Unmanned aerial vehicle low-cost flight path planning method based on noise protection area

Abstract

The invention discloses an unmanned aerial vehicle low-cost flight path planning method based on a noise protection area. The method comprises the steps that a total cost objective function is established based on the sum of the total energy consumption cost and the total noise cost of the whole flight process of an unmanned aerial vehicle; and based on the total cost objective function and a preset annular noise protection area of a building, on the basis that the unmanned aerial vehicle avoids the noise protection area, planning an optimal operation path of the unmanned aerial vehicle with the minimum total cost at a preset flight height. Based on the combination of the total energy consumption cost, the total noise cost and the Dijkstra algorithm, and based on the arrangement of the noise protection area of the building, an efficient and safe operation path with the minimum noise influence and the lowest cost under the preset flight height can be planned for the unmanned aerial vehicle in the urban low-altitude airspace.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and in particular to a low-cost UAV track planning method based on a noise protection zone. Background Art

[0002] In recent years, with the development of Unmanned Aerial Vehicle (UAV) technology, Electric Vertical Take-Off and Landing Unmanned Aerial Vehicle (EVTL) has received widespread attention. It has the advantages of vertical take-off and landing, no need to rely on traditional runway take-off and landing, high maneuverability, and high efficiency. At the same time, the site used for take-off and landing occupies a small area, saving regional resources and having the potential for zero emissions in the future, which is in line with the requirements for UAV operation in urban airspace.

[0003] In urban environments, due to the large number of buildings and complex layouts, the operation of drones in such airspaces places dual requirements on efficiency and safety for path planning algorithms. The core goal of drone path planning is to find an optimal flight path from the starting point to the end point while meeting the operating specifications, ensuring that the drone can complete the designated task safely and efficiently. To address this problem, researchers have proposed a variety of algorithmic solutions, and existing solutions can generally be classified into two categories: static path planning and dynamic path planning.

[0004] Static path planning algorithms usually optimize for a pre-known and unchanging environment layout, while dynamic path planning algorithms can adapt to dynamically changing factors in the environment, such as moving obstacles, thereby providing real-time flight path adjustments for drones. There are many buildings in the city and the situation is complicated. The operation of drones in urban airspace often requires efficient path planning algorithms to improve their safety and flight efficiency. The path planning of drones refers to the drone seeking an optimal flight trajectory from the starting position to the target position under the condition that its operating requirements are met. It can ensure that the drone completes its flight mission successfully. There are many algorithms proposed for the drone path planning problem, which can be roughly divided into two categories: static path planning algorithms and dynamic path planning algorithms. Static path planning algorithms include: A* algorithm, Dijkstra algorithm, Voronoi graph method, rapid exploration random number algorithm, ant colony algorithm, particle swarm algorithm, genetic algorithm, etc.

[0005] Dynamic path planning algorithms include: ripple diffusion algorithm, artificial potential field method, D* algorithm, etc. There are many studies in the field of UAV path planning at present, and new algorithms are constantly proposed, improved and integrated. For example, some people propose to use the grid method combined with the improved A* algorithm, store the cost of the evaluation function in the grid, and use dynamic step search to plan the flight path of the UAV. There are also literatures that propose that the spatial discretization of the grid method greatly increases the number of nodes and links in the model, and proposes a ripple diffusion algorithm that comprehensively considers the performance constraints of the UAV. There are also literatures that propose new improvement principles for the traditional Voronoi diagram and combine it with the ant colony algorithm to improve the efficiency of planning paths. A public literature proposes an improved ant colony algorithm that introduces the whale algorithm to reduce the difficulty of path search and improve the convergence speed of the algorithm in response to the problem that the ant colony algorithm is prone to local convergence when planning paths. There are literatures that propose an improved gray wolf optimization algorithm that uses new strategies such as Cat mapping and introduces nonlinear convergence factors to plan paths faster and more accurately. Some literatures have proposed a new EHT-RRT algorithm based on exploration, inspiration and transfer to improve the planned path, which is effective in three-dimensional path planning. Some literatures have used the intelligent water drop algorithm combined with the fusion ant colony optimization algorithm and a new node selection strategy to improve the algorithm convergence rate and path planning efficiency. Some literatures have proposed a two-layer simulated annealing algorithm to improve the operation efficiency in response to the traffic congestion problem of urban logistics drones.

[0006] The above literature focuses on the problem of UAV path planning, focusing on comprehensively considering the performance parameters of the UAV to reduce operational risks, and improving the convergence rate and speed through algorithm improvements to enhance the efficiency and accuracy of path planning. Through these methods, the study plans the optimal operation trajectory for the UAV.

[0007] However, these studies have limitations in considering the actual impact of drone operations, especially ignoring the noise issues that may be generated by drone operations and the consideration of their operating costs. These factors are extremely critical in real-world applications. Noise issues may affect social acceptance, while cost issues are directly related to the economic feasibility of drone operations.

[0008] The above literature studies the path planning of drones, considers the performance of drones, reduces the danger of drone operation, improves the convergence rate and speed of algorithms, improves the efficiency and accuracy of path planning, and uses algorithms to plan the optimal path for drone operation, but rarely considers the impact of drone operation in reality - the noise problem generated by drones and the cost of operation. Whether the noise problem of drones meets the needs of cities and the recognition of the public is one of the main issues. At present, people are paying more and more attention to the noise problem of drones, but there are few studies on related aspects. Most of them simply include the noise problem in the cost of drone operation and combine it with other costs for research. Summary of the invention

[0009] The purpose of the present invention is to overcome the shortcomings and defects of the prior art and to provide a low-cost trajectory planning method for UAVs based on noise protection zones. The concept of noise protection zones is proposed, and the noise impact of UAVs can be reduced by adding noise protection zones to buildings, thereby optimizing the noise problem of UAVs while ensuring the safety of UAV operation. In addition, a planning method combining the Dijkstra algorithm with the improved Dubins algorithm geometric path planning is used, and a total cost objective function is introduced to achieve the planning of the optimal operation path of the UAV with the lowest cost and lowest noise.

[0010] A low-cost trajectory planning method for an unmanned aerial vehicle based on a noise protection zone comprises the following steps:

[0011] The total cost objective function is established based on the sum of the total energy consumption cost and the total noise cost of the entire flight process of the UAV;

[0012] Based on the total cost objective function and the preset annular noise protection zone of the building, an optimal operation path of the drone with the minimum total cost at a preset flight altitude is planned on the basis that the drone avoids the noise protection zone.

[0013] The noise protection zone is an annular area set outside the building below the drone flight path, located between the circumscribed circle of the building and the concentric outer circle outside the circumscribed circle;

[0014] When a drone flies outside the noise protection zone, it has no noise impact on the buildings within the noise protection zone. When it is on the outer circle of the noise protection zone, it begins to have noise impact on the buildings. The noise impact of the drone on the buildings increases with the depth of the noise protection zone.

[0015] The width of the annular zone of the noise protection zone is determined according to the noise range / distance at which the noise of the drone affects the building.

[0016] The width of the annular zone of the noise protection zone is determined according to the following formula:

[0017]

[0018] Where S represents the width of the annular noise protection zone, T mi is the thrust of each motor of the drone, A is the propeller disc area of ​​the drone, θ is the angle between the drone and the observation point, M t is the blade tip Mach number, P h is the absorbed power, B is the number of propeller blades, R' is the propeller radius, P0 is the reference sound pressure, m is the harmonic order, P SPL represents the sound pressure level, and n represents the number of motors in the drone.

[0019] The expression of the total cost objective function is as follows:

[0020] C=δ1ln(E N )+δ2ln(S P ), where C represents the total cost objective function, S P is the total noise cost of the drone during the entire flight process, E N It is the total energy consumption cost of the UAV throughout the entire flight process, δ1 and δ2 represent weight coefficients respectively, and ln is a function.

[0021] Among them, the total noise cost of the UAV during the entire flight process is S P The noise cost to the people below is calculated by the following formula:

[0022]

[0023] S P，i =S O *PPL i ,

[0024] PP i L=ψ P.i.AVG S d ,

[0025] In the above formula, j represents the number of segments into which the planned UAV route is divided, and the population density of each segment is the average population density of the segment; S P，i represents the noise cost of the drone on flight segment i, S O represents the noise cost of drones to people, S d Segment i widens to both sides d s The area after s represents the horizontal distance between the drone and the person, ψ P.i.AVG represents the average population density on the drone flight segment i; PP i L represents the population on the drone flight segment i.

[0026] Among them, the total energy consumption cost E of the UAV during the entire flight process is N The expression is as follows:

[0027] E N =E U +E H +E D , where E H represents the energy consumption cost of the drone during horizontal flight, E U With E D They are the energy consumption cost of the UAV's vertical take-off and vertical descent respectively.

[0028] Among them, the method of planning the optimal operation path of the drone with the minimum total cost at a preset flight altitude based on the total cost objective function and the preset noise protection zone is to traverse and calculate the total energy consumption cost of each track among all tracks between the starting point and the end point of the drone outside the noise protection zone, and select the track with the lowest total energy consumption cost as the optimal operation path of the drone with the minimum total cost at the preset flight altitude.

[0029] Each flight path is composed of a plurality of straight segments at predetermined flight altitudes between the starting point and the end point of the UAV, and the straight segments are tangent to the outer circle of the annular noise protection zone set outside each building / building complex on the flight path of the UAV.

[0030] Among them, the Dijkstra algorithm and the improved Dubins algorithm are used for planning to plan the optimal operation path of the UAV with the minimum total cost; among them, when the flight segments of two adjacent noise protection areas intersect, the improved Dubins algorithm is used to smooth the flight segments of the two adjacent noise protection areas.

[0031] The smoothing of the flight segments of two adjacent noise protection zones includes drawing a tangent line from the starting point of the UAV across the front noise protection zone directly to the rear noise protection zone on the same side of the intersecting flight segment, and forming a smooth flight segment between the starting point and the outer circle of the rear noise protection zone, and the straight line of the smooth flight segment is not tangent to the front noise protection zone.

[0032] When planning the trajectory of the drone of the present invention, a noise protection zone is set up, and the noise impact generated by the operation of the drone is reduced through the noise protection zone model; the noise cost of the drone to the ground people is taken into account in the operation cost of the drone, so that the consideration of the drone in the trajectory planning is more comprehensive, and the impact of the noise generated by the operation of the drone is effectively reduced.

[0033] When planning the trajectory of the UAV of the present invention, a method for improving Dubins path planning is proposed in terms of enriching path selection, optimizing path smoothness, etc. The improved Dubins path planning has abundant optional paths, shortens the flight distance of the UAV, and improves the operation efficiency. At the same time, based on the total energy consumption cost, the total noise cost and the Dijkstra algorithm, and based on the setting of the noise protection zone of the building, an efficient and safe operation path with the minimum noise impact and the lowest cost at a preset flight altitude is planned for the UAV in the low-altitude airspace of the city. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The present invention is a flow chart of the low-cost trajectory planning method for unmanned aerial vehicles based on the noise protection zone.

[0035] Figure 2A , Figure 2B as well as Figure 2C It is a schematic diagram of setting a circumscribed circle on the outer periphery of a building according to the present invention.

[0036] Figure 3 The figure shows a schematic diagram of calculating the center point and radius of the circumscribed circle of a building on the XOY plane.

[0037] Figure 4 This is a schematic diagram of the altitude layer division of the drone operating airspace.

[0038] Figure 5 Schematic diagram of the width S of the noise protection zone added to the building model.

[0039] Figure 6 Schematic diagram of a drone route passing through a noise protection zone.

[0040] Figure 7 This is a schematic diagram of a UAV flying at a predetermined altitude.

[0041] Figure 8 Schematic diagram of the noise impact on people when a drone flies at a predetermined altitude.

[0042] Fig. 9 Schematic diagram of a drone's flight path around a single building noise protection zone.

[0043] Fig.10 Schematic diagram of a drone flying from location A to location B and bypassing multiple buildings.

[0044] Fig.11 Schematic diagram of the trajectory planning of a drone when it passes through two buildings from the starting point to the end point.

[0045] Fig.12 Schematic diagram of smoothing the selected track line.

[0046] Fig.13 Schematic diagram of the distribution of buildings under the drone airspace during the simulation test.

[0047] Fig.14A , Fig. 14B The comparison between the noise-free minimum cost path of the present invention and the best path obtained by the A* algorithm among all the planned paths Figure 1 .

[0048] Fig.15A , Fig. 15B FIG2 is a comparison diagram of the noise-free minimum cost path of the present invention and the best path obtained according to the A* algorithm among all planned paths.

[0049] Fig.16A , Fig. 16B The comparison between the noise-free minimum cost path of the present invention and the best path obtained by the A* algorithm among all the planned paths Figure 3 .

[0050] Fig.17A , Fig. 17B The comparison between the noise-free minimum cost path of the present invention and the best path obtained by the A* algorithm among all the planned paths Figure 4 . DETAILED DESCRIPTION

[0051] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] In view of the noise problem of drones mentioned in the background technology, the present invention proposes a noise protection zone. By planning the drone path according to the noise protection zone, low noise impact or even noise-free operation of drones can be achieved. In addition, since cost determines the economic efficiency of drone operation, the present invention proposes a total cost objective function, which considers the energy consumption cost and the noise cost of low altitude cities to people, so as to evaluate the total cost of the planned path.

[0053] Traditional Dubins geometric path planning is to continuously link tangents between two building models to seek the optimal path from the starting point to the target point. The present invention improves the traditional Dubins, constructs all tangents between each building model, and uses an improved path optimization method to optimize the tangents that do not conform to the actual situation, making the route more efficient and smooth.

[0054] In the present invention, the UAV refers to a vertical take-off and landing UAV, and its trajectory planning means that when the UAV needs to go from one point to another in the city to perform a task, the starting point and the end point are known, and a vertical take-off and landing UAV is used to plan an operation path for it to meet the operation requirements of the UAV. Among them, the maximum operating altitude of the UAV does not exceed 120m; the UAV vertical take-off and landing stage only considers the energy consumption cost.

[0055] Please refer to Figure 1 As shown, the low-cost trajectory planning method for a UAV based on a noise protection zone in an embodiment of the present invention includes the steps of: establishing a total cost objective function based on the sum of the total energy consumption cost and the total noise cost of the UAV throughout the entire flight process; based on the total cost objective function and a preset annular noise protection zone of a building, planning an optimal operation path for the UAV with the minimum total cost at a preset flight altitude on the basis of the UAV avoiding the noise protection zone.

[0056] Among them, the noise protection zone is an annular area set up outside the building below the drone's flight path, located between the circumscribed circle of the building and the concentric outer circle outside the circumscribed circle; when the drone flies outside the noise protection zone, it has no noise impact on the buildings within the noise protection zone, and when it is on the outer circle of the noise protection zone, it begins to have noise impact on the buildings. The noise caused by the drone on the building increases with the depth of the drone entering the noise protection zone.

[0057] The width of the annular zone of the noise protection zone is determined according to the noise range / distance at which the noise of the drone affects the building.

[0058] In this application, when establishing a noise protection zone, it is achieved by modeling the city where the drone is operating according to the parameters of the buildings: first, from the top-down angle of the building, a circumscribed circle is added to the periphery, the center of the circumscribed circle is the center of the building, and the distance from the midpoint to the farthest point on the boundary of the building is the radius of the circumscribed circle, see Figure 2A , Figure 2B as well as Figure 2C As shown, Figure 2A , Figure 2B The circumscribed circle set outside a single building of different shapes is shown. Since the location, height and other parameters of the buildings in the city are known, the styles are diverse, and the shapes of different buildings are also very different. In order to plan the path more efficiently, a circumscribed circle is added to the periphery of each building from a bird's-eye view. The center of the circumscribed circle is the center of the building, and the distance from the center point to the farthest point on the boundary of the building is the radius of the circumscribed circle. This model can effectively avoid the risk of conflict between drones and buildings. It should be noted that in some cases there are building clusters in the city, where the distances between the buildings are very close. This small group of buildings can be regarded as a whole. See Figure 2CAs shown in FIG. 1 , the center of the building complex area (including three rectangular buildings in a top view) is used as the center of the circle for modeling. The modeled building model is as follows: Figure 2A , Figure 2B as well as Figure 2C As shown, the aforementioned building model is brought into the coordinate system, and the model on the XOY plane is as follows Figure 3 As shown, the center and radius of the building's circumscribed circle can be obtained. Figure 3 As shown, quadrilateral A1A2A3A4 is the projection of a building or a group of buildings on the XOY plane, and each point corresponds to a corresponding coordinate, such as Figure 3 As shown, circle O0 is the center of the circumscribed circle of a building or a group of buildings, r is the radius of the circumscribed circle of a building or a group of buildings, and the center coordinates (X0, Y0) and radius r can be obtained based on the coordinates of the four vertices of the building or the group of buildings. The noise protection zone is a concentric outer circle arranged outside the circumscribed circle, and the annular area between the circumscribed circle and the outer circle is the noise protection zone of the building. When planning a path or track, the drone at a predetermined flight altitude is made to fly outside the noise protection zone.

[0059] Since drones fly at a preset altitude, when planning the route, it is necessary to divide the urban airspace environment in which the drones operate, assign operating altitude layers to the drones, and plan their flight paths according to the drones' operating altitudes.

[0060] Among them, the reasonable allocation of operating altitude layers for drones can effectively improve the working efficiency of drones and reduce the operating costs of drones. Specifically, by obtaining the heights of all buildings in the drone operating airspace, the altitude layers are reasonably divided, and the height interval between two altitude layers is a height range. The division is as follows: Figure 4 shown.

[0061] See also Figure 4 As shown, when the UAV flies at different altitudes within the same altitude range, the positions of the buildings within the altitude range remain unchanged, and the planned path remains unchanged. Figure 4 The airspace where drones can operate is divided into four altitude levels: h1, h2, h3, and h4. The altitude level h0 in the figure indicates the highest altitude at which drones cause noise impact to pedestrians on the ground. When drones operate below the altitude h0, they will cause noise impact to pedestrians on the ground. When drones operate above the altitude h0, they will no longer cause noise impact to pedestrians on the ground. The altitudes corresponding to h1, h2, h3, and h4 can be specifically divided according to different cities and regions.

[0062] In the present application, the noise range (distance) of the drone noise affecting the building, that is, the width S of the noise protection zone, is calculated and then the width S of the noise protection zone is added to the building model. Then, based on the drone avoiding the noise protection zone, the optimal operation path of the drone with the lowest total cost at a preset flight altitude is planned based on the present application.

[0063] The invention proposes a noise protection zone and adds a noise protection zone to the building, which can greatly reduce the noise impact of drone operation on the building. The noise protection zone can enable drones to operate noiselessly and at low cost above a certain height, and to operate at low noise and low cost below a certain height, providing protection for the operation of drones in the city.

[0064] Among them, the drone noise in the present invention mainly considers the drone noise as the running rotation noise.

[0065] The rotation noise P of a single motor propeller of a drone rel It is expressed as:

[0066]

[0067] Where, ρ0 = 1.225 kg / m 3 is the air density at standard atmospheric sea level, B represents the number of propeller blades of the drone, R` represents the propeller radius of the drone, M t is the blade tip Mach number, P h To absorb power, m is the harmonic order, P SPL represents the sound pressure level, P0 is the reference sound pressure, S represents the width of the annular noise protection zone (i.e., the difference between the radius of the circumscribed circle and the radius of the outer circle), A represents the propeller disc area of ​​the drone, and T mi is the thrust of each motor of the UAV, and θ represents the angle between the UAV and the observation point.

[0068] Therefore, in this application, the noise range (distance) of the drone noise affecting the building, that is, the width S of the noise protection zone is calculated as follows:

[0069]

[0070] In the formula, n represents the number of motors of the drone;

[0071] Using the Burgers thrust model, the thrust of each motor of the drone is obtained as T mi :

[0072]

[0073] V tip=Π*N*d (5)

[0074] Where V ∞ Indicates the airspeed of the drone, V tip represents the propeller tip speed of the drone, N represents the number of revolutions per second of the propeller, d represents the propeller diameter, S b represents the total area of ​​propeller blades, η represents the normalized thrust ratio of work to kinetic energy, ρ represents the air density, and Π represents the calculation model for calculating the wingtip speed in the thrust model.

[0075] Add the noise protection zone width S to the building model, such as Figure 5 As shown in the figure, the center of the building model is the center of the building, point O, r is the radius from point O to the circumcircle of the building, and R is the radius from point O to the outer circle of the protection zone. The annular area from the circumcircle of the building to the outer circle of the protection zone is the added building noise protection zone. When the drone flies outside the protection zone, it has no noise impact on the building. When it is on the outer circle of the noise protection zone, it starts to have noise impact on the building. Moreover, the noise increases with the depth of the protection zone. The conflict diagram is shown in the figure below. Figure 6 As shown, Figure 6 In the figure, the straight line with an arrow passing through the annular area is the flight track of the UAV.

[0076] The energy consumption of drones during operation is one of the costs that need to be considered. During the operation, the operation of drones is divided into three stages: horizontal operation stage, vertical take-off stage and vertical landing stage. The output power and energy consumption of the drone battery are different in different flight stages.

[0077] When the drone is in the vertical take-off and landing stage, the thrust generated by the motor is mainly used to overcome its own gravity. The output power P of the battery in this stage L for:

[0078]

[0079] Where W P , W M , W ESC They represent the working efficiency of the propeller, motor and ESC of the drone respectively. M represents the total weight of the drone. g represents the acceleration of gravity. τ is the correction factor.

[0080] When the drone is in the horizontal operation stage, the thrust generated by the motor is mainly to overcome the air resistance. The output power P of the battery in this horizontal flight stage H for:

[0081]

[0082] Where σ represents the frontal area, V Hrepresents the horizontal velocity, C D0 represents the zero lift drag coefficient and k represents the induced drag factor.

[0083] The battery output power of the drone during operation is related to the air density. When the drone is operating at an altitude of H1, the flight altitude is H2, and the operating distance of the drone is L, Figure 7 As shown, the energy consumption E of the drone during horizontal flight H The expression is:

[0084]

[0085] The air density is:

[0086]

[0087] ρ0 represents the air density under standard atmospheric conditions, T0 represents the temperature under standard atmospheric conditions, ζ represents the air gas constant, K T Represents the temperature gradient below the top of the troposphere; the air density in different urban areas and at different altitudes varies accordingly. According to the BADA manual, the air density and temperature at the standard atmospheric sea level are ρ0 = 1.225 kg / m 3 , T0=15℃ / 288.15K. The air density and temperature T decrease with the increase of altitude. For every 100 meters increase in altitude, the temperature drops by 0.6℃. The expression of temperature T decreasing with the increase of altitude is T=T0-0.6(H+H a ), where H a represents the altitude, H represents the operating altitude of the drone, and the expression for the change in air density is:

[0088] The speed of the drone when taking off vertically is V U , the landing speed is V D , and the energy consumption E for vertical takeoff and descent is obtained U With E D The expression is:

[0089]

[0090]

[0091] In the above formula, ρ U , ρ D They represent the air density when the UAV takes off vertically and the air density when the UAV lands vertically, and t represents the flight time of the UAV.

[0092] Therefore, the total energy consumption cost E of the UAV proposed in this application during the entire operation process is N for:

[0093] E N =E U +E H +E D (14)

[0094] The noise cost model considers the noise impact on people on the ground when the drone is operating at a low altitude. Therefore, when the drone is executing the flight plan, the noise cost must also be considered. The noise impact of the drone will decrease as the operating altitude increases. After reaching a certain altitude, its impact on people on the ground will not be considered. When the drone flies at a low altitude, especially when passing through crowded areas such as public places, the noise of the drone will have a greater impact on people on the ground. The noise impact of drones on people is as follows: Figure 8 shown. Figure 8 h is the height of the drone above the person's head, d s The horizontal distance between a person and a drone.

[0095] Sound propagation is approximately spherical diffusion, so the noise cost of drone operation to people is S O It is expressed as:

[0096]

[0097] In the formula, ω represents the conversion coefficient from sound intensity to sound level, S C Represents the reference noise of the drone.

[0098] Population density is a key factor in measuring the noise impact of drones on people. Assuming that the center of the area where the planned path is located is the area with the highest population density (busy area), the population density gradually decreases from the center to the outside in a circle. In order to better evaluate the population density distribution in the urban area, the gravity model is used, and the expression is as follows:

[0099]

[0100] In the above formula, ψ P represents the population density in the area with a radius of r1 from the city center, r1 represents the distance from the city center, ψ P.AVG The average population density within the entire urban area.

[0101] In the subsequent calculation, the planned route is divided into j segments, and the population density of each segment is the average population density of the segment, so the total noise cost is:

[0102] PPL i =ψ P.i.AVG S d (17)

[0103] S P，i =S O *PPLi (18)

[0104]

[0105] In the formula, j represents the number of segments into which the planned UAV route is divided, and the population density of each segment is the average population density of the segment; S P，i represents the noise cost of the drone on flight segment i, S O represents the noise cost of drones to people, S d Segment i widens to both sides d s The area after s represents the horizontal distance between the drone and the person, ψ P.i.AVG represents the average population density on the drone flight segment i; PP i L represents the population on the drone flight segment i.

[0106] In the present invention, the total cost objective function C is proposed, which is the sum of the energy consumption cost and the noise cost of the UAV operation. The total cost objective function expression is:

[0107] C=δ1ln(E N )+δ2ln(S P ) (20)

[0108] The present invention is based on the noise protection zone and combines Dubins path planning with Dijkstra algorithm to plan the optimal operation path of the UAV with the minimum total cost C.

[0109] Dubins path planning is a widely applicable path planning method, and the planned path has the advantages of being smooth and efficient. Dijkstra algorithm is a widely used method that can find the shortest path in an area based on the lowest cost, and can efficiently find the lowest cost path from the starting point to any node in the area. The present invention efficiently combines Dijkstra algorithm with Dubins path planning, and can plan a path with no (low) noise impact and the lowest cost for the drone at each altitude.

[0110] Based on the noise-free operation of drones, it is necessary to avoid the circular area of ​​the noise protection zone. Using the idea of ​​Dubins path planning, the drone path is composed of the tangent and arc of the circle. Under the condition of ensuring safety and noise-free, a drone operation path with the lowest cost and shortest distance is obtained. The operation of drones needs to avoid passing through the noise protection zone as much as possible. Therefore, the drone bypasses the noise protection zone of a single building in the following ways: Fig. 9 As shown: Fig. 9 A schematic diagram of the drone's path is shown. Fig. 9The middle circle CEFD is the outer circle of the protection zone, with point O as the center, point A as the starting point, and point B as the target point. Flying from point A to point B requires bypassing the noise protection zone of the building. There are two ways to make a tangent from a point to a circle, that is, there are two paths from point A to point B, namely tangent AC→arc CE→tangent EB and tangent AD→arc DF→tangent FB. The line segments from the center of the circle to each tangent point form a certain angle, OC and OE form an angle α, and OD and OF form an angle β.

[0111] Assuming that the distance from point A to the center O is L1, the distance from point B to the center O is L2, and the straight line length of the AB path is L3, the calculation model or expression for each path length is:

[0112] The distance from point A to point B via points C and E is:

[0113]

[0114] The distance from point A to point B via points D and F is:

[0115]

[0116] The above method is only a way for a drone to bypass a single building. In an urban environment, there are many buildings. A schematic diagram of bypassing multiple buildings when flying from location A to location B (only the optimal tangent path is retained) is shown as follows: Fig.10 shown. Fig.10 Figure 2 shows a schematic diagram of Dubins path planning. Fig.10 There are three buildings in the figure, whose centers are O1, O2, and O3 respectively. There are two tangent points on the outer circles, namely C, D, E, F, G, and H. Multiple line segments are connected to form the planned flight path from location A to location B, bypassing multiple buildings. The traditional Dubins method plans the path based on the shortest distance of the UAV flight. If the operating cost of the UAV is considered, it needs to be improved. For this purpose, the present invention uses a geometric method to draw two tangent lines from a point to the circle. The two circles (see Fig.11 As shown in the figure, the centers are O1 and O2 respectively. There are four common tangents between them. The two ends of each tangent are regarded as two nodes. The circle is the noise protection zone, and the tangent is the flight path of the drone. The tangent paths between the noise protection zones of all buildings are found. According to the cost model, the cost information of each path is obtained. Finally, the Dijkstra algorithm is used to find the best path from the starting point through each node to the target point according to the cost information. The improved Dubins path planning method of this application has many nodes and paths, which greatly enriches the optionality of the path. The planning diagram is shown in the figure below. Fig.11 shown. Fig.11 The improved Dubins path planning diagram is shown. Fig.11This is just a simple example. A total of four optional paths are planned from the starting point to the end point (as the number of buildings increases, the number of optional paths gradually increases).

[0117] In some specific cases, the tangents between the two protection zone models may intersect, which makes the UAV's running path not smooth and increases the running cost, so it needs to be smoothed. For the tangents drawn by two different protection zone models (the tangents are on the same side), if the tangents do not intersect, the Dubins idea can be used to plan the path normally. If the tangents intersect, they are smoothed, that is, in the direction of that side, the tangent can be drawn directly from the starting point across the protection zone to the next protection zone, making the path smoother, shortening the UAV's running distance, and reducing the UAV's running cost. The smoothing diagram is shown below. Fig.12 shown. Fig.12 FIG. 4 shows a schematic diagram of improved Dubins path smoothing processing. As shown by the red line in the figure, the line segments where the two boxes intersect in the above figure are smoothed to form a line segment.

[0118] Simulation test

[0119] Parameter settings

[0120] The model parameters of the noise protection zone are shown in Table 1. UAVs are operating in the air, and buildings have a certain isolation effect on noise. Therefore, the noise impact of UAVs in the air is set to 120 dB, and the weight coefficients in the total cost C are δ1 = 0.75 and δ2 = 0.25:

[0121] Table 1 Noise protection zone model parameters

[0122]

[0123] Unit conversion of parameters in Table 1: 1ft = 0.3048m, 1dyn / cm 2 =0.1Pa, 1lb=4.4482N.

[0124] The energy consumption cost model parameters are shown in Table 2:

[0125] Table 2 Energy consumption cost model parameters

[0126]

[0127] The noise cost model parameters are shown in Table 3:

[0128] Table 3 Noise cost model parameters

[0129]

[0130] If the altitude exceeds a certain level, there will be no noise impact on the people on the ground. In this application, the noise cost is defined as 50dB. According to relevant literature, the flight altitude corresponding to the noise cost is 28m.

[0131] Select an area in a city with a range of 800m×800m, a starting point coordinate of (43, 724), and an end point coordinate of (744, 67). Model the buildings in the area (no noise protection zone is added to the building models in the figure) and label them, such as Fig.13 shown. Fig.13 A schematic diagram of building distribution is shown, in which each circle represents the circumscribed circle of a building, which are numbered 1-18 in total. The flight track of the drone is from the origin to the end point, passing through the building groups or areas where the buildings are located, which are numbered 1-18 in total.

[0132] Among them, the parameter information of the building is shown in Table 4:

[0133] Table 4 Building parameters

[0134]

[0135] Path Planning

[0136] According to the flight altitude layer division of the airspace, it is divided into four altitude layers: 20m, 40m, 60m, and 80m, corresponding to the four altitude intervals of (20m, 40m), (40m, 60m), (60m, 80m), and above 80m. The Dijkstra algorithm and Dubins path planning algorithm are used to plan the path of the UAV at each altitude layer.

[0137] In the following description, the red lines in the corresponding figures are the flight trajectory lines finally optimized according to the method of the present invention and the best path planning obtained according to the A* algorithm.

[0138] (1) When the flight altitude is 20m, remove the buildings below this altitude and keep the buildings numbered 1, 5, 6, 7, 10, 11, 13, 15, 16, and 18. When the flight altitude is 20m, the noise protection zone is S = 8m, S O =72dB, in the building circumference plus the protection zone, the noise-free minimum cost path of the present invention is compared with the optimal path planning result obtained by the A* algorithm among all the planned paths. Fig.14A , Fig. 14B As shown:

[0139] Fig.14A , Fig. 14BThe comparison chart of the path planning results at the 20m altitude layer is shown. The 20m altitude layer is lower than the altitude threshold. At this altitude layer, the total cost considers energy consumption cost and noise cost. The noise cost S of the optimal path calculated according to the method of the present invention is P =22896dB, energy cost E N =1641.088J, total cost C = 9.79. Noise cost S obtained by A* algorithm P =25416dB, energy cost E N =1880.695J, the total cost is C=9.92.

[0140] (2) When the flight altitude is 40m, remove the buildings below this altitude and keep the buildings numbered 1, 7, 11, 13, 16, and 18. When the flight altitude is 40m, the noise protection zone S=17m is calculated according to the method of the present invention. The noise-free minimum cost path among all the planned paths is compared with the optimal path planning result obtained by the A* algorithm, with the protection zone added to the circumscribed circle of the building. Fig.15A , Fig. 15B As shown:

[0141] Fig.15A , Fig. 15B The comparison chart of path planning results at 40m altitude is shown. There is no noise cost impact at 40m altitude, so the total cost only considers the energy cost. The energy cost E of the optimal path calculated according to the method of the present invention is N =21280.96J, total cost C = 7.47. Energy consumption cost E obtained by A* algorithm N =23377.33J, the total cost is C=7.55.

[0142] (3) When the flight altitude is 60m, remove the buildings below this altitude and keep the buildings numbered 1, 7, 11, 13, 16, and 18. When the flight altitude is 60m, the noise protection zone S=26m is calculated according to the method of the present invention. The noise-free minimum cost path among all planned paths is compared with the optimal path planning result obtained by the A* algorithm, with the building circumscribed circle plus the protection zone. Fig.16A , Fig. 16B As shown:

[0143] Fig.16A , Fig. 16B The comparison chart of the path planning results at the 60m altitude layer is shown. There is no noise cost impact at the 60m altitude layer, so the total cost only considers the energy cost. The energy cost of the optimal path is calculated according to the method of the present invention, E N =26374.79J Total cost C = 7.64. Energy consumption cost E obtained by A* algorithm N =29012.33J, the total cost is C=7.71.

[0144] (4) When the flight altitude is 80m, remove the buildings below this altitude and keep the buildings numbered 7, 11, 16, and 18. When the flight altitude is 80m, the noise protection zone S=34m is calculated according to the method of the present invention. The noise-free minimum cost path among all the planned paths is compared with the optimal path planning result obtained by the A* algorithm, with the protection zone added to the circumscribed circle of the building. Fig.17A , Fig. 17B As shown:

[0145] Fig.17A , Fig. 17B The comparison chart of path planning results at 80m altitude is shown. There is no noise cost impact at 80m altitude, so the total cost only considers the energy cost. The energy cost E of the optimal path calculated by the method of the present invention is N =31324.97J, total cost C = 7.76. Energy consumption cost E obtained by A* algorithm N =34404.99J, the total cost is C = 7.84. The selection of the optimal path at different altitudes according to the Dijkstra algorithm and the resulting costs are shown in Table 5 below:

[0146] Table 5 Dijkstra algorithm calculation results

[0147] Flight altitude layer / m Horizontal flight power / W Noise costs Total energy consumption / J Energy cost Total Cost 20m 184.66 2.51 16410.88 7.28 9.79 40m 188.1 0 21280.96 7.47 7.47 60m 191.79 0 26374.79 7.64 7.64 80m 195.74 0 31324.97 7.76 7.76

[0148] The selection of the optimal path at different altitudes according to the A* algorithm and the resulting costs are shown in Table 6 below:

[0149] Table 6 Calculation results of A* algorithm

[0150] Flight altitude layer / m Horizontal flight power / W Noise costs Total energy consumption / J Energy cost Total Cost 20m 184.66 2.54 18806.95 7.38 9.92 40m 188.1 0 23377.33 7.55 7.55 60m 191.79 0 29012.33 7.71 7.71 80m 195.74 0 34404.99 7.84 7.84

[0151] According to the comparison of the data in Table 5 and Table 6, it can be concluded that the algorithm proposed in the present invention has a lower cost than the A* algorithm. The optimal altitude layer selected for the operation of the UAV is 40m, and the minimum cost C of the path at this altitude layer is 7.47.

[0152] Simulation studies have shown that the technology or method proposed in the present invention can achieve low-cost operation at high altitudes without noise in urban airspace, and low-cost operation at low altitudes with short flight time and high efficiency. Among them, when the weights of the total cost target cost function are 0.75 and 0.25, the optimal operating altitude of the drone is 40m, with no noise impact and the lowest cost.

[0153] In summary, it can be seen that the technology or method proposed in the present invention proposes a noise protection zone, obtains data through the noise protection zone model, and minimizes the noise impact of the drone on the environment and people. The technology or method proposed in the present invention improves the Dubins path planning method, so that the comprehensive cost model can be used for path planning, adding nodes, optimizing the smoothness of the path, shortening the flight path, and realizing low-cost operation. In addition, compared with the results of the A* algorithm, the advantages of this article's comprehensive search path and lower operating cost are highlighted. The technology or method proposed in the present invention enriches the cost objective function considerations, making the consideration of the drone operating cost more comprehensive and more in line with the actual situation. The technology or method proposed in the present invention not only considers the impact of drones based on noise protection zones on the surrounding environment, but also considers the noise impact of drones on the ground when operating at low altitudes in the cost model, making the research on noise impact more comprehensive.

[0154] In summary, the technology or method proposed in the present invention is a drone operation path planning method that is more in line with actual needs, safer and more efficient. It significantly reduces the noise impact caused by drones flying in urban environments and effectively reduces their operating costs, providing innovative solutions and support for the application of drones in complex urban spaces.

[0155] The basic principles and main features of the present invention and the advantages of the present invention are shown and described above. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention.

[0156] Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the claims are included in the present invention.

[0157] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A low-cost trajectory planning method for UAVs based on noise protection areas, characterized in that: Includes steps: The total cost objective function is established based on the sum of the total energy consumption cost and the total noise cost of the entire flight process of the UAV; Based on the total cost objective function and the preset annular noise protection zone of the building, an optimal operation path of the drone with the minimum total cost at a preset flight altitude is planned on the basis that the drone avoids the noise protection zone.

2. According to claim 1, the low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas is characterized in that: The noise protection zone is an annular area set outside the building below the drone flight path area, located between the circumscribed circle of the building and the concentric outer circle outside the circumscribed circle; When a drone flies outside the noise protection zone, it has no noise impact on the buildings within the noise protection zone. When it is on the outer circle of the noise protection zone, it begins to have noise impact on the buildings. The noise impact of the drone on the buildings increases with the depth of the noise protection zone.

3. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 1 is characterized in that: The width of the annular zone of the noise protection zone is determined according to the noise range / distance at which the noise of the drone affects the building.

4. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 3 is characterized in that: The width of the annular zone of the noise protection zone is determined according to the following formula: Where S represents the width of the annular noise protection zone, T mi is the thrust of each motor of the drone, A is the propeller disc area of ​​the drone, θ is the angle between the drone and the observation point, M t is the blade tip Mach number, P h is the absorbed power, B is the number of propeller blades, R' is the propeller radius, P0 is the reference sound pressure, m is the harmonic order, P SPL represents the sound pressure level, and n represents the number of motors in the drone.

5. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 1 is characterized in that: The expression of the total cost objective function is as follows: C=δ1ln(E N )+δ2ln(S P ), where C represents the total cost objective function, S P is the total noise cost of the drone during the entire flight process, E N It is the total energy consumption cost of the UAV throughout the entire flight process, δ1 and δ2 represent weight coefficients respectively, and ln is a function.

6. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 5 is characterized in that: The total noise cost of the drone during the entire flight process is S P The noise cost to the people below is calculated by the following formula: S P，i =S O *PPL i , PP i L=ψ P.i.AVG S d , In the above formula, j represents the number of segments into which the planned UAV route is divided, and the population density of each segment is the average population density of the segment; S P，i represents the noise cost of the drone on flight segment i, S O represents the noise cost of drones to people, S d Segment i widens to both sides d s The area after s represents the horizontal distance between the drone and the person, ψ P.i.AVG represents the average population density on the drone flight segment i; PP i L represents the population on the drone flight segment i.

7. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 5 is characterized in that: The total energy consumption cost of the drone during the entire flight process is E N The expression is as follows: E N =E U +E H +E D , where E H represents the energy consumption cost of the drone during horizontal flight, E U With E D They are the energy consumption cost of the UAV's vertical take-off and vertical descent respectively.

8. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection zones according to any one of claims 1 to 7, characterized in that: The method of planning the optimal operation path of the UAV with the minimum total cost at a preset flight altitude based on the total cost objective function and the preset noise protection zone is to traverse and calculate the total energy consumption cost of each track among all tracks between the starting point and the end point of the UAV outside the noise protection zone, and select the track with the lowest total energy consumption cost as the optimal operation path of the UAV with the minimum total cost at the preset flight altitude. Each flight path is composed of a plurality of straight segments at predetermined flight altitudes between the starting point and the end point of the UAV, and the straight segments are tangent to the outer circle of the annular noise protection zone set outside each building / building complex on the flight path of the UAV.

9. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 8 is characterized in that: The Dijkstra algorithm and the improved Dubins algorithm are used for planning to plan the optimal operation path of the UAV with the minimum total cost; among them, when the flight segments of two adjacent noise protection zones intersect, the improved Dubins algorithm is used to smooth the flight segments of the two adjacent noise protection zones.

10. The low-cost trajectory planning method for unmanned aerial vehicles based on noise protection areas according to claim 9 is characterized in that: The smoothing process of the flight segments of two adjacent noise protection zones includes drawing a tangent line from the starting point of the UAV across the front noise protection zone directly to the rear noise protection zone on the same side of the intersecting flight segment, and forming a smooth flight segment between the starting point and the outer circle of the rear noise protection zone, wherein the straight line of the smooth flight segment is not tangent to the front noise protection zone.

Images