A Multi-UCAV Cooperative Real-Time Trajectory Planning Method

Abstract

The present invention provides a method for collaborative real-time trajectory planning of multiple UCAVs, which includes the following steps: constructing a battlefield environment, including three-dimensional terrain and threat range; establishing dynamic models of each UCAV under flight altitude constraint conditions, flight speed constraint conditions, flight threat constraint conditions, and flight anti-collision constraint conditions, and obtaining the control quantities of each UCAV; establishing an objective function for collaborative real-time trajectory planning of multiple UCAVs; and determining the collaborative real-time trajectory of multiple UCAVs based on the I-MRFO algorithm. The method for collaborative real-time trajectory planning of multiple UCAVs provided by the present invention has the characteristics of very good collaboration and real-time performance, strong adaptability, fast response, and high accuracy, and can be widely applied to the field of multi-aircraft collaboration.

Description

Technical Field

[0001] The present invention relates to the technology of trajectory planning, and particularly to a method for collaborative real-time trajectory planning of multiple UCAVs. Background Art

[0002] At present, collaborative trajectory planning of multiple UCAVs (Unmanned Combat Aerial Vehicles) has become the core and hot topic of UAV combat flight. For example, tasks of multiple UCAVs attacking multiple targets and real-time trajectory planning, collaborative trajectory planning of multiple UCAVs based on biogeography optimization, collaborative route planning algorithms of multiple UCAVs, collaborative UCAV trajectory planning for jamming suppression optimized based on genetic algorithms, and so on. These trajectory planning methods have planned the collaborative trajectories of multiple UCAVs from various different perspectives. However, they all have problems such as slow response, low accuracy, and poor stability.

[0003] In the prior art, there are problems such as slow response, low accuracy, and poor stability in the collaborative trajectory planning of multiple UCAVs. Summary of the Invention

[0004] In view of this, the main object of the present invention is to provide a method for collaborative real-time trajectory planning of multiple UCAVs with good stability, fast response, high accuracy, and high survival rate.

[0005] To achieve the above object, the technical solution proposed by the present invention is as follows:

[0006] A method for collaborative real-time trajectory planning of multiple UCAVs includes the following steps:

[0007] Step 1, construct a battlefield environment, including: three-dimensional terrain, threat range.

[0008] Step 2, establish dynamic models of each UCAV under flight altitude constraint conditions, flight speed constraint conditions, flight threat constraint conditions, and flight anti-collision constraint conditions, and obtain the control amounts of each UCAV.

[0009] Step 3, establish an objective function for collaborative real-time trajectory planning of multiple UCAVs.

[0010] Step 4, determine the collaborative real-time trajectory of multiple UCAVs based on the I-MRFO algorithm.

[0011] In summary, for the multi-UCAV cooperative real-time trajectory planning method of the present invention, first, a UCAV dynamics model is established according to the battlefield environment and necessary constraint conditions, and the flight control quantities of each UCAV are obtained based on this dynamics model; secondly, in order to reach the target point quickly and smoothly, a multi-UCAV cooperative real-time trajectory planning objective function is established, and under this objective function, based on the I-MRFO algorithm, the current flight position of the UCAV and the estimated flight position of the UCAV at the next moment are determined. The multi-UCAV cooperative real-time trajectory planning method of the present invention adopts an improved manta ray foraging optimization (I-MRFO) algorithm, and uses the control quantities determined by the UCAV dynamics model as input quantities, intermediate quantities, and output quantities, so that the cooperation and real-time performance between each UCAV are very good, the adaptability is very strong, the response is fast, and the accuracy is high, meeting the limitations of various battlefield situations and its own characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic diagram of the overall process of the multi-UCAV cooperative real-time trajectory planning method of the present invention.

[0013] Figure 2 It is a schematic diagram of the first risk avoidance method of UCAV under the flight threat constraint conditions of the present invention.

[0014] Figure 3 It is a schematic diagram of the second risk avoidance method of UCAV under the flight threat constraint conditions of the present invention.

[0015] Figure 4 It is a schematic diagram of the third risk avoidance method of UCAV under the flight threat constraint conditions of the present invention.

[0016] Figure 5 It is a schematic diagram of the relative distance of each UCAV during flight in the embodiment of the present invention.

[0017] Figure 6 It is a schematic diagram of the change in the height difference between each UCAV and the ground in the embodiment of the present invention.

[0018] Figure 7 It is a schematic diagram of the change in the height of each UCAV with the ground during flight in the embodiment of the present invention.

[0019] Figure 8 It is a schematic diagram of the change in the flight speed of each UCAV in the embodiment of the present invention.

[0020] Figure 9 It is a schematic diagram of the time-consuming situation of real-time trajectory planning in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] To make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Figure 1 This is a schematic diagram of the overall process of the multi-UCAV collaborative real-time trajectory planning method described in the present invention. As Figure 1 shown, a multi-UCAV (Unmanned Combat Aerial Vehicle) collaborative real-time trajectory planning method described in the present invention includes the following steps:

[0023] Step 1: Construct the battlefield environment, including: three-dimensional terrain, threat range.

[0024] Step 2: Establish the dynamic models of each UCAV under the flight altitude constraint condition, flight speed constraint condition, flight threat constraint condition, and flight anti-collision constraint condition, and obtain the control quantities of each UCAV.

[0025] Step 3: Establish the multi-UCAV collaborative real-time trajectory planning objective function.

[0026] Step 4: Determine the multi-UCAV collaborative real-time trajectory based on the I-MRFO (Improve Manta Ray Foraging Optimization) algorithm.

[0027] In summary, the multi-UCAV collaborative real-time trajectory planning method described in the present invention first establishes the dynamic model of the UCAV according to the battlefield environment and necessary constraint conditions, and obtains the flight control quantities of each UCAV based on this dynamic model; secondly, in order to reach the target point quickly and smoothly, the multi-UCAV collaborative real-time trajectory planning objective function is established, and under this objective function, the current flight position of the UCAV and the estimated flight position of the UCAV at the next moment are determined based on the I-MRFO algorithm. The multi-UCAV collaborative real-time trajectory planning method described in the present invention adopts the improved manta ray foraging optimization (I-MRFO) algorithm, and uses the control quantities determined by the UCAV dynamic model as input quantities, intermediate quantities, and output quantities, so that the collaboration and real-time performance among the UCAVs are very good, the self-adaptability is very strong, the response is fast, and the accuracy is high, meeting the limitations of various battlefield situations and its own characteristics.

[0028] In the present invention, the three-dimensional terrain is established through a digital elevation map, Shuttle Radar Topography Mission (SRTM) data, and two-dimensional cubic convolution interpolation. In practical applications, when a UCAV executes a combat mission, it will encounter complex undulating terrains. Especially for a UCAV that needs to conduct a low-altitude penetration, the terrain environment is particularly important. Therefore, the present invention uses a digital elevation map and SRTM data to simulate the three-dimensional terrain of the battlefield. However, since this three-dimensional terrain only includes the height information at the sampling points and does not contain the height information at any position within the three-dimensional terrain, two-dimensional cubic convolution interpolation is used to establish the height information at various positions within the three-dimensional terrain. Here, the two-dimensional cubic convolution interpolation method is a prior art and will not be elaborated herein.

[0029] In step 1 of the present invention, the determination of the threat range includes the following steps:

[0030] Step 11: Pre-acquire the deployment azimuth of the enemy air defense system and the performance of each weapon.

[0031] Step 12: Determine the position of the threat source and the planar threat radius according to the deployment azimuth and the performance of each weapon.

[0032] Step 13: The threat range is the spatial range enveloped by a cylinder with the position of the threat source as the center, the planar threat radius as the radius, and an infinite height.

[0033] In practical applications, when each UCAV executes a combat mission, in addition to being affected by the natural terrain environment, it will also encounter threats from threat sources such as enemy radar detection, anti-aircraft guns, and surface-to-air missiles. In addition, compared with ballistic missiles, UCAVs mainly rely on low-altitude penetration to carry out strike missions: when encountering a threat source's fire attack, UCAVs usually turn close to the ground for evasion instead of climbing to get rid of it.

[0034] In the present invention, the flight altitude constraint condition is: Con1 = max{h - h max , h min - h} < 0; where h is the real-time flight altitude of the UCAV, h max is the set maximum flight altitude of the UCAV, and h min is the set minimum flight altitude of the UCAV. In practical applications, UCAVs can effectively utilize the shielding effect of mountain terrains by flying at low altitudes during mission execution, which helps to reduce the probability of being detected by enemy radars. In addition, due to the performance limitations of UCAVs themselves, there is a maximum flight altitude h max ; on the other hand, when UCAVs execute missions in complex mountain terrains, too low a flight altitude increases the probability of crashing due to collision with mountains. To reduce the flight safety hazards of UCAVs, a minimum flight altitude h min also needs to be set.

[0035] In the present invention, the flight speed constraint condition is: Con2 = max{v - v max , v min - v} < 0; where v is the real-time flight speed of the UCAV, v max is the set maximum flight speed of the UCAV, and v min is the set minimum flight speed of the UCAV. In practical applications, due to the maneuverability constraints of the UCAV itself, there is a maximum flight speed v max ; at the same time, a fixed-wing UAV cannot hover in the air and needs to keep flying during the mission execution, so there is also a minimum flight speed v min .

[0036] The determination of the flight threat constraint condition includes the following steps:

[0037] Step 21: Determine whether each UCAV encounters an enemy threat during flight: If so, determine the current position O of each UCAV, the threat source position O1, the threat radius R0 of the threat source as the center of the threat range, the real-time speed direction, and the included angle α0 between the line connecting the current position of the UCAV and the threat source position, and the included angle α1 between the line connecting the current position of the UCAV and the threat source position and the tangent on the side close to the current position of the UCAV between the line connecting the current position of the UCAV and the center threat range radiated by the threat source;

[0038] Step 22: Take the included angle α0 as the speed included angle and the included angle α1 as the tangential included angle; according to the size relationship between the speed included angle and the tangential included angle, determine the way for the UCAV to avoid the threat source: when the speed included angle > the tangential included angle, the UCAV flies straight to avoid the threat source; when the speed included angle < the tangential included angle, the UCAV turns to fly to avoid the threat source, and the turning radius satisfies where O2 represents the center of the turning arc when the UCAV turns, |OO2| represents the maximum turning radius, and |OO1| represents the distance between the current position of the UCAV and the threat source position.

[0039] Figure 2 is a schematic diagram of the first risk avoidance method of the UCAV under the flight threat constraint condition of the present invention. Figure 3 is a schematic diagram of the second risk avoidance method of the UCAV under the flight threat constraint condition of the present invention. Figure 4 is a schematic diagram of the third risk avoidance method of the UCAV under the flight threat constraint condition of the present invention. As Figure 2 shown, when α0 > α1, the UCAV can avoid the strike of the threat source by flying along the direction of the real-time flight speed v. As Figure 3As shown in the figure, when α0 < α1, the UCAV needs to turn to avoid being struck by the threat source. At the same time, when the turning radius satisfies , the UCAV must turn along the dotted arc line. At this time, the turning radius |OO2| is the largest, and the turning arc line shown by the dotted line is tangent to the envelope line of the circular threat range shown by the solid line. The turning arc line is the maximum turning for the UCAV to avoid being struck. As Figure 4 shown in the figure, when α0 < α1, the UCAV needs to turn, and when the turning radius satisfies , the turning arc line shown by the dotted line is separated from the envelope line of the circular threat range shown by the solid line.

[0040] In the present invention, the flight anti-collision constraint condition is: Con4 = max{d - d max , d min - d} < 0; where d is the real-time distance between UCAVs, d max is the maximum communication distance between UCAVs, and d min is the minimum safety distance between UCAVs. In practical applications, in order to avoid collisions between UCAVs during flight, it is necessary to set the minimum safety distance d min between UCAVs; on the other hand, when UCAVs fly in coordination, they need to maintain information sharing and communication, and there is a maximum communication distance d max between UCAVs.

[0041] In the present invention, the dynamic models of the UCAVs are based on the inertial coordinate system and the track coordinate system, as follows:

[0042]

[0043] Among them, is the coordinate of the j-th UCAV in the inertial coordinate system, is the coordinate of the j-th UCAV in the track coordinate system. N represents the total number of UCAVs, j = 1, 2,..., N, and both j and N are natural numbers; v j represents the real-time speed of the j-th UCAV, m j represents the mass of the j-th UCAV, g represents the acceleration due to gravity, γ j represents the track inclination angle of the j-th UCAV, represents the yaw angle of the j-th UCAV, α j represents the angle of attack of the j-th UCAV, μ j represents the roll angle of the j-th UCAV, T j represents the engine thrust of the j-th UCAV, D j represents the aerodynamic drag of the j-th UCAV, L jDenote the aerodynamic lift of the j-th UCAV; moreover, the engine thrust T of the j-th UCAV j , the aerodynamic drag D of the j-th UCAV j , and the aerodynamic lift L of the j-th UCAV j respectively satisfy the following relationships:

[0044] T j = ε j T jmax

[0045]

[0046] wherein, T jmax represents the maximum thrust of the j-th UCAV at a certain moment in units of 1000 lb, and ε j represents the throttle position of the j-th UCAV at a certain moment, the air density ρ = 1.225e -h / 9300 , S represents the cross-sectional area of the UCAV, and C jD represents the aerodynamic drag coefficient of the j-th UCAV at a certain moment, and C jL represents the aerodynamic lift coefficient of the j-th UCAV at a certain moment; the values of C jD and C jL are respectively as follows:

[0047] C jD = (-0.043 + 0.1369·α j )sinα j + (0.131 + 3.0825·α j )cosα j

[0048] C jL = (0.043 - 0.1369·α j )cosα j + (0.131 + 3.0825·α j )sinα j

[0049] According to the above dynamic model of the j-th UCAV, the control quantity Control = [ε j , μ j , α j at a certain moment of the j-th UCAV is obtained.

[0050] In the present invention, the multi-UCAV cooperative real-time trajectory planning objective function minJ is as follows:

[0051] minJ = δ1Con g + δ2Con h + δ3Con c+ηgPen

[0052] Among them, the flight path cost The flight altitude cost The flight coordination cost δ1 represents the weight coefficient of the flight path cost, δ2 represents the weight coefficient of the flight altitude cost, and δ3 represents the weight coefficient of the flight coordination cost; the penalty function i = 1, 2, 3, τ1 represents the value of whether the flight path cost condition is satisfied, τ2 represents the value of whether the flight coordination cost condition is satisfied, and τ3 represents the value of whether the flight path cost condition is satisfied; η represents the penalty factor; represents the real-time position of the jth UCAV at time t, B j represents the target position of the jth UCAV, represents the real-time flight altitude of the jth UCAV at time t, represents the distance of the jth UCAV from its own target point at time t, represents the distance of the kth UCAV from its own target point at time t; the values of the weight coefficient δ1 of the flight path cost, the weight coefficient δ2 of the flight altitude cost, and the weight coefficient δ3 of the flight coordination cost are as follows:

[0053]

[0054] Among them, k is a natural number, and 1 ≤ k ≤ N, k ≠ j.

[0055] In practical applications, the flight path cost Con g is composed of the distances from the track points of each UCAV to their respective target points, and is used to guide the UCAV to fly closer to the target point as soon as possible; the flight altitude cost Con h is the height difference between the track points of each UCAV and the terrain where they are located, and is used to guide the UCAV to fly as close to the ground as possible on the premise of ensuring that it does not crash and fall, reducing the probability of being detected by radar; the flight coordination cost Con c is used to ensure that each UCAV can reach the target point to perform tasks within a short time interval.

[0056] In the present invention, step 4 specifically includes the following steps:

[0057] Step 41: Initialize and generate NP individuals for controlling N UCAVs; let p = 1, and set the maximum number of iterations p max ; among them, each individual includes 3N elements, p represents the current number of iterations, P, NP, p max are all natural numbers.

[0058] Step 42: Randomly generate the first random number rand1 and the second random number rand2.

[0059] Step 43: Obtain the adaptive control parameter

[0060] Step 44: Compare the current iteration number p with the maximum iteration number p max as follows: When p ≤ p max execute Step 45.

[0061] Step 45: Determine whether rand1 < 0.5 holds: If it holds, execute Step 46; if not, execute Step 47.

[0062] Step 46: Compare Coef with the first random number rand1: If Coef > rand1, then

[0063]

[0064] wherein, each element in the first random vector r1 and the second random vector r2 is uniformly distributed with values ranging from 0 to 1; the e-th individual among the NP individuals in the p-th generation the e-th individual among the NP individuals in the (p + 1)-th generation the individual with the best performance among the NP individuals in the p-th generation both e and q are natural numbers, and e = 1, 2,..., NP, q ∈ {1, 2,..., NP}; the first weight coefficient β1 takes the value of: Here, the third random number rand3 is a uniformly distributed random number, and rand3 ∈ [0, 1];

[0065] If Coef ≤ rand1, then

[0066]

[0067] wherein, represents the reference point of the NP individuals in the p-th generation, and here, respectively represent the 3 individuals with the best performance among the NP individuals in the p-th generation, and the additional candidate reference points wherein, each element in the third random vector r3, the fourth random vector r4, the fifth random vector r5, the sixth random vector r6, and the seventh random vector r7 is uniformly distributed with values ranging from 0 to 1.

[0068] Step 47: Determine whether the second random number rand2 < 0.5 holds: If it holds, then

[0069]

[0070] wherein, the second weight coefficient The eighth random vector r8, the ninth random vector r9, the tenth random vector r 10 , the eleventh random vector r 11 All elements in are uniformly distributed with values ranging from 0 to 1;

[0071] If it does not hold, then

[0072]

[0073] Among them, mean represents the mean vector, the bias vector w follows a normal distribution, w~N(0, cov); cov represents the weighted covariance matrix of the dominant UCAV population; the values of the mean vector mean and the weighted covariance matrix cov are as follows:

[0074]

[0075] Here, The third weight coefficient

[0076] Step 48: Calculate the multi-UCAV cooperative real-time trajectory planning objective function of the e-th individual among the NP individuals in the (p + 1)-th generation, and update the iteration count of the e-th individual;

[0077]

[0078] Among them, S represents the individual roll range influence coefficient, S = 2; the fourth random number rand4 ∈ [0, 1], and the fifth random number rand5 ∈ [0, 1].

[0079] Step 49: Calculate the multi-UCAV cooperative real-time trajectory planning objective function of the e-th individual in the (p + 2)-th generation:

[0080]

[0081] Step 410: When q s > q max Stop running and stop trajectory planning; otherwise, let p = p + 1, p + 1 = p + 2, and return to Step 42.

[0082] In practical applications, the p-th generation is the P-th iteration of the above algorithm. In the present invention, each individual includes the control vectors of N aircraft, and each aircraft has three control quantities, so each individual contains 3N elements. The NP individuals refer to NP different 3N-dimensional control vectors that regard N aircraft as a whole.

[0083] In the present invention, the above-mentioned I-MRFO algorithm adopted in step 4 enables the UCAV to expand the search range of the target or threat source between the upper and lower boundaries of the search space. The random reference point is selected from the elite pool set composed of the current best three individuals, enabling the I-MRFO algorithm to converge quickly. And creatively, three manta ray foraging optimization strategies, namely chain foraging, spiral foraging, and loop-the-loop foraging, are ingeniously integrated to establish an improved manta ray foraging optimization algorithm based on the distributed estimation algorithm. This enables the estimation of the track points at the next moment in the UCAV cooperative real-time track to have high adaptability to terrain following and target tracking. At the same time, it also has advantages such as fast response, strong real-time performance, and high accuracy, improving the application development performance of the I-MRFO algorithm.

[0084] Embodiment

[0085] In this embodiment, the takeoff points, target positions, and threat sources of the three UCAVs have been obtained, and the specific information is shown in Tables 1 and 2.

[0086] Table 1 Information on the takeoff points and target positions of each UCAV

[0087]

[0088] Table 2 Threat source information

[0089]

[0090] In this embodiment, the time window for the real-time track planning of each UCAV is 1 second, that is, the next position state of each UCAV is planned within 1 second. The tasks of each UCAV are set in a battlefield space of (100×100) km. The maximum flight altitude of each UCAV is 5 km, and the minimum flight altitude is 0.2 km; the maximum flight speed of each UCAV is 0.8 Ma (Mach), and the minimum flight speed is 0.3 Ma; the angle of attack range of each UCAV is [-15°, 15°], and the roll angle range is [-60°, 60°]; the minimum anti-collision distance between each UCAV is 0.2 km, and the maximum communication distance is 30 km; take δ1 = 0.3, δ2 = 0.6, δ3 = 0.1, and the maximum number of iterations q max = 3000, the search population NP = 30, and the problem dimension Dim = 3 UCAVs × 3 control variables = 9. The simulation stops when all UCAVs reach the target point.

[0091] Figure 5 It is a schematic diagram of the relative distances of each UCAV during flight in the embodiment of the present invention. As Figure 5As shown in the figure, the relative distance between each UCAV during the flight is [0.84, 18.82] km, which is much larger than the safe flight distance of UCAVs and meets the maximum communication distance between each UCAV. This ensures that each UCAV can fly safely and exchange and share information.

[0092] Figure 6 Schematic diagram of the change of the height difference between each UCAV and the ground in the embodiment of the present invention. Figure 7 Schematic diagram of the change of altitude of each UCAV during flight in the embodiment of the present invention. Figure 6 , Figure 7 As shown in the figure, in this embodiment, the maximum altitude of each UCAV during flight is 0.31km, and the minimum altitude is 0.2km, which meets the flight altitude constraint. At the same time, during the entire flight, the altitude of each UCAV and the corresponding terrain does not change much, indicating that each UCAV can follow the terrain well and can achieve ground penetration.

[0093] Figure 8 Schematic diagram of the flight speed change of each UCAV in the embodiment of the present invention. Figure 9 Schematic diagram of the time consumption of real-time track planning in an embodiment of the present invention. Figure 8 As shown in the figure, in the embodiment of the present invention, each UCAV quickly increases its flight speed after taking off, which is conducive to reducing the flight time. At the same time, the flight speeds of the three UCAVs are all between [0.35, 0.8] Ma, satisfying the flight speed constraint. The arrival times of the three UCAVs are 435s, 435s and 442s respectively. That is to say, under the guidance of the flight coordination cost, each UCAV adjusts its own speed, so that each UCAV can reach the mission target position within a short time interval. Figure 9 As shown, using the I-MRFO method, the time window of each step meets the time constraint of 1 second, which meets the needs of real-time trajectory planning.

[0094] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-UCAV collaborative real-time trajectory planning method, characterized in that, The real-time trajectory planning method comprises the following steps: Step 1: Construct the battlefield environment, including three-dimensional terrain and threat range; Step 2: Establish the dynamics model of each UCAV under the flight altitude constraint condition, flight speed constraint condition, flight threat constraint condition, and flight collision avoidance constraint condition, and obtain the control amount of each UCAV; Step 3: Establish the objective function of multi-UCAV collaborative real-time trajectory planning; Step 4: Determine the real-time trajectory of multiple UCAVs in collaboration based on the I-MRFO algorithm; The step 4 specifically includes the following steps: Step 41: Initialize and generate NP individuals for controlling N UCAVs; let p = 1 and set the maximum number of iterations p max ; where each individual includes 3N elements, p represents the current iteration number, and p, NP, p max are all natural numbers; Step 42, randomly generate a first random number rand1 and a second random number rand2; Step 43, obtain the adaptive control parameter Step 44. Compare the current iteration number p with the maximum iteration number p max as follows: When p ≤ p max , execute Step 45; Step 45, determine whether rand1<0.5 holds: if so, execute step 46; if not, execute step 47; Step 46: Compare Coef with the first random number rand1: If Coef>rand1, then Among them, each element in the first random vector r1 and the second random vector r2 follows a uniform distribution with values ranging from 0 to 1; the e-th individual among the NP individuals in the p-th generation the e-th individual among the NP individuals in the (p + 1)-th generation the individual with the best performance among the NP individuals in the p-th generation Both e and q are natural numbers, and e = 1, 2, …, NP, q ∈ {1, 2, …, NP}; the value of the first weight coefficient β1 is: Here, the third random number rand3 is a uniformly distributed random number, and rand3 ∈ [0, 1]; If Coef ≤ rand1, then Among them, represents the reference point of the NP individuals in the p-th generation, and Here, respectively represent the 3 individuals with the best performance among the NP individuals in the p-th generation, plus the candidate reference points Among them, each element in the third random vector r3, the fourth random vector r4, the fifth random vector r5, the sixth random vector r6, and the seventh random vector r7 follows a uniform distribution with values ranging from 0 to 1; Step 47: Determine whether the second random number rand2<0.5 holds: If so, then Among them, the second weight coefficient the eighth random vector r8, the ninth random vector r9, the tenth random vector r 10 , the eleventh random vector r 11 each element of which follows a uniform distribution with values ranging from 0 to 1; If not, then where mean represents the mean vector, the bias vector w follows a normal distribution, w ∼ N(0, cov); cov represents the weighted covariance matrix; the values of the mean vector mean and the weighted covariance matrix cov are as follows: Here, The third weight coefficient Step 48, calculating the UCAV collaborative real-time trajectory planning objective function of the e-th individual among the p+1-th generation NP individuals, and updating the number of iterations of the e-th individual; Wherein, S represents the UCAV roll range influence coefficient, S=2; the fourth random number rand4∈[0,1], the fifth random number rand5∈[0,1]; Step 49, calculate the multi-UCAV collaborative real-time trajectory planning objective function of the e-th individual of the p+2th generation: Step 410: When q s >q max , end the operation and stop the trajectory planning; otherwise, set p = p + 1, p + 1 = p + 2, and return to Step 42.

2. The multi-UCAV collaborative real-time trajectory planning method according to claim 1, wherein, The three-dimensional terrain is established by digital elevation maps, space shuttle radar terrain-based mapping mission data, and two-dimensional cubic convolution interpolation; Determination of the threat scope includes the following steps: Step 11: Obtain the deployment position and weapon performance of the enemy air defense system in advance; Step 12: Determine the location of the threat source and the plane threat radius based on the deployment orientation and the performance of each weapon; Step 13: The threat range is the space enclosed by a cylinder with the position of the threat source as the center, the plane threat radius as the radius, and an infinite height.

3. The multi-UCAV collaborative real-time trajectory planning method according to claim 2, wherein The flight altitude constraint condition is: Con1 = max{h - h max , h min - h} < 0; where h is the real-time flight altitude of the UCAV, h max is the set maximum flight altitude of the UCAV, and h min is the set minimum flight altitude of the UCAV; The flight speed constraint condition is: Con2 = max{v - v max , v min - v} < 0; where v is the real-time flight speed of the UCAV, v max is the set maximum flight speed of the UCAV, and v min is the set minimum flight speed of the UCAV; The determination of the flight threat constraint condition comprises the following steps: Step 21, determine whether each UCAV encounters an enemy threat during flight: if so, determine the current position O of each UCAV, the position O1 of the threat source, the threat range of the threat radius R0 radiated by the threat source, the real-time speed direction and the angle α0 between the line connecting the current position of the UCAV and the position of the threat source, and the angle α1 between the line connecting the current position of the UCAV and the position of the threat source and the tangent between the current position of the UCAV and the threat range of the threat source radiated by the threat source and the tangent close to the current position of the UCAV; Step 22: Take the included angle α0 as the velocity included angle and the included angle α1 as the tangential included angle; determine the way for the UCAV to avoid the threat source according to the size relationship between the velocity included angle and the tangential included angle: when the velocity included angle > the tangential included angle, the UCAV flies straight to avoid the threat source; when the velocity included angle < the tangential included angle, the UCAV turns to fly to avoid the threat source, and the turning radius satisfies where O2 represents the center of the turning arc when the UCAV turns, |OO2| represents the maximum turning radius, and |OO1| represents the distance between the current position of the UCAV and the position of the threat source; The flight anti-collision constraint condition is: Con4 = max{d - d max , d min - d} < 0; where d is the real-time distance between UCAVs, d max is the maximum communication distance between UCAVs, d min is the minimum safety distance between UCAVs; The UCAV dynamic models are based on the inertial coordinate system and the track coordinate system, as follows: Among them, is the coordinate of the j-th UCAV in the inertial coordinate system, is the coordinate of the j-th UCAV in the track coordinate system, N represents the total number of UCAVs, j = 1, 2, …, N, and both j and N are natural numbers; v j represents the real-time speed of the j-th UCAV, m j represents the mass of the j-th UCAV, g represents the acceleration due to gravity, γ j represents the track inclination angle of the j-th UCAV, θ j represents the yaw angle of the j-th UCAV, α j represents the angle of attack of the j-th UCAV, μ j represents the roll angle of the j-th UCAV, T j represents the engine thrust of the j-th UCAV, D j represents the aerodynamic drag of the j-th UCAV, L j represents the aerodynamic lift of the j-th UCAV; moreover, the engine thrust T j of the j-th UCAV, the aerodynamic drag D j of the j-th UCAV, and the aerodynamic lift L j of the j-th UCAV respectively satisfy the following relationships: T j = ε j T jmax Among them, T jmax represents the maximum thrust of the j-th UCAV at a certain moment in units of 1000 lb, and ε j represents the throttle position of the j-th UCAV at a certain moment. The air density ρ = 1.225e -h / 9300 , S represents the cross-sectional area of the UCAV, and C jD represents the aerodynamic drag coefficient of the j-th UCAV at a certain moment, and C jL represents the aerodynamic lift coefficient of the j-th UCAV at a certain moment; C jD , C jL take the following values respectively: C jD = (-0.043 + 0.1369·α j )sinα j + (0.131 + 3.0825·α j )cosα j C jL = (0.043 - 0.1369·α j ) cosα j + (0.131 + 3.0825·α j ) sinα j According to the dynamic model of the j-th UCAV above, the control quantity of the j-th UCAV at a certain moment is obtained as Control = [ε j , μ j , α j .

4. The multi-UCAV collaborative real-time trajectory planning method according to claim 3, wherein The objective function minJ of multi-UCAV collaborative real-time trajectory planning is as follows: minJ = δ1Con g + δ2Con h + δ3Con c + ηgPen Among them, the flight path cost flight altitude cost flight cooperation cost δ1 represents the weight coefficient of the flight path cost, δ2 represents the weight coefficient of the flight altitude cost, and δ3 represents the weight coefficient of the flight cooperation cost; the penalty function τ1 represents the value of whether the flight path cost condition is satisfied, τ2 represents the value of whether the flight cooperation cost condition is satisfied, and τ3 represents the value of whether the flight path cost condition is satisfied; η represents the penalty factor; represents the real-time position of the j-th UCAV at time t, B j represents the target position of the j-th UCAV, represents the real-time flight altitude of the j-th UCAV at time t, represents the distance of the j-th UCAV from its own target point at time t, represents the distance of the k-th UCAV from its own target point at time t; the values of the weight coefficient δ1 of the flight path cost, the weight coefficient δ2 of the flight altitude cost, and the weight coefficient δ3 of the flight cooperation cost are as follows: Wherein, k is a natural number, and 1≤k≤N, k≠j.

Images