Comprehensive decision-making method and system for flight path planning of unmanned aerial vehicle

Abstract

The invention relates to an unmanned aerial vehicle flight path planning comprehensive decision method and system, and the method comprises the steps: building a two-dimensional plane coordinate system for a target flight region according to the flight height of an unmanned aerial vehicle, and solving a minimum circumcircle for all obstacles in a two-dimensional plane; establishing an obstacle threat model of the unmanned aerial vehicle according to the minimum circumcircle of all obstacles in the two-dimensional plane; constructing a flight energy consumption model of the unmanned aerial vehicle according to the hovering power when the unmanned aerial vehicle hovers and the induction power when the unmanned aerial vehicle hovers; constructing a flight curvature model of the unmanned aerial vehicle according to the difference value of the direction angles between the adjacent path nodes of the unmanned aerial vehicle; establishing a target decision model of the unmanned aerial vehicle according to the obstacle threat model, the flight energy consumption model and the flight curvature model of the unmanned aerial vehicle; setting a flight starting point and a flight ending point of the unmanned aerial vehicle in a two-dimensional plane Extending path nodes of the unmanned aerial vehicle by using the target decision model from the flight starting point of the unmanned aerial vehicle; updating the parameters of the target decision model according to the obstacle threat of the unmanned aerial vehicle and the accumulated length of the current path every time the path node is expanded; and obtaining a final flight path of the unmanned aerial vehicle until the distance from the expanded path node to the flight end point of the unmanned aerial vehicle is smaller than a set threshold value. According to the method, the detail expression capability and the algorithm flexibility of route planning are effectively improved, and efficient route planning considering safety, economy and feasibility is realized in a complex two-dimensional scene.

Description

Technical Field

[0001] The present invention belongs to the field of unmanned aerial vehicle (UAV) path planning, and in particular relates to a comprehensive decision-making method and system for UAV trajectory planning. Background Art

[0002] With technological advancements, drones, with their high maneuverability, flexibility, and low cost, have shown tremendous potential in both military and civilian fields. Their applications encompass a wide range of areas, including wireless communication assistance, target tracking and identification, agricultural plant protection, and disaster relief. Autonomous flight capabilities are crucial for these missions, and trajectory planning, as a core technology enabling autonomous flight, enables drones to efficiently avoid various obstacles and ensure flight safety. Obstacle avoidance involves the process of detecting and responding to obstacles in an unknown environment while autonomously flying. This involves adjusting the flight attitude based on the size, shape, and location of the obstacle, combined with pre-defined obstacle avoidance strategies. This process involves key factors such as environmental perception, trajectory planning, and flight control. Trajectory planning involves planning the optimal collision-free path from a drone's starting point to its destination. This involves a system optimization problem that considers complex constraints, including the drone's flight time, energy consumption, autonomous obstacle avoidance, and communication efficiency.

[0003] The following major issues currently exist in the field of UAV trajectory planning: Regarding obstacle modeling, traditional methods often directly analyze irregular obstacles based on their original geometric form, resulting in high computational complexity for threat analysis and difficulty in quickly assessing path safety. When constructing energy consumption models, some technologies fail to fully incorporate the UAV's dynamic characteristics and ignore the mapping relationship between key parameters such as hovering power and induction power and actual flight energy consumption. This results in discrepancies between energy consumption predictions and the UAV's actual energy consumption, making it difficult to accurately support path optimization for long-duration missions. Regarding flight curvature constraints, existing technologies often use continuous curvature models to describe path smoothness, resulting in deviations between the model and the UAV's actual steering capabilities (such as the maximum rotation angle limit). The planned path may exceed the physical performance range of the aircraft, affecting flight feasibility. Furthermore, some algorithms lack dynamic parameter adjustment mechanisms based on obstacle threat and path length. Consequently, the weight parameters in the target decision model cannot be adaptively adjusted with environmental changes. This results in insufficient precision in balancing indicators such as obstacle avoidance and energy consumption control during multi-objective optimization, making it difficult to generate an optimal path that balances safety and economy in complex scenarios. Summary of the Invention

[0004] In order to solve the problems existing in the background technology, one aspect of the present invention provides a comprehensive decision-making method for UAV trajectory planning, comprising:

[0005] S1: Establish a two-dimensional plane coordinate system for the target flight area according to the flight altitude of the UAV, and find the minimum circumscribed circle of all obstacles in the two-dimensional plane;

[0006] S2: Establish an obstacle threat model for the drone based on the minimum circumscribed circle of all obstacles in a two-dimensional plane;

[0007] S3: Construct a flight energy consumption model of the UAV based on the hovering power and the induced power of the UAV when hovering;

[0008] S4: Construct the flight curvature model of the UAV based on the difference in direction angles between adjacent path nodes of the UAV;

[0009] S5: Establish a target decision model for the UAV based on the obstacle threat model, flight energy consumption model, and flight curvature model of the UAV;

[0010] S6: Set the starting point and end point of the UAV's flight on the two-dimensional plane; use the target decision model to expand the UAV's path nodes from the UAV's flight starting point;

[0011] S7: Each time a path node is extended, the parameters of the target decision model are updated according to the obstacle threat of the UAV and the cumulative length of the current path; until the distance from the extended path node to the UAV flight endpoint is less than the set threshold, the final flight path of the UAV is obtained.

[0012] Another aspect of the present invention also provides a UAV trajectory planning comprehensive decision-making system, which includes a memory and a processor; the memory is used to store an application; the processor is used to run the application and execute the UAV trajectory planning comprehensive decision-making method.

[0013] Another aspect of the present invention provides a computer storage medium, on which a remote monitoring program is stored. When the remote monitoring program is executed by a processor, the method for comprehensive decision-making for unmanned aerial vehicle trajectory planning is implemented.

[0014] The present invention has at least the following beneficial effects

[0015] In order to solve the problems in the prior art such as large modeling errors of irregular obstacles, deviations in mapping between energy consumption models and actual energy consumption, mismatches between curvature constraints and the physical performance of drones, and insufficient dynamic adaptability, the present invention simplifies irregular obstacles into minimum circumscribed circle modeling, which not only reduces the computational complexity of threat analysis through geometric simplification, but also improves path safety by reserving safety space using the circumscribed circle; the energy consumption model is constructed by fully considering the relationship between the hovering power and induction power of the drone and the actual energy consumption, thereby solving the problem of energy consumption prediction deviation in long-duration missions; discrete directional angle differences are used instead of continuous curvature constraints to make path smoothness modeling more consistent with the physical performance of the drone, such as the maximum rotation angle, and avoid the planned path exceeding the capability range of the aircraft; a dynamic parameter adjustment mechanism based on obstacle threat and path length is introduced to achieve an adaptive balance between safety and energy consumption indicators in multi-objective optimization; the present invention effectively improves the detail expression ability and algorithm flexibility of trajectory planning, and realizes efficient path planning that takes into account safety, economy and feasibility in complex two-dimensional scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the method flow of the present invention;

[0017] Figure 2 Schematic diagram of the simulation of the present invention. DETAILED DESCRIPTION

[0018] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. The following embodiments and features in the embodiments can be combined with each other unless there is a conflict. Combining the UAV obstacle threat model, flight energy consumption model and flight curvature model into a complete UAV trajectory planning model and establishing a multi-objective optimization problem can effectively solve the comprehensive performance problem of UAV trajectory planning.

[0019] See also Figure 1 One aspect of the present invention provides a comprehensive decision-making method for UAV trajectory planning, comprising:

[0020] S1: Establish a two-dimensional plane coordinate system for the target flight area according to the flight altitude of the UAV, and find the minimum circumscribed circle of all obstacles in the two-dimensional plane;

[0021] In this embodiment, step S1 transforms irregular obstacles into regular geometric shapes by establishing a two-dimensional coordinate system and calculating the minimum circumscribed circle of the obstacle. This effectively addresses the high computational complexity of threat analysis and low safety assessment efficiency caused by the complex shape of the obstacles in traditional methods. Its core effect lies in: geometric simplification reduces the complexity of obstacle modeling, thereby reducing the computational complexity of threat assessment in subsequent path planning; at the same time, the enveloping characteristics of the minimum circumscribed circle reserve sufficient safety distance for the drone.

[0022] For example, if a polygonal building (such as a pentagon) exists in the target flight area, traditional methods require calculating the distance between the drone and each edge to assess the threat. However, S1 solves the pentagon's minimum circumscribed circle (radius R), simplifying the obstacle into a circle with center coordinates O and radius R. In this case, the drone only needs to ensure that the distance between the path node and point O is greater than R + safety margin to avoid collision risk. The computational complexity is reduced from processing five edges to processing one circle, significantly improving modeling efficiency.

[0023] S2: Establish an obstacle threat model for the drone based on the minimum circumscribed circle of all obstacles in a two-dimensional plane;

[0024] Preferably, the obstacle threat model of the drone includes:

[0025]

[0026] O i =max{O i1 ,O i2 ,O i3 ,...,O ik ,...,O im}

[0027]

[0028] Where n represents the number of path nodes of the UAV path; r k represents the minimum circumscribed circle radius of the kth obstacle; d ik represents the distance from the i-th path node to the center of the minimum circumscribed circle of the k-th obstacle; max represents the maximum value; O i represents the obstacle threat of the i-th path node; O represents the obstacle threat of the UAV path.

[0029] In this embodiment, step S2 establishes a threat model based on the minimum circumscribed circle of the obstacle, quantifying the threat of irregular obstacles as the difference between the distance from the path node to the center of the circumscribed circle and the radius. This effectively addresses the cumbersome threat assessment and insufficient safety redundancy caused by the complex obstacle contours in traditional methods. By taking the maximum threat value from the circumscribed circle of the path node to all obstacles, this model ensures a more rigorous assessment of the safe distance between the drone path and obstacles. This not only simplifies the computational complexity of threat analysis from processing complex contours point by point to calculating the distance between the centers of the circles, improving modeling efficiency, but also reserves sufficient safety space for the flight path through the envelope characteristics of the circumscribed circle. Compared with the original contour modeling, it has higher safety redundancy. Moreover, the threat parameters can be updated in real time as the path expands, providing precise constraints for dynamic path optimization.

[0030] S3: Construct a flight energy consumption model of the UAV based on the hovering power and the induced power of the UAV when hovering;

[0031] Preferably, the flight energy consumption model of the UAV includes:

[0032]

[0033] Among them, L i represents the path length from the i-1th path node to the i-th path node; x i and y i represents the coordinates of the i-th path node in the two-dimensional coordinate system; L represents the path length of the UAV path; x i-1 and y i-1 represents the coordinates of the i-1th path node in the two-dimensional coordinate system; n represents the number of path nodes of the UAV path; P0 represents the hovering power of the UAV when it is hovering; δ represents the cross-sectional drag coefficient of the UAV, ρ represents the air density, s represents the rotor solidity of the UAV, A represents the rotor disc area of ​​the UAV, v represents the flight speed of the UAV, Ω represents the blade angular velocity of the UAV, and R represents the blade radius of the UAV; P i represents the induced power of the drone when it is hovering; k is the incremental correction factor of the induced power; W is the total weight of the drone; P fly Indicates the flight power of the drone; U tip represents the propeller tip speed, v0 is the induced speed of the UAV, and d0 is the reference flight speed; E flyall represents the flight energy consumption of the UAV; T represents the flight time of the UAV.

[0034] In this embodiment, step S3 constructs an energy consumption model by combining the drone's hovering power and induction power, effectively solving the problem of deviation between the energy consumption model and the actual flight energy consumption mapping in traditional technologies. The model accurately associates path length with parameters such as hovering power and induction power, reducing energy consumption prediction errors and providing reliable quantitative support for path optimization for long-duration missions.

[0035] For example, when the drone's hovering power is 100 watts and the induced power is calculated as 50 watts using the dynamic formula, if a certain path section is 100 meters long and the flight speed is 10 meters per second, the energy consumption of this section can be calculated according to the model to be (100+50)×(100÷10)=1500 joules. Compared with traditional models that do not consider dynamic parameters, this calculation method is more in line with the actual energy consumption of the drone, avoiding unreasonable path planning problems caused by energy consumption estimation deviations, and ensuring that path optimization in long-duration missions takes into account both energy economy and flight feasibility.

[0036] S4: Construct the flight curvature model of the UAV based on the difference in direction angles between adjacent path nodes of the UAV;

[0037] Preferably, the flight curvature model of the UAV includes:

[0038] C i =|θ i -θ i-1 |

[0039] C=max{|θ2-θ1|,|θ3-θ2|,...,|θ i -θ i-1 |,...,|θ n -θ n-1 |}

[0040] Among them, C i represents the curvature of the i-th path node; C represents the curvature of the current path; θ i-1 represents the direction angle of the UAV at the i-1th path node; θ i Indicates the direction angle of the UAV at the i-th path node; max indicates the maximum value.

[0041] In this embodiment, step S4 constructs the angular difference between adjacent path nodes of the UAV as a flight curvature model, effectively solving the problem of mismatch between the traditional continuous curvature model and the actual steering capability of the UAV. The model replaces the continuous curvature constraint with the discretized angular difference, making the path smoothness modeling more consistent with the physical properties of the UAV, such as the maximum rotation angle, avoiding the planned path from exceeding the steering capability of the body, reducing the modeling complexity of the curvature constraint, and improving the accuracy of the path feasibility assessment.

[0042] For example, when the drone's heading angle at a certain path node is 30° and the heading angle at the next node is 60°, the heading angle difference is 30°. If the drone's maximum rotation angle is 45°, then the path segment meets the curvature constraint; if the difference exceeds 45°, the model will automatically exclude the extended node. Compared with the traditional continuous curvature model that requires solving complex differential equations, this discretization process significantly improves modeling efficiency and engineering practicality.

[0043] S5: Establish a target decision model for the UAV based on the obstacle threat model, flight energy consumption model, and flight curvature model of the UAV;

[0044] Preferably, the target decision model of the UAV includes:

[0045]

[0046] C2:C=max{|θ2-θ1|,|θ3-θ2|,...,|θ i -θ i-1 |,...,|θ n -θ n-1 |}≤θ max

[0047]

[0048] C4:O i ≤threshold

[0049] Where P represents the target decision model; h represents the target optimization problem; Min represents the minimum value; ω1, ω2 and ω3 represent weight parameters; E0 represents the initial energy of the UAV; threshold represents the dynamically adjusted parameter; θ max Indicates the preset maximum direction angle.

[0050] In this embodiment, step S5 constructs a target decision model by integrating the obstacle threat model, the flight energy consumption model, and the flight curvature model. This transforms the multi-objective optimization problem into a unified mathematical expression with weighted parameters, effectively addressing the difficulty in coordinating and optimizing safety, energy consumption, and path smoothness in traditional algorithms. This model dynamically adjusts the priority of each metric using weight parameters ω1, ω2, and ω3, and integrates a dynamic threshold to achieve adaptive parameter updates. This allows path planning to balance obstacle avoidance safety (e.g., obstacle threat value O), energy efficiency (e.g., path length L), and flight feasibility (e.g., curvature C) in complex scenarios.

[0051] For example, in a certain planning scenario, ω1 = 0.4 (obstacle threat weight), ω2 = 0.3 (energy consumption weight), and ω3 = 0.3 (curvature weight) are set. When the threat value of candidate path node A is O = 0.2, the path length is L = 50 meters, and the curvature is C = 30°, and the O value of node B is O = 0.1, L = 60 meters, and C = 25°, the target decision model calculates h = 0.4 × 0.2 + 0.3 × (50 / 100) + 0.3 × (30 / 45) ≈ 0.33 and h = 0.4 × 0.1 + 0.3 × (60 / 100) + 0.3 × (25 / 45) ≈ 0.32, and selects node B with the smaller h as the extension node. This demonstrates the model's ability to collaboratively optimize multiple objectives. Furthermore, during the path extension process, the threshold parameter can be dynamically adjusted based on the real-time obstacle threat and cumulative path length, further enhancing the flexibility of the planning strategy.

[0052] S6: Set the starting point and end point of the UAV's flight on the two-dimensional plane; use the target decision model to expand the UAV's path nodes from the UAV's flight starting point;

[0053] Preferably, the method of expanding the path nodes of the UAV by using the target decision model includes:

[0054] A temporary path is constructed by selecting candidate expansion nodes from the neighboring nodes of the current path node, calculating the target optimization problem h, and selecting the candidate expansion node that minimizes the target optimization problem h as the final expansion node, wherein the neighboring nodes of the current node include: points on a circle with the current node as the center and a preset step size as the radius.

[0055] In this embodiment, step S6 effectively solves the problems of strong search blindness and low multi-objective optimization efficiency in traditional path planning by setting the flight start and end points and expanding the path nodes based on the target decision model. This step limits the neighbor node search range of the current node (such as discrete points within a circular area) with a preset step size as the radius. By constructing a temporary path and calculating the h value of the target optimization problem (comprehensive obstacle threat, energy consumption, curvature and other indicators), it ensures that the optimal node is selected in each expansion step, reducing the computational complexity of the path search, and improving the adaptability of the path to complex environments through a dynamic parameter adjustment mechanism.

[0056] For example, when the starting point of the drone is (4, -1) and the end point is (11, 17), and the preset step length is 5 meters, the neighbor nodes of the current path node (6, 2) are points evenly distributed on the circle with (6, 2) as the center and 5 meters as the radius (such as (11, 2), (6, 7), etc.). If the threat value of candidate expansion node A (11, 2) is O = 0.1, the path length L = 5 meters, and the curvature C = 20°, and the O of node B (6, 7) is O = 0.2, L = 5 meters, and C = 30°, the weight parameter is set. With the values ​​ω1 = 0.4, ω2 = 0.3, and ω3 = 0.3, it can be calculated that h of node A = 0.4×0.1+0.3×(5 / 100)+0.3×(20 / 45)≈0.163, and h of node B = 0.4×0.2+0.3×(5 / 100)+0.3×(30 / 45)≈0.243. Finally, node A with a smaller h value is selected to expand the path. This process achieves efficient and scientific path planning by limiting the search range and multi-objective quantitative evaluation.

[0057] S7: Each time a path node is expanded, the parameters of the target decision model are updated according to the obstacle threat to the UAV and the cumulative length of the current path; until the distance from the expanded path node to the UAV's flight destination is less than the set threshold, the UAV's final flight path is obtained;

[0058] Preferably, the parameters of the updated target decision model include:

[0059] threshold(d)=threshold(d-1)+reward O +reward L -penalty

[0060] reward O =scale O *exp(-O)

[0061]

[0062] Among them, scale O 、scale L and scale are the obstacle impact reward factor, cumulative path reward factor, and penalty factor, respectively; exp represents the exponential function; threshold(d) represents the dynamic adjustment parameter during the current expansion; and threshold(d-1) represents the dynamic adjustment parameter during the previous expansion.

[0063] Step S7 dynamically updates the target decision model parameters (such as threshold) according to the real-time obstacle threat and the cumulative path length after each path node expansion, effectively solving the problem that the fixed parameters in the traditional algorithm cannot adapt to environmental changes. This mechanism effectively solves the problem that the fixed parameters in the traditional algorithm cannot adapt to environmental changes by adjusting the obstacle impact reward factor (scale O ), cumulative path reward factor (scale L ) and penalty factor (scale) to adjust the threshold in real time, so that the model can adaptively balance obstacle avoidance safety and energy economy in multi-objective optimization, improve the environmental adaptability of path planning, and reduce path redundancy caused by fixed parameters.

[0064] For example, when the obstacle threat value O of a path is reduced from 0.5 to 0.3 after expansion, and the cumulative path length L increases from 80 meters to 90 meters, if scale O =0.3, scale L =0.3, scale=0.4, the reward can be calculated O =scale O *exp(-O)=0.222, Assuming the previous threshold was 5, the current threshold = 5 + 0.222 + 0.0033 - 0.12 ≈ 5.1053. The updated parameters will make subsequent node expansion more inclined to choose paths with lower threats while also taking into account path length optimization, ensuring that the optimal path that balances safety and efficiency is dynamically generated in complex scenarios.

[0065] Figure 2 This is the simulation result diagram of the present invention, and the parameter settings are shown in Table 1:

[0066] Table 1 Simulation parameters

[0067]

[0068] According to the simulation results Figure 2 As can be seen, the path effectively avoids obstacles, ensures that the rotation angle at each step does not exceed the maximum angle, and balances the path length to find the optimal path. This is a very feasible trajectory planning method.

[0069] Another aspect of the present invention also provides a UAV trajectory planning comprehensive decision-making system, which includes a memory and a processor; the memory is used to store an application; the processor is used to run the application and execute the UAV trajectory planning comprehensive decision-making method.

[0070] Another aspect of the present invention provides a computer storage medium, on which a remote monitoring program is stored. When the remote monitoring program is executed by a processor, the method for comprehensive decision-making for unmanned aerial vehicle trajectory planning is implemented.

[0071] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct 10RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), etc.

[0072] In summary, the present invention addresses the problems in the prior art such as large modeling errors of irregular obstacles, deviations in mapping energy consumption models and actual energy consumption, mismatches between curvature constraints and the physical performance of drones, and insufficient dynamic adaptability. The irregular obstacles are simplified into minimum circumscribed circle modeling, which not only reduces the computational complexity of threat analysis through geometric simplification, but also uses the circumscribed circle to reserve a safety space to improve path safety. The energy consumption model is constructed by fully considering the relationship between the hovering power and induction power of the drone and the actual energy consumption, thereby solving the problem of energy consumption prediction deviation in long-duration missions. The continuous curvature constraint is replaced by discrete directional angle difference, so that the path smoothness modeling is more in line with the physical performance of the drone, such as the maximum rotation angle, to avoid the planned path exceeding the capability range of the aircraft. A dynamic parameter adjustment mechanism based on obstacle threat and path length is introduced to achieve an adaptive balance between safety and energy consumption indicators in multi-objective optimization. The present invention effectively improves the detail expression ability and algorithm flexibility of trajectory planning, and realizes efficient path planning that takes into account safety, economy and feasibility in complex two-dimensional scenes.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A comprehensive decision-making method for UAV trajectory planning, characterized by: include: S1: Establish a two-dimensional plane coordinate system for the target flight area according to the flight altitude of the UAV, and find the minimum circumscribed circle of all obstacles in the two-dimensional plane; S2: Establish an obstacle threat model for the drone based on the minimum circumscribed circle of all obstacles in a two-dimensional plane; S3: Construct a flight energy consumption model of the UAV based on the hovering power and the induced power of the UAV when hovering; S4: Construct the flight curvature model of the UAV based on the difference in direction angles between adjacent path nodes of the UAV; S5: Establish a target decision model for the UAV based on the obstacle threat model, flight energy consumption model, and flight curvature model of the UAV; S6: Set the starting point and end point of the UAV's flight on the two-dimensional plane; use the target decision model to expand the UAV's path nodes from the UAV's flight starting point; S7: Each time a path node is extended, the parameters of the target decision model are updated according to the obstacle threat of the UAV and the cumulative length of the current path; until the distance from the extended path node to the UAV flight endpoint is less than the set threshold, the final flight path of the UAV is obtained.

2. The comprehensive decision-making method for UAV trajectory planning according to claim 1 is characterized in that: The obstacle threat model of the drone includes: The i =max{O i1 ,O i2 ,O i3 ,...,O ik ,...,O im } Where n represents the number of path nodes of the UAV path; r k represents the minimum circumscribed circle radius of the kth obstacle; d ik represents the distance from the i-th path node to the center of the minimum circumscribed circle of the k-th obstacle; max represents the maximum value; O i represents the obstacle threat of the i-th path node; O represents the obstacle threat of the UAV path.

3. The comprehensive decision-making method for UAV trajectory planning according to claim 2 is characterized in that: The flight energy consumption model of the UAV includes: Among them, L i represents the path length from the i-1th path node to the i-th path node; x i and y i represents the coordinates of the i-th path node in the two-dimensional coordinate system; L represents the path length of the UAV path; x i-1 and y i-1 represents the coordinates of the i-1th path node in the two-dimensional coordinate system; n represents the number of path nodes of the UAV path; P0 represents the hovering power of the UAV when it is hovering; δ represents the cross-sectional drag coefficient of the UAV, ρ represents the air density, s represents the rotor solidity of the UAV, A represents the rotor disc area of ​​the UAV, v represents the flight speed of the UAV, Ω represents the blade angular velocity of the UAV, and R represents the blade radius of the UAV; P i represents the induced power of the drone when it is hovering; k is the incremental correction factor of the induced power; W is the total weight of the drone; P fly Indicates the flight power of the drone; U tip represents the propeller tip speed, v0 is the induced speed of the UAV, and d0 is the reference flight speed; E flyall represents the flight energy consumption of the UAV; T represents the flight time of the UAV.

4. The comprehensive decision-making method for UAV trajectory planning according to claim 3 is characterized in that: The flight curvature model of the UAV includes: C i =|θ i -θ i-1 | C=max{|θ2-θ1|,|θ3-θ2|,...,|θ i -θ i-1 |,...,|θ n -θ n-1 |} Among them, C i represents the curvature of the i-th path node; C represents the curvature of the path; θ i-1 represents the direction angle of the UAV at the i-1th path node; θ i Indicates the direction angle of the UAV at the i-th path node; max indicates the maximum value.

5. The method for comprehensive decision-making of UAV trajectory planning according to claim 4, characterized in that: The target decision model of the UAV includes: C2:C=max{|θ2-θ1|,|θ3-θ2|,...,|θ i -θ i-1 |,...,|θ n -θ n-1 |}≤θ max C4:O i ≤threshold Where P represents the target decision model; h represents the target optimization problem; Min represents the minimum value; ω1, ω2 and ω3 represent weight parameters; E0 represents the initial energy of the UAV; threshold represents the dynamically adjusted parameter; θ max Indicates the preset maximum direction angle.

6. The method for comprehensive decision-making of UAV trajectory planning according to claim 5, characterized in that: The method of expanding the path nodes of the UAV by using the target decision model includes: A temporary path is constructed by selecting candidate expansion nodes from the neighboring nodes of the current path node, calculating the target optimization problem h, and selecting the candidate expansion node that minimizes the target optimization problem h as the final expansion node, wherein the neighboring nodes of the current node include: points on a circle with the current node as the center and a preset step size as the radius.

7. The method for comprehensive decision-making of UAV trajectory planning according to claim 5, characterized in that: The parameters of the updated target decision model include: threshold(d)=threshold(d-1)+reward O +reward L -penalty reward O =scale O *exp(-O) Among them, scale O 、scale L and scale are the obstacle impact reward factor, cumulative path reward factor, and penalty factor, respectively; exp represents the exponential function; threshold(d) represents the dynamic adjustment parameter during the current expansion; and threshold(d-1) represents the dynamic adjustment parameter during the previous expansion.

8. A comprehensive decision-making system for UAV trajectory planning, characterized by: The system includes a memory and a processor; the memory is used to store an application; the processor is used to run the application and execute the comprehensive decision-making method for unmanned aerial vehicle trajectory planning as described in any one of claims 1 to 7.

9. A computer storage medium, characterized in that The computer storage medium stores a remote monitoring program, which, when executed by the processor, implements a comprehensive decision-making method for UAV trajectory planning according to any one of claims 1 to 7.

Images