An unmanned aerial vehicle path planning method and device, an electronic device, and a storage medium

Abstract

The application relates to the technical field of unmanned planes, and discloses an unmanned plane path planning method and device, electronic equipment and a storage medium, the method comprising the following steps: collecting multi-modal data; calculating a comprehensive dynamic entropy under a current scene, constructing and updating a historical communication gain map; using a Kalman filtering model to establish a state equation and an observation equation, identifying dynamic parameters in real time, and taking the dynamic parameters as kinematic hard constraints of path planning; using segmented quantum bit coding to perform quantum coding on the unmanned plane path, integrating the historical communication gain map and the kinematic hard constraints into a fitness function, taking the maximum fitness as an optimization target, and obtaining an optimal path coordinate sequence; and performing unmanned plane path planning according to the optimal path coordinate sequence; the segmented quantum bit coding is used for quantum coding of the unmanned plane path, the maximum fitness is taken as the optimization target, a closed loop from data sensing to path execution is formed, and optimal path planning of the unmanned plane is realized.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and specifically to a UAV trajectory planning method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the rapid development and application of drone technology, more and more urban airspaces are becoming drone mission execution spaces. However, open airspace is subject to temporary factors such as weather changes, bird / balloon intrusions, no-fly zones, or temporary occupation. Against this backdrop, how to plan drone flight paths has become an urgent problem to be solved. Summary of the Invention

[0003] This invention provides a method, apparatus, electronic device, and storage medium for planning unmanned aerial vehicle (UAV) flight paths, in order to solve the problem of how to plan UAV flight paths.

[0004] In a first aspect, the present invention provides a method for planning the trajectory of an unmanned aerial vehicle (UAV), the method comprising:

[0005] Collect multimodal data during the preparation or execution of UAV operations. The multimodal data includes obstacle information, dynamic target information, and airspace permission information. Based on multimodal data, calculate the comprehensive dynamic entropy in the current scenario, and construct and update the historical communication gain map; State equations and observation equations are established using a Kalman filter model, dynamic parameters are identified in real time, and the dynamic parameters are used as kinematic hard constraints for trajectory planning. The UAV trajectory is quantum-encoded using segmented qubit encoding. The historical communication gain map and kinematic hard constraints are incorporated into the fitness function. The optimal trajectory coordinate sequence is obtained by taking the maximum fitness as the optimization objective. UAV trajectory planning is performed based on the optimal trajectory coordinate sequence.

[0006] This invention comprehensively perceives the surrounding environment of a UAV by collecting multimodal data, uses integrated dynamic entropy to characterize the dynamic level of the airspace, constructs and updates a historical communication gain map to provide communication basis for trajectory planning, uses a Kalman filter model to identify dynamic parameters in real time to avoid situations where a UAV is theoretically flyable but actually not. It uses segmented qubit encoding to perform quantum encoding of the UAV trajectory, with the maximum fitness as the optimization objective, to obtain the optimal trajectory coordinate sequence for UAV trajectory planning, forming a closed loop from data perception to trajectory execution, and realizing the optimal trajectory planning of the UAV.

[0007] In one optional implementation, calculating the comprehensive dynamic entropy of the current scene includes: Calculate the static heterogeneity entropy based on the heterogeneity of obstacle types and sizes within the UAV's operating airspace; Calculate the dynamic fluctuation entropy based on the motion uncertainty of dynamic targets within the UAV operating airspace and the frequency of no-fly zone changes; Set balance coefficients for static and dynamic scenarios respectively, and then weight and fuse the static heterogeneous entropy and dynamic fluctuation entropy according to the balance coefficients to obtain the comprehensive dynamic entropy under the current scenario.

[0008] This invention utilizes static heterogeneous entropy to quantify the complexity of static airspace and dynamic fluctuation entropy to quantify the uncertainty of dynamic scenarios. It integrates static heterogeneous entropy and dynamic fluctuation entropy into a comprehensive dynamic entropy, which can cope with the heterogeneity of static environment and respond to the uncertainty of dynamic environment, thereby improving the adaptability of UAVs in complex airspace operations.

[0009] In one alternative implementation, constructing and updating a historical communication gain map includes: Initialize the historical communication gain map based on prior communication data of the UAV's operating area; The historical communication gain values ​​of the grid in the historical communication gain map are calculated using the weighted average method; The historical communication gain values ​​in the grid of the historical communication gain map are updated periodically and / or conditionally. The historical communication gain map is updated by matching the dynamically updated historical communication gain values ​​with the grid in the historical communication gain map.

[0010] This invention converts communication data into a gridded historical communication gain map by initializing and updating the historical communication gain map, providing communication reliability constraints for UAV trajectory planning.

[0011] In one optional implementation, a Kalman filter model is used to establish state equations and observation equations, identify dynamic parameters in real time, and use these dynamic parameters as kinematic hard constraints for trajectory planning, including: Obtain the UAV's position and flight attitude angles, and construct a UAV kinematic model based on the UAV's position and flight attitude angles; The kinematic model of the UAV is transformed into state equations and observation equations using the Kalman filter model, and the dynamic parameters are identified in real time using velocity and turning radius as the state variables to be identified. The output speed and turning radius are used as hard kinematic constraints.

[0012] This invention identifies dynamic parameters through a Kalman filter model, providing hard kinematic constraints for trajectory planning and laying the foundation for the kinematic feasibility of the planned trajectory.

[0013] In one optional implementation, the UAV trajectory is quantum-encoded using segmented qubit encoding. Historical communication gain maps and kinematic hard constraints are incorporated into the fitness function, with maximizing fitness as the optimization objective, to obtain the optimal trajectory coordinate sequence, including: The drone's flight path is divided into sub-paths, and quantum bit encoding is used to encode the sub-paths and initialize the quantum population. The sub-paths include takeoff, cruise, operation, and landing. Adjusting the quantum gate rotation angle based on comprehensive dynamic entropy; The kinematic hard constraints are transformed into parameter deviation values ​​of kinematic feasibility costs. Track gain values ​​are extracted from the communication gain map and transformed into historical communication gain costs. Based on the fitness function, all individuals in the quantum population are optimized, and the coordinate sequence corresponding to the track with the highest fitness is determined as the optimal track coordinate sequence.

[0014] This invention uses quantum bit encoding to divide the UAV trajectory into sub-trajectories, enabling independent control of parameters at each stage of the trajectory. Based on the comprehensive dynamic entropy, the quantum gate rotation angle is adjusted for scenario-based adaptive adjustment. The fitness function is used to optimize quantum individuals, so that the final trajectory satisfies kinematic and communication constraints while ensuring the executability of the trajectory.

[0015] In one alternative implementation, the fitness function is as follows:

[0016] in, For fitness, For path length cost, The weighting coefficients for path length cost. For the sake of safety, The weighting coefficient for security costs. As a cost of dynamic constraints, These are the weighting coefficients for the dynamic constraint cost. For the sake of kinematic feasibility, The weighting coefficients for the kinematic feasibility cost. This comes at the cost of historical communication gains. This is the weighting coefficient for the historical communication gain cost.

[0017] This invention integrates path length, security, dynamic constraints, kinematic feasibility, and historical communication gain by designing a fitness function to achieve multi-dimensional constraints and realize multi-objective integrated evaluation.

[0018] In one optional implementation, optimization is performed on all individuals in the quantum population based on a fitness function, and the coordinate sequence corresponding to the track with the highest fitness is determined as the optimal track coordinate sequence, including: By combining the comprehensive dynamic entropy to set the screening threshold, the population with fitness greater than or equal to the screening threshold is retained; Select individuals with high fitness from the retained population, and entangle the qubits of the selected individuals with the qubits of randomly selected ordinary individuals until the population converges to a stable state. The quantum individual with the highest fitness is selected from the converged population as the optimal solution. The qubit encoding of the optimal solution is analyzed to obtain the optimal track coordinate sequence.

[0019] This invention achieves scenario-based adaptive quantum population optimization by screening quantum populations, thereby optimizing flight paths and providing data support for UAV flight path planning.

[0020] Secondly, the present invention provides a drone trajectory planning device, the device comprising: The data acquisition module is used to collect multimodal data during the preparation or execution of UAV operations. The multimodal data includes obstacle information, dynamic target information, and airspace permission information. The calculation module is used to calculate the comprehensive dynamic entropy of the current scenario based on multimodal data, and to build and update the historical communication gain map; The identification module is used to establish state equations and observation equations using the Kalman filter model, identify dynamic parameters in real time, and use the dynamic parameters as kinematic hard constraints for trajectory planning. The optimization module is used to quantum encode the UAV trajectory using segmented qubit encoding, incorporate the historical communication gain map and kinematic hard constraints into the fitness function, and obtain the optimal trajectory coordinate sequence with the maximum fitness as the optimization objective. The planning module is used to plan the UAV trajectory based on the optimal trajectory coordinate sequence.

[0021] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the UAV trajectory planning method of the first aspect or any corresponding embodiment described above.

[0022] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the unmanned aerial vehicle (UAV) trajectory planning method of the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of UAV flight path planning and deployment according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the UAV trajectory planning method according to an embodiment of the present invention; Figure 3 This is a system framework diagram for implementing a UAV trajectory planning method according to an embodiment of the present invention; Figure 4 This is a structural block diagram of an unmanned aerial vehicle (UAV) trajectory planning device according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0027] In related technologies, quantum genetic algorithms have been attempted for UAV trajectory planning. However, while quantum genetic algorithms exhibit significant advantages in static airspace trajectory planning due to their quantum superposition and entanglement properties, they struggle to adapt to dynamic, heterogeneous airspace. In addition, improved versions of traditional genetic algorithms and particle swarm optimization algorithms have been used in dynamic airspace planning research. However, these improvements primarily focus on operator fine-tuning and parameter correction, failing to establish a linkage mechanism with the dynamic complexity of the airspace, resulting in slow convergence speeds and weak global optimization capabilities.

[0028] This invention provides a method for UAV trajectory planning. It utilizes multimodal perception to quantify the dynamic complexity of the airspace, providing a reference for the adaptive adjustment of quantum genetic algorithm parameters. It establishes an online identification model of UAV kinematic parameters, enhances the airspace dynamic constraint response capability of quantum entanglement operators, integrates the historical grid communication quality of the current airspace, constructs a communication gain map, improves the reliability of UAV operation communication, and realizes real-time trajectory planning for UAVs operating in dynamic heterogeneous airspace.

[0029] For example, a schematic diagram of drone flight path planning and deployment is shown below. Figure 1 As shown, the deployment scheme consists of a task drone 1, a flight path 2, a dynamic target 3, a ground station 4, and a static target 5. The task drone 1 performs its mission according to the planned flight speed and flight path 2. The ground station 4 is powered on and can display the current flight status of the drone.

[0030] According to an embodiment of the present invention, an embodiment of a UAV trajectory planning method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] This embodiment provides a method for planning the trajectory of an unmanned aerial vehicle (UAV). Figure 2 This is a flowchart of a UAV trajectory planning method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Collect multimodal data of the UAV operation preparation process or operation execution process.

[0032] In this embodiment of the invention, during the preparation or execution of UAV operations, i.e. before takeoff or during operation, the UAV's onboard multi-source sensors comprehensively perceive multimodal data in the UAV trajectory planning scenario. This multimodal data includes: static obstacle location and size information collected by the UAV's onboard lidar; dynamic target locations such as birds, balloons, and other UAVs collected by the UAV's onboard visual camera; dynamic target speeds collected by the UAV's onboard millimeter-wave radar; real-time airspace permission information such as temporary no-fly zone coordinates and flight altitude restrictions obtained through the API interface provided by the low-altitude traffic management system; real-time meteorological data collected by the UAV's onboard climate sensor; and communication gain data collected by the UAV's onboard communication module.

[0033] Step S202: Based on multimodal data, calculate the comprehensive dynamic entropy in the current scenario, and construct and update the communication gain map.

[0034] In this embodiment of the invention, the comprehensive dynamic entropy is calculated based on the collected multimodal data. The comprehensive dynamic entropy is a mixed quantitative index of the static heterogeneous characteristics and dynamic fluctuation characteristics of the UAV operating airspace. The higher the value, the higher the dynamic complexity of the airspace.

[0035] A communication gain map is constructed, which is a historical statistical summary of the communication quality of each grid in the UAV's operational airspace. This map reflects the stability of communication status in different airspaces and provides a communication reliability reference for flight path planning. To ensure the real-time performance and accuracy of the communication gain map, the constructed map is updated.

[0036] Step S203: Use the Kalman filter model to establish the state equation and observation equation, identify the dynamic parameters in real time, and use the dynamic parameters as the kinematic hard constraints for trajectory planning.

[0037] In this embodiment of the invention, a Kalman filter model is used to establish state equations and observation equations. Velocity and turning radius are used as state variables to be identified. Dynamic parameters are identified in real time and used as kinematic hard constraints for trajectory planning to ensure that the planned trajectory is within the physical flight capability range of the UAV.

[0038] Step S204: Integrate the communication gain map and kinematic hard constraints into the fitness function, and use the maximum fitness as the optimization objective to obtain the optimal track coordinate sequence.

[0039] In this embodiment of the invention, a fitness function is constructed, and the communication gain map and kinematic hard constraints are integrated into the fitness function to perform trajectory optimization. The optimization objective is to maximize the fitness, and the quantum individual with the highest fitness is taken as the optimal solution for trajectory optimization. The optimal quantum individual is then analyzed to obtain the optimal trajectory coordinate sequence.

[0040] Step S205: Plan the UAV trajectory based on the optimal trajectory coordinate sequence.

[0041] In this embodiment of the invention, the UAV is guided to perform planned operations based on the optimal track coordinate sequence.

[0042] The UAV trajectory planning method provided in this embodiment comprehensively perceives the surrounding environment of the UAV by collecting multimodal data, uses integrated dynamic entropy to characterize the dynamic level of the airspace, constructs and updates a historical communication gain map to provide communication basis for trajectory planning, uses a Kalman filter model to identify dynamic parameters in real time to avoid situations where the UAV is theoretically flyable but not actually flyable, and uses segmented qubit encoding to perform quantum encoding of the UAV trajectory, with the maximum fitness as the optimization objective, to obtain the optimal trajectory coordinate sequence for UAV trajectory planning, forming a closed loop from data perception to trajectory execution, and realizing the optimal trajectory planning of the UAV.

[0043] This embodiment provides a method for planning the trajectory of an unmanned aerial vehicle (UAV), which includes the following steps: Step S301: Collect multimodal data of the UAV operation preparation process or operation execution process.

[0044] Please see details Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0045] Step S302: Based on multimodal data, calculate the comprehensive dynamic entropy in the current scenario, and construct and update the communication gain map.

[0046] Specifically, the calculation of the comprehensive dynamic entropy in the current scenario in step S302 above includes: Step S3021: Calculate the static heterogeneity entropy based on the heterogeneity of obstacle types and sizes within the UAV's operating airspace. Step S3022: Calculate the dynamic fluctuation entropy based on the motion uncertainty of dynamic targets within the UAV's operating airspace and the frequency of no-fly zone changes; Step S3023: Set the balance coefficients for the static scene and the dynamic scene respectively. Based on the balance coefficients, the static heterogeneous entropy and the dynamic fluctuation entropy are weighted and fused to obtain the comprehensive dynamic entropy under the current scene.

[0047] In this embodiment of the invention, based on the abnormal intrusion of obstacles and dynamic targets in the operational airspace, the operational scenarios of the UAV are divided into low-dynamic scenarios, medium-dynamic scenarios, and high-dynamic scenarios. Specifically, low-dynamic scenarios are generally located in open suburban areas, with a high proportion of static obstacles, no dynamic targets, and no changes in no-fly zones. Medium-dynamic scenarios are generally located at the urban-rural fringe, with a small number of dynamic targets and occasional changes in no-fly zones. High-dynamic scenarios are generally located in the core urban area, with multiple dynamic targets and frequent changes in no-fly zones.

[0048] Taking into account both static heterogeneity and dynamic fluctuations, a hybrid entropy model is designed: Static heterogeneous entropy ( H 1): Based on the heterogeneity of obstacle types and sizes within the UAV operating airspace, define... ,in, P ( i ) is the first i The proportion of obstacles in the airspace is obtained through statistical analysis of sensor data.

[0049] Dynamic fluctuation entropy ( H 2): Based on the motion uncertainty of dynamic targets within the UAV operating airspace and the frequency of no-fly zone changes, define... ,in, P ( v ) represents the velocity distribution probability of dynamic obstacles. FThis refers to the number of times the no-fly zone changes within a unit of time. This is the weighting coefficient (typically 0.4).

[0050] Comprehensive dynamic entropy ( H ):definition ,in, To balance the coefficients, a static scene is generally chosen. Dynamic scenes H∈[0,1], the higher the entropy value, the higher the spatial dynamic complexity.

[0051] It should be noted that if static heterogeneous entropy H 1 and dynamic fluctuation entropy H If 2 exceeds [0, 1], it is mapped to the interval [0, 1] through linear normalization.

[0052] By using static heterogeneous entropy to quantify the complexity of static airspace and dynamic fluctuation entropy to quantify the uncertainty of dynamic scenarios, static heterogeneous entropy and dynamic fluctuation entropy are integrated into a comprehensive dynamic entropy, which can not only cope with the heterogeneity of static environment but also respond to the uncertainty of dynamic environment, thereby improving the adaptability of UAVs in complex airspace operations.

[0053] Specifically, the construction and updating of the communication gain map in step S302 above includes: Step S3024: Initialize the historical communication gain map based on the prior communication data of the UAV operation area; Step S3025: Calculate the historical communication gain values ​​of the grids in the historical communication gain map using the weighted average method; Step S3026: Update the historical communication gain values ​​of the grid in the historical communication gain map periodically and / or conditionally. Step S3027: Update the historical communication gain map by matching the dynamically updated historical communication gain values ​​with the grids in the historical communication gain map.

[0054] In this embodiment of the invention, the communication gain map is a historical statistical summary and accumulation of the communication quality of each grid within the UAV's operational airspace. It is used to reflect the stability of communication status in different airspaces and to provide a communication reliability reference for flight path planning. The construction and updating process is as follows: (1) Initialize the historical communication gain map: Based on the base station coverage area of ​​the UAV operation area, historical operation communication records and other prior communication data, initialize the historical communication gain map; (2) Data acquisition and supplementation: During the operation of the UAV, the communication module collects the communication gain data of each location of the UAV in real time, such as RSRP (Reference Signal Receiving Power), SINR (Signal to Interference plus Noise Ratio), packet loss rate, etc., and obtains the corresponding grid labels in combination with GPS or high-precision positioning technology as a source of historical data supplementation. (3) Calculation of communication gain value: The historical communication gain value of the grid in the historical communication gain map is calculated using the weighted average method. The core of the weighted average is that the weight of recent data is higher than that of distant data. The formula for calculating the historical communication gain value is: , G 1~ G n These are the communication gain measurements for this grid at different times. w 1~ w n For the weights at corresponding times, follow ,in, The attenuation coefficient is 0.02. t_now For the current moment, t k For the first k The measurement time should be chosen to ensure that recent communication status has a greater impact on historical gain values. (4) Dynamic Update: The historical communication gain values ​​of the grids in the historical communication gain map are dynamically updated using periodic updates, conditional updates, or a combination of periodic and conditional updates. For example, the update frequency of periodic updates is 2Hz, that is, once every 0.5 seconds. Conditional updates are performed when the communication gain measurement value of a grid deviates from the current historical value by more than 20%, thereby ensuring the timeliness of the historical communication gain map. (5) Update the historical communication gain map: accurately match the dynamically updated historical communication gain values ​​with the grid in the historical communication gain map.

[0055] It should be noted that invalid data due to sensor malfunctions, abrupt changes in communication gain caused by electromagnetic interference, and other abnormal values ​​are filtered out using a 3D filter. After outliers are removed according to the criteria, a weighted average is calculated to ensure the accuracy of the historical communication gain map.

[0056] By initializing and updating the historical communication gain map, the communication data is transformed into a gridded historical communication gain map, providing communication reliability constraints for UAV trajectory planning.

[0057] Step S303: Use the Kalman filter model to establish the state equation and observation equation, identify the dynamic parameters in real time, and use the dynamic parameters as the kinematic hard constraints for trajectory planning.

[0058] Specifically, step S303 includes: Step S3031: Obtain the UAV position and flight attitude angles, and construct a UAV kinematic model based on the UAV position and flight attitude angles; Step S3032: The kinematic model of the UAV is transformed into state equations and observation equations using the Kalman filter model, and the dynamic parameters are identified in real time using velocity and turning radius as the state variables to be identified. Step S3033: The output speed and turning radius are used as kinematic hard constraints.

[0059] In this embodiment of the invention, the UAV trajectory planning under the traditional static quantum genetic algorithm does not take into account the influence of the real environment and load on the UAV's motion parameters. The flight trajectory under ideal conditions will produce large errors and bring potential hazards.

[0060] A kinematic parameter identification model for the UAV is established to acquire real-time kinematic parameters under the influence of wind resistance, load, etc., and to reconstruct and plan the UAV's real flight path. Based on the phase displacement and velocity of the UAV output by the onboard optical flow odometry, the flight attitude angle output by the IMU (Inertial Measurement Unit), and the absolute position and velocity of the UAV output by the RTK, the UAV's position and flight attitude angle are determined.

[0061] A kinematic model of the UAV is established based on its position and flight attitude angles. For a multi-rotor UAV, the kinematic model is as follows:

[0062] in,( x , y , z ( ) represents the location of the drone. v For flight speed, The pitch angle, This is the heading angle.

[0063] Using a Kalman filter model, the kinematic model is transformed into state equations and observation equations, with velocity... v Turning radius R ( , (where is the rate of change of heading angle) is the state variable to be identified, and the state equation is as follows: ,in, Here is the state transition matrix. For the control matrix, For motor control quantities, For process noise, the observation equation is as follows: ,in, C For the observation matrix, To observe noise.

[0064] The identification results are output, and the output speed and turning radius are used as hard kinematic constraints. It should be noted that, to ensure the real-time performance of the identification results, the identification results are updated every 10ms, and the updated speed and turning radius are used as hard kinematic constraints.

[0065] By identifying dynamic parameters through a Kalman filter model, kinematic hard constraints are provided for trajectory planning, laying the foundation for the kinematic feasibility of the planned trajectory.

[0066] Step S304: The UAV trajectory is quantum-encoded using segmented qubit encoding. The communication gain map and kinematic hard constraints are incorporated into the fitness function. The optimal trajectory coordinate sequence is obtained with the maximum fitness as the optimization objective.

[0067] Specifically, step S304 includes: Step S3041: Divide the UAV track into sub-tracks, use qubit encoding to perform quantum encoding on the sub-tracks, and initialize the quantum population. Step S3042: Adjust the quantum gate rotation angle based on the comprehensive dynamic entropy; Step S3043: Transform the kinematic hard constraints into parameter deviation values ​​of kinematic feasibility costs, extract track gain values ​​from the communication gain map and transform them into historical communication gain costs, optimize all individuals in the quantum population based on the fitness function, and determine the coordinate sequence corresponding to the track with the highest fitness as the optimal track coordinate sequence.

[0068] In this embodiment of the invention, segmented qubit encoding is used to divide the UAV trajectory into four independent and controllable segmented units: takeoff, cruise, operation, and landing. This preserves the parallel search capability of multiple solutions brought by quantum superposition states, as well as the precise control of parameters at each stage of the trajectory, and is applicable to the kinematic constraints differences of different stages such as UAV takeoff and cruise.

[0069] For each segmented unit, sub-segments can be dynamically added according to the actual situation to optimize the trajectory planning. For example, if an occasional obstacle or abnormal drone intrudes into the airspace during the operation, the operation can be increased to 2-3 sub-segments. Each trajectory segment is allocated 4 qubits (1 qubit group), which are mapped to key trajectory parameters to achieve a full-dimensional representation of "position-altitude-velocity". The specific allocation is as follows: Plane coordinates ( x , y ): Occupies 2 qubits and uses binary encoding to map geographic coordinates; Flight altitude (z Occupies 1 qubit, adaptable to different operating altitude requirements of drones (such as power line inspection). z =5-10m, low-altitude logistics z =10-20m); Flight speed ( v ): Occupies 1 qubit, mapping the corresponding segment of flight speed (such as the takeoff segment). v =2-5m / s, cruising section v =5-15m / s).

[0070] The quantum encoding of a single track is as follows:

[0071] in, : k It is a segmented index (1~N, corresponding to takeoff / cruise, etc.). m The ground state index (0~15, corresponding to 16 combinations of qubits) ); For the 4-qubit first m ground state (e.g.) m =0 corresponds to until m =15 corresponds to The normalization condition is that all ground state coefficients within each track segment satisfy... (Probability conservation in a single segment), rather than all segments sharing a single coefficient.

[0072] Converting the physical parameters of sub-tracks into quantum encoding forms to realize the quantum state expression of tracks and complete the initialization of the quantum population is the foundation of quantum genetic algorithms.

[0073] By combining comprehensive dynamic entropy, the quantum gate rotation angle step size is adaptively adjusted. The approach prioritizes scenario-based logic adaptation over formula-based quantitative calculation. The rotation angle step size formula is used to accurately calculate the step size value under different entropy values, while the adjustment logic clarifies the step size adaptation target under different dynamic scenarios. The rotation angle step size formula is as follows:

[0074] in, This is the initial step size (typically taken as 0.05π). This is the attenuation coefficient (typically taken as 1.2, controlling the rate of change of the step size with the entropy value).

[0075] It should be noted that, in order to meet the needs of UAV trajectory planning in different scenarios, especially UAV trajectory optimization in dynamic scenarios, the rotation angle values ​​are as follows under different scenario logics: When H < 0.3 (low dynamics), the value is 0.03π (adjustable); When 0.3≤H<0.7 (medium dynamic), the value is 0.05π (adjustable); When H≥0.7 (high dynamic), the value is 0.08π (adjustable).

[0076] The quantum gate uses the Hadamard gate to achieve population initialization (uniform distribution) and the X gate (Pauli-X) to achieve mutation, dynamically selecting the rotation direction based on the fitness value and entropy value.

[0077] By quantum-encoding the sub-tracks of the UAV's flight path, independent control of parameters at each stage of the flight path is achieved. The quantum gate rotation angle is adjusted based on the comprehensive dynamic entropy to perform scenario-based adaptive adjustment. The fitness function is used to optimize the quantum individuals, so that the final flight path satisfies the kinematic and communication constraints while ensuring the executability of the flight path.

[0078] To address the constraints of path shortest distance, safety, obstacles and dynamic targets, kinematic feasibility, and historical communication gain in UAV operations, a fitness function with multi-dimensional constraints is designed as follows:

[0079] in, For fitness, For path length cost, The weighting coefficients for path length cost. For the sake of safety, The weighting coefficient for security costs. As a cost of dynamic constraints, These are the weighting coefficients for the dynamic constraint cost. For the sake of kinematic feasibility, The weighting coefficients for the kinematic feasibility cost. This comes at the cost of historical communication gains. This is the weighting coefficient for the historical communication gain cost.

[0080] The deviation between the kinematic parameters of the actual flight of the UAV and the parameters of the planned trajectory is quantified, thereby transforming the hard kinematic constraints into parameter deviation values ​​of kinematic feasibility costs. The larger the deviation, the higher the kinematic feasibility cost and the lower the trajectory feasibility.

[0081] The gain values ​​of the grids traversed by the UAV's flight path are extracted from the communication gain map and converted into historical communication gain costs. The lower the gain value, the higher the historical communication gain cost and the lower the reliability of the flight path communication.

[0082] With the goal of maximizing fitness, optimization is performed on all individuals in the quantum population, and the coordinate sequence corresponding to the track with the highest fitness is determined as the optimal track coordinate sequence.

[0083] Specifically, step S3043 above, which involves optimizing all individuals in the quantum population based on the fitness function and determining the coordinate sequence corresponding to the track with the highest fitness as the optimal track coordinate sequence, includes: Step S30431: Combine the comprehensive dynamic entropy to set the screening threshold and retain the population with fitness greater than or equal to the screening threshold; Step S30432: Select individuals with high fitness from the retained population, and perform entanglement operations on the qubits of the selected individuals and the qubits of randomly selected ordinary individuals until the population converges to a stable state. Step S30433: Select the quantum individual with the highest fitness from the converged population as the optimal solution, analyze the quantum bit encoding of the optimal solution, and obtain the optimal trajectory coordinate sequence.

[0084] In this embodiment of the invention, the fitness value of each individual in the population is calculated, and a screening threshold T is set: T = T0 × (1 + H), where T0 is the basic threshold (generally taken as 0.6). The population with fitness greater than or equal to the screening threshold is retained, and inferior individuals are eliminated to avoid population redundancy.

[0085] To enhance the population's ability to quickly respond to sudden constraints such as no-fly zones and abrupt changes in wind speed, individuals with high fitness rankings are selected; for example, the top 20% of individuals by fitness are selected. The qubits of these selected individuals are entangled with the qubits of randomly selected ordinary individuals to generate new entangled individuals. This process continues until the population converges to a stable state, where fitness no longer significantly increases. Entangled individuals inherit the superior genes of elite individuals while introducing diversity from ordinary individuals, enabling rapid adaptation to sudden constraints.

[0086] The quantum individual with the highest fitness is selected from the converged population as the optimal solution. Its qubit encoding is analyzed. According to the segmented encoding rule, the quantum ground state of the optimal quantum individual is mapped one by one to the core parameters such as the plane coordinates and flight altitude of each track segment. According to the segmented order of takeoff, cruise, operation and landing, the core parameters of each segment are continuously spliced ​​and smoothed to obtain the optimal track coordinate sequence of the UAV in the current airspace.

[0087] By dynamically controlling the population selection intensity through comprehensive dynamic entropy, and introducing quantum entanglement to specifically enhance the information correlation between individuals, the system balances population convergence efficiency and diversity, solving the problems of premature convergence and insufficient response to sudden constraints in traditional quantum genetic algorithms in dynamic scenarios. This achieves synergistic optimization where "entropy guides the evolutionary direction and entanglement operators enhance the anti-disturbance capability".

[0088] By screening quantum populations, we can achieve scenario-based adaptive optimization of quantum populations, realize trajectory optimization, and provide data support for UAV trajectory planning.

[0089] Step S305: Plan the UAV trajectory based on the optimal trajectory coordinate sequence.

[0090] Please see details Figure 2 Step S205 of the illustrated embodiment will not be described again here.

[0091] The UAV trajectory planning method provided in this embodiment incorporates multimodal perception methods for meteorological and communication gain, accurately capturing multimodal information such as airspace obstacles, weather disturbances, and communication quality. This enhances the perception of dynamic airspace complexity during UAV operations, broadens the adaptability of UAVs in complex scenarios, and, based on the algorithm framework of "dynamic perception of airspace environment - dynamic kinematic identification - communication constraint fusion - adaptive quantum genetic algorithm," enables UAVs to dynamically perceive scene complexity and corrects the parameters of the quantum genetic algorithm, achieving real-time performance and robustness in trajectory planning.

[0092] like Figure 3 As shown, Figure 3 The system framework diagram for implementing the UAV trajectory planning method includes an input layer, a processing layer, a planning layer, and an output layer.

[0093] The input layer includes the IMU, RTK, visual radar, and communication module carried by the operational UAV, as well as the low-altitude management system and meteorological system carried by the airspace environment. Among them, the IMU and RTK provide the UAV's flight attitude angle, position, and speed; the visual radar senses static and dynamic targets; the communication module collects communication data; the low-altitude management system provides no-fly zones and airspace permissions; and the meteorological system provides environmental data such as wind speed and visibility.

[0094] The processing layer includes online identification of kinematic parameters, which uses a Kalman filter model to output the UAV's kinematic parameters, including flight speed. v Turning radius R .

[0095] The planning layer evaluates the trajectory through qubit encoding, rotation angle adjustment, screening threshold, and fitness function.

[0096] The output layer outputs the optimal trajectory coordinate sequence of the UAV.

[0097] This embodiment also provides a UAV trajectory planning device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0098] This embodiment provides a drone trajectory planning device, such as... Figure 4As shown, it includes: The acquisition module 401 is used to acquire multimodal data during the preparation or execution of UAV operations. The multimodal data includes obstacle information, dynamic target information, and airspace permission information. The calculation module 402 is used to calculate the comprehensive dynamic entropy in the current scenario based on multimodal data, and to build and update the historical communication gain map; The identification module 403 is used to establish state equations and observation equations using the Kalman filter model, identify dynamic parameters in real time, and use the dynamic parameters as kinematic hard constraints for trajectory planning. The optimization module 404 is used to quantum encode the UAV trajectory using segmented qubit encoding, integrate the historical communication gain map and kinematic hard constraints into the fitness function, and obtain the optimal trajectory coordinate sequence with the maximum fitness as the optimization objective. Planning module 405 is used to plan the UAV trajectory based on the optimal trajectory coordinate sequence.

[0099] In some alternative implementations, the computing module 402 includes: The first computing unit is used to calculate the static heterogeneity entropy based on the heterogeneity of obstacle types and sizes within the UAV's operating airspace. The second calculation unit is used to calculate the dynamic fluctuation entropy based on the motion uncertainty of dynamic targets in the UAV's operating airspace and the frequency of changes in no-fly zones; The weighted fusion unit is used to set the balance coefficients for static and dynamic scenes respectively. Based on the balance coefficients, the static heterogeneous entropy and dynamic fluctuation entropy are weighted and fused to obtain the comprehensive dynamic entropy under the current scene.

[0100] In some alternative implementations, the computing module 402 further includes: The initialization unit is used to initialize the historical communication gain map based on prior communication data of the UAV's operating area. The third calculation unit is used to calculate the historical communication gain value of the grid in the historical communication gain map using the weighted average method; The first update unit is used to periodically update and / or conditionally update the historical communication gain values ​​of the grid in the historical communication gain map. The second update unit is used to update the historical communication gain map by matching the dynamically updated historical communication gain values ​​with the grid in the historical communication gain map.

[0101] In some alternative implementations, the identification module 403 includes: The model building unit is used to obtain the UAV's position and flight attitude angles, and to build a UAV kinematic model based on the UAV's position and flight attitude angles; The identification unit is used to transform the UAV kinematic model into state equations and observation equations using the Kalman filter model, and to identify dynamic parameters in real time using velocity and turning radius as the state variables to be identified. Define the element to use the output speed and turning radius as kinematic hard constraints.

[0102] In some alternative implementations, the identification module 403 further includes: The coding unit is used to divide the UAV trajectory into sub-trajectories, and uses qubits to encode the sub-trajectories for quantum encoding and quantum population initialization. The sub-trajectories include takeoff, cruise, operation, and landing. The adjustment unit is used to adjust the quantum gate rotation angle based on the comprehensive dynamic entropy; The optimization unit is used to transform kinematic hard constraints into parameter deviation values ​​of kinematic feasibility costs, extract track gain values ​​from the communication gain map and transform them into historical communication gain costs, optimize all individuals in the quantum population based on the fitness function, and determine the coordinate sequence corresponding to the track with the highest fitness as the optimal track coordinate sequence.

[0103] In some alternative implementations, the optimization unit includes: The reserved sub-unit is used to set the screening threshold by combining the comprehensive dynamic entropy, and retain the population with fitness greater than or equal to the screening threshold; Entangled subunits are used to select individuals with high fitness from the retained population, and to entangle the qubits of the selected individuals with the qubits of randomly selected ordinary individuals until the population converges to a stable state. The analytical subunit is used to select the quantum individual with the highest fitness from the converged population as the optimal solution, analyze the quantum bit encoding of the optimal solution, and obtain the optimal track coordinate sequence.

[0104] The UAV trajectory planning device provided in this embodiment of the invention can execute the UAV trajectory planning method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0105] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0106] The following is a detailed reference. Figure 5The diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from memory 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the electronic device. The processor 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0107] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 5 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0108] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a memory 508, or installed from a ROM 502. When the computer program is executed by the processor 501, it performs the functions defined in the UAV trajectory planning method of the embodiments of the present invention.

[0109] Figure 5 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0110] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the UAV trajectory planning method shown in the above embodiments is implemented.

[0111] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0112] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended invention.

Claims

1. A method for planning the trajectory of an unmanned aerial vehicle (UAV), characterized in that, The method includes: Collect multimodal data during the preparation or execution of UAV operations, including obstacle information, dynamic target information, and airspace permission information; Based on the multimodal data, calculate the comprehensive dynamic entropy in the current scenario, and construct and update the historical communication gain map; State equations and observation equations are established using a Kalman filter model, dynamic parameters are identified in real time, and the dynamic parameters are used as kinematic hard constraints for trajectory planning. The UAV trajectory is quantum-encoded using segmented qubit encoding. The historical communication gain map and the kinematic hard constraints are incorporated into the fitness function. The optimal trajectory coordinate sequence is obtained by taking the maximum fitness as the optimization objective. UAV trajectory planning is performed based on the optimal trajectory coordinate sequence.

2. The method according to claim 1, characterized in that, The calculation of the comprehensive dynamic entropy in the current scenario includes: Calculate the static heterogeneity entropy based on the heterogeneity of obstacle types and sizes within the UAV's operating airspace; Calculate the dynamic fluctuation entropy based on the motion uncertainty of dynamic targets within the UAV operating airspace and the frequency of no-fly zone changes; Balance coefficients are set for static and dynamic scenarios respectively. The static heterogeneous entropy and the dynamic fluctuation entropy are weighted and fused according to the balance coefficients to obtain the comprehensive dynamic entropy under the current scenario.

3. The method according to claim 1, characterized in that, The construction and updating of the historical communication gain map includes: Initialize the historical communication gain map based on prior communication data of the UAV's operating area; The historical communication gain values ​​of the grid in the historical communication gain map are calculated using the weighted average method; The historical communication gain values ​​in the grid of the historical communication gain map are updated periodically and / or conditionally. The historical communication gain map is updated by matching the dynamically updated historical communication gain values ​​with the grid in the historical communication gain map.

4. The method according to claim 1, characterized in that, The process of establishing state and observation equations using a Kalman filter model, identifying dynamic parameters in real time, and using these dynamic parameters as kinematic hard constraints for trajectory planning includes: Obtain the UAV's position and flight attitude angles, and construct a UAV kinematic model based on the UAV's position and flight attitude angles; The kinematic model of the UAV is transformed into state equations and observation equations using a Kalman filter model, and dynamic parameters are identified in real time using velocity and turning radius as state variables to be identified. The output speed and turning radius are used as hard kinematic constraints.

5. The method according to claim 1, characterized in that, The method involves quantum encoding the UAV trajectory using segmented qubit encoding, incorporating the historical communication gain map and the kinematic hard constraints into the fitness function, and using maximum fitness as the optimization objective to obtain the optimal trajectory coordinate sequence, including: The drone's flight path is divided into sub-paths, and quantum bit encoding is used to encode the sub-paths for quantum population initialization. The sub-paths include takeoff, cruise, operation, and landing. Adjust the quantum gate rotation angle based on the aforementioned comprehensive dynamic entropy; The kinematic hard constraints are transformed into parameter deviation values ​​of kinematic feasibility costs. Track gain values ​​are extracted from the communication gain map and transformed into historical communication gain costs. Based on the fitness function, all individuals in the quantum population are optimized, and the coordinate sequence corresponding to the track with the highest fitness is determined as the optimal track coordinate sequence.

6. The method according to claim 5, characterized in that, The fitness function is as follows: in, For fitness, For path length cost, The weighting coefficients for path length cost. For the sake of safety, The weighting coefficient for security costs. As a cost of dynamic constraints, These are the weighting coefficients for the dynamic constraint cost. For the sake of kinematic feasibility, The weighting coefficients for the kinematic feasibility cost. This comes at the cost of historical communication gains. This is the weighting coefficient for the historical communication gain cost.

7. The method according to claim 5, characterized in that, The process of optimizing all individuals in the quantum population based on a fitness function, and determining the coordinate sequence corresponding to the track with the highest fitness as the optimal track coordinate sequence, includes: By combining the comprehensive dynamic entropy to set a screening threshold, the population with fitness greater than or equal to the screening threshold is retained; Select individuals with high fitness from the retained population, and entangle the qubits of the selected individuals with the qubits of randomly selected ordinary individuals until the population converges to a stable state. The quantum individual with the highest fitness is selected from the converged population as the optimal solution. The qubit encoding of the optimal solution is analyzed to obtain the optimal track coordinate sequence.

8. A drone trajectory planning device, characterized in that, The device includes: The data acquisition module is used to collect multimodal data during the preparation or execution of UAV operations. The multimodal data includes obstacle information, dynamic target information, and airspace permission information. The calculation module is used to calculate the comprehensive dynamic entropy in the current scenario based on the multimodal data, and to construct and update the historical communication gain map; The identification module is used to establish state equations and observation equations using the Kalman filter model, identify dynamic parameters in real time, and use the dynamic parameters as kinematic hard constraints for trajectory planning. The optimization module is used to quantum encode the UAV trajectory using segmented qubit encoding, integrate the historical communication gain map and the kinematic hard constraints into the fitness function, and obtain the optimal trajectory coordinate sequence with the maximum fitness as the optimization objective. The planning module is used to plan the UAV trajectory based on the optimal trajectory coordinate sequence.

9. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory stores computer instructions, and the processor executes the UAV trajectory planning method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the UAV trajectory planning method according to any one of claims 1 to 7.

Images