the key parameters to obtain a structured path parameter set.

[0159] The model deviation report generation module is used to input real-time environmental feedback data and structured path parameter sets into a preset model for matching analysis and to obtain a model deviation report.

[0160] The local adjustment path generation module is used to generate local adjustment paths based on the model deviation report and preset optimization level rules.

[0161] In one embodiment, the model bias report generation module includes:

[0162] The dynamic environment parameter extraction unit is used to extract dynamic environment parameters from real-time environmental feedback data and input them into a preset UAV dynamics model to obtain theoretical linear acceleration and theoretical angular velocity.

[0163] The residual value calculation unit is used to compare the theoretical linear acceleration, theoretical angular velocity and the actual linear acceleration and actual angular velocity in the real-time environmental feedback data, and calculate the residual values ​​of the actual linear acceleration and the actual angular velocity.

[0164] The energy consumption deviation rate calculation unit is used to predict the total energy consumption of the dynamic obstacle avoidance path according to the preset energy consumption model, and compare it with the real-time battery consumption data in the structured path parameter set to calculate the energy consumption deviation rate.

[0165] The risk increment generation unit is used to calculate the risk score based on the location and speed of new obstacles in the structured path parameter set, and compare it with the preset risk level table to generate risk increments.

[0166] The model deviation report generation unit is used to generate a model deviation report based on the residual values ​​of actual linear acceleration, actual angular velocity, energy consumption deviation rate, and risk increment.

[0167] In one embodiment, the initial flight path generation module includes:

[0168] The dynamic flight path parameter generation unit is used to generate dynamic flight path parameters based on the preliminary dataset using the following formula:

[0169]

[0170]

[0171] in, For dynamic flight path parameters, For the initial dataset, Let be the total cost function. For obstacle avoidance weights, The current three-dimensional position coordinates of the drone. For the first The three-dimensional position coordinates of the obstacle As energy consumption weight, Let be the instantaneous energy consumption rate function of the UAV;

[0172] The candidate path generation unit is used to generate a candidate path based on dynamic flight path parameters using A. * The algorithm generates multiple candidate paths;

[0173] The total path cost generation unit is used to generate the corresponding total path cost based on the candidate paths using the following formula:

[0174]

[0175]

[0176]

[0177] in, The total cost of the path. For the actual cost, For the remaining cost, This is the timeout penalty coefficient. The current node time For portions exceeding the deadline, This represents the Euclidean distance between adjacent nodes. Distance weights The time increment to reach the current node. As time weight, Distance to the nearest obstacle Risk weighting Weighted by minimum remaining time. The current three-dimensional position coordinates, The target's three-dimensional position coordinates, For maximum speed, Weighted by remaining time;

[0178] The initial flight path determination unit is used to compare the total path cost and determine the candidate path with the minimum total path cost as the initial flight path.

[0179] In one embodiment, the energy consumption data prediction module includes:

[0180] The energy consumption data prediction unit is used to predict energy consumption data based on the initial flight path using the following formula:

[0181]

[0182]

[0183] in, This is a forecast of energy consumption. For total power, Let velocity be the function. For acceleration function, For height function, Reserve power for emergencies. Based on power consumption, To assist in power consumption, For motor efficiency, air density, The drag coefficient, For windward area, For flight speed, For the quality of drones, It is the acceleration due to gravity. For the climb angle, For acceleration;

[0184] The high-energy-consumption flight segment determination unit is used to construct an energy consumption heat map of the initial flight path if the energy consumption data exceeds a threshold, and to determine the high-energy-consumption flight segment based on the energy consumption heat map.

[0185] The optimized path parameter generation unit is used to analyze high-energy-consuming flight segments, obtain energy consumption analysis results, and generate optimized path parameters based on the energy consumption analysis results.

[0186] The optimized path generation unit is used to adjust the initial flight path according to the optimized path parameters to generate an optimized path with reduced energy consumption data.

[0187] In one embodiment, the security indicator generation module includes:

[0188] The critical node data extraction unit is used to extract critical node data from the optimized path and obtain multi-source threat data;

[0189] The node risk parameter generation unit is used to perform correlation analysis between key node data and multi-source threat data to obtain node risk parameters;

[0190] The node risk value generation unit is used to generate node risk values ​​based on node risk parameters and a preset risk level table, using the following formula:

[0191]

[0192] in, The node risk value. Let i be the weight of the i-th type of threat. For the risk level of the i-th type of threat, For environmental risk coefficient, For environmental risks;

[0193] The security index calculation unit is used to calculate the security index based on the node risk value and the security margin adjustment mechanism.

[0194] The safe path generation unit is used to optimize the safe distance of the optimized path based on the node risk value if the safety index is lower than the threshold, and generate a safe path.

[0195] In one embodiment, the security index calculation unit includes

[0196] The safety index calculation subunit is used to calculate the safety index. The calculation formula for the safety index is as follows:

[0197]

[0198]

[0199]

[0200]

[0201] in, For safety indicators, Minimum obstacle distance The adjusted safe distance, To adjust the remaining energy, To adjust the time tolerance, Energy consumption threshold The deadline for the task. The maximum node risk value. The original safe distance. For expansion coefficient, The safety indicators from the previous moment. As the energy buffer coefficient, This represents the time relaxation factor.

[0202] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the above-described UAV low-altitude flight path dynamic planning method.

[0203] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0204] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0205] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A dynamic planning method for unmanned aerial vehicle low-altitude air route, characterized in that, The method comprises: acquiring real-time dynamic environment change data and obstacle distribution information of a low-altitude flight area, constructing a preliminary data set, and generating an initial flight path based on the preliminary data set combined with time constraints; predicting energy consumption data according to the initial flight path, and if the energy consumption data exceeds a threshold, optimizing the initial flight path to obtain an optimized path with reduced energy consumption data; quantifying risk assessment of the optimized path, obtaining a safety index combined with a safety margin adjustment mechanism, and if the safety index is lower than a threshold, adjusting the safety distance of the optimized path to generate a safe path; monitoring the flight conditions of the safe path, generating a condition monitoring update report, and acquiring new obstacle position information according to the condition monitoring update report to generate a dynamic obstacle avoidance path; generating an execution instruction set for the UAV according to the dynamic obstacle avoidance path; the execution instruction set is used to instruct the UAV to fly according to the dynamic obstacle avoidance path; wherein the generation of the initial flight path based on the preliminary data set comprises: generating dynamic flight path parameters according to the preliminary data set using the following formula: ； ； wherein, is a dynamic flight path parameter, is a preliminary data set, is a total cost function, is an obstacle avoidance weight, is a current three-dimensional position coordinate of the UAV, is a three-dimensional position coordinate of the th obstacle, is an energy consumption weight, is an instantaneous energy consumption rate function of the UAV; According to the dynamic flight path parameters, a plurality of candidate paths are generated using an A * algorithm. generating the corresponding path total cost according to the candidate path using the following formula: ； ； ； wherein, is the path total cost, is the actual cost, is the remaining cost, is the timeout penalty coefficient, is the current node time part exceeding the deadline, is the Euclidean distance between adjacent nodes, is the distance weight, is the time increment to reach the current node, is the time weight, is the distance to the nearest obstacle, is the risk weight, is the minimum remaining time weight, is the current three-dimensional position coordinate, is the target three-dimensional position coordinate, is the maximum speed, is the remaining time weight; comparing the path total costs to determine the candidate path with the minimum path total cost as the initial flight path.

2. The method of claim 1, wherein, After generating the execution instruction set for the UAV, the method further comprises: acquiring real-time environment feedback data when the UAV is flying on the dynamic obstacle avoidance path; extracting key parameters from the dynamic obstacle avoidance path and processing the key parameters to obtain a structured path parameter set; inputting the real-time environment feedback data and the structured path parameter set into a preset model for matching analysis to obtain a model deviation report; generating a local adjustment path according to the model deviation report combined with a preset optimization level rule.

3. The method of claim 2, wherein, The inputting of the real-time environment feedback data and the structured path parameter set into a preset model for matching analysis to obtain a model deviation report comprises: extracting dynamic environment parameters from the real-time environment feedback data and inputting them into a preset UAV dynamics model to obtain theoretical linear acceleration and theoretical angular velocity; comparing the theoretical linear acceleration, the theoretical angular velocity, and the actual linear acceleration and actual angular velocity in the real-time environment feedback data to calculate residual values of the actual linear acceleration and the actual angular velocity; predicting the total energy consumption of the dynamic obstacle avoidance path according to a preset energy consumption model and comparing it with real-time battery consumption data in the structured path parameter set to calculate an energy consumption deviation rate; calculating a risk score according to new obstacle positions and speeds in the structured path parameter set and comparing it with a preset risk level table to generate a risk increment; generating the model deviation report according to the residual values of the actual linear acceleration and the actual angular velocity, the energy consumption deviation rate, and the risk increment.

4. The method of claim 3, wherein, The risk assessment of the optimized path is quantified, a safety margin adjustment mechanism is combined to obtain a safety index, and if the safety index is lower than a threshold value, a safety distance adjustment is performed on the optimized path to generate a safety path, including: Key node data is extracted from the optimized path, and multi-source threat data is obtained; The key node data and the multi-source threat data are associated and analyzed to obtain node risk parameters; According to the node risk parameters, in combination with the preset risk level table, the following formula is used to generate node risk values: ； wherein, is a node risk value, is a weight of the ith threat, is a risk level of the ith threat, is an environmental risk coefficient, is an environmental risk; According to the node risk values, in combination with the safety margin adjustment mechanism, the safety index is calculated; If the safety index is lower than a threshold value, the optimized path is optimized according to the node risk values to generate the safety path.

5. The method of claim 4, wherein, The calculation formula of the safety index is: ； ； ； ； wherein, is a safety index, is a minimum obstacle distance, is an adjusted safety distance, is an adjusted remaining energy, is an adjusted time margin, is an energy consumption threshold, is a mission deadline, is a maximum node risk value, is a pre-adjusted safety distance, is an expansion coefficient, is a previous time safety index, is an energy buffer coefficient, is a time relaxation coefficient.

6. The method of claim 1, wherein, According to the initial flight path, the energy consumption data is predicted, and if the energy consumption data exceeds a threshold value, the initial flight path is optimized to obtain an optimized path with reduced energy consumption data, including: According to the initial flight path, the following formula is used to predict energy consumption data: ； ； wherein, is an energy consumption prediction value, is a total power, is a speed function, is an acceleration function, is an altitude function, is an emergency reserve power, is a base power consumption, is an auxiliary power consumption, is a motor efficiency, is an air density, is a drag coefficient, is a wind area, is a flight speed, is a UAV mass, is a gravitational acceleration, is a climb angle, is an acceleration; If the energy consumption data exceeds a threshold value, an energy consumption heat map of the initial flight path is constructed, and according to the energy consumption heat map, a high energy consumption segment is determined; The high energy consumption segment is analyzed to obtain energy consumption analysis results, and according to the energy consumption analysis results, optimized path parameters are generated; According to the optimized path parameters, the initial flight path is adjusted to generate the optimized path with reduced energy consumption data.

7. An unmanned aerial vehicle low-altitude air route dynamic planning device, characterized in that, The device includes various functional modules required to implement the method of any one of claims 1 to 6.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images