CN113268087A - Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment - Google Patents
Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment Download PDFInfo
- Publication number
- CN113268087A CN113268087A CN202110635113.9A CN202110635113A CN113268087A CN 113268087 A CN113268087 A CN 113268087A CN 202110635113 A CN202110635113 A CN 202110635113A CN 113268087 A CN113268087 A CN 113268087A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- constraint
- ant colony
- colony algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 103
- 238000000034 method Methods 0.000 title claims abstract description 46
- 239000003016 pheromone Substances 0.000 claims abstract description 61
- 230000007246 mechanism Effects 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims abstract description 16
- 241000257303 Hymenoptera Species 0.000 claims description 33
- 238000004088 simulation Methods 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 20
- 230000008859 change Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 241000282461 Canis lupus Species 0.000 abstract description 11
- 238000005457 optimization Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000011217 control strategy Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/12—Target-seeking control
Abstract
The invention relates to a flight path planning method for multi-unmanned aerial vehicle cooperative work based on an improved ant colony algorithm in a multi-constraint complex environment, which effectively solves the problems of long iteration time, slow convergence speed or early falling into a local optimal solution or even algorithm stagnation of the existing ant colony algorithm in the agricultural process; the technical scheme comprises the following steps: on the basis of the traditional ant colony algorithm, a wolf colony distribution principle is introduced for reference, a roulette mechanism is introduced at the same time, two operation modes of a random pheromone volatilization coefficient, a random pheromone and a heuristic factor importance degree coefficient are added on the basis, the performance of the ant colony algorithm is improved, the operation time is reduced, the situation that the unmanned aerial vehicle is trapped in a local optimal solution is avoided, the unmanned aerial vehicle is re-planned in real time when the unmanned aerial vehicle faces a sudden threat in a dynamic environment, and the intelligent degree of the unmanned aerial vehicle is improved.
Description
Technical Field
The invention belongs to the technical field of multi-unmanned aerial vehicle flight path planning, and particularly relates to a flight path planning method based on the cooperative work of multiple unmanned aerial vehicles based on an improved ant colony algorithm in a multi-constraint complex environment.
Background
In the multi-unmanned aerial vehicle flight path planning design under the multi-constraint complex environment, under the condition of meeting the maneuvering performance of each unmanned aerial vehicle, conflict constraint among the unmanned aerial vehicles and threat constraint of an external complex environment, research in the field is mainly focused on two aspects of a control mode and an optimization algorithm before a route with the minimum total cost from a starting point to a task point is planned for each unmanned aerial vehicle according to task requirements;
in the aspect of control, a centralized mode is a main method for unmanned aerial vehicle cluster collaborative track planning, and hierarchical planning and centralized adjustment are main control strategies of the method, but the centralized control strategy requires that a planning center and all unmanned aerial vehicles keep smooth communication at any time, the difficulty is high in a complex environment, and a distributed decentralized collaborative control scheme cannot solve the problem of time collaborative constraint conditions among tracks;
in the aspect of optimization algorithm, many scholars also provide a plurality of algorithms at present, but most of the algorithms are adaptive expansion of the algorithm of a single unmanned aerial vehicle, and the constraint condition of multi-machine anti-collision and the multi-machine cooperative target are added on the basis of an original model, so that the essence of the algorithm is consistent with the flight path planning of the single unmanned aerial vehicle;
in summary, the current multi-unmanned aerial vehicle flight path planning research under complex environment is not complete, a unified and acknowledged solution and algorithm are not formed, and especially under the cooperation of three-dimensional multiple machines, if a scientific and efficient autonomous cooperative control technology is lacked, the overall efficiency of the multi-unmanned aerial vehicle system is reduced, and the risks of conflict and collision among unmanned aerial vehicles occur;
in view of the above, a flight path planning method based on improved ant colony algorithm for multi-unmanned aerial vehicle cooperative work in a multi-constraint complex environment is provided to solve the above problems.
Disclosure of Invention
Aiming at the situation and overcoming the defects of the prior art, the invention provides a flight path planning method for multi-unmanned aerial vehicle cooperative work based on an improved ant colony algorithm in a multi-constraint complex environment, which introduces a wolf colony distribution principle for reference and introduces a roulette mechanism, adds two operation modes of a random pheromone volatilization coefficient, a random pheromone and a heuristic factor importance coefficient on the basis, respectively carries out simulation calculation on three ant colony algorithms with different improvement degrees, compares and verifies the feasibility of the improved ant colony algorithm in solving the flight path planning problem of the unmanned aerial vehicle and the superiority of the improved ant colony algorithm compared with the traditional ant colony algorithm, finally applies the improved ant colony algorithm to the simulation of the multi-unmanned aerial vehicle cooperative flight path planning, and obtains a proper flight path for the multi-unmanned aerial vehicle flight path planning.
The flight path planning method based on the improved ant colony algorithm and used for the cooperative work of the multiple unmanned aerial vehicles in the multi-constraint complex environment is characterized by comprising the following steps of:
s1: determining the optimal flight path of the unmanned aerial vehicle based on an improved ant colony algorithm, wherein the optimal flight path comprises the properties of updating pheromones, randomness of ant path finding processes, introduction of a random pheromone volatilization coefficient mechanism, and real-time change of pheromone concentration importance coefficient and heuristic factor importance coefficient;
s2: introducing a correlation factor when planning the tracks of the multiple unmanned aerial vehicles and determining a track cost function based on the correlation factor;
s3: setting constraint conditions when planning the tracks of the multiple unmanned aerial vehicles;
s4: determining the flight environment of the unmanned aerial vehicle and establishing an environment model of the flight space of the unmanned aerial vehicle by using a grid method.
Preferably, the step S1 includes the following steps:
s1-1: the ants with shorter found paths are classified into one class, and the concentration of pheromones carried by the ants is increased; meanwhile, ants with poor found paths are classified into another type, the concentration of pheromones carried by the ants is reduced, and the updating method for obtaining the pheromones based on the mechanism is as follows:
in the formulae (1-6) and (1-7), L*And L**Representing the optimal path and the worst path found by the ant colony in the iteration; delta and omega respectively represent the number of ants for finding the optimal path and the worst path;
s1-2: taking each node which the ant may go to as each sector of the wheel disc, selecting the sector area represented by the node with high probability to be larger, and simultaneously using rand (0,1) to generate a random number between 0 and 1 to replace a pointer, rotating the pointer, wherein the node represented by the area where the pointer stays is the node of the ant moving next step;
s1-3: a random pheromone volatilization coefficient mechanism is introduced, and when each iteration starts, a random value based on the range is given to the pheromone volatilization coefficient, as shown in the formula (1-8):
ρ＝rand×0.5+1 (1-8)
s1-4: endowing the property that the pheromone concentration importance degree coefficient and the heuristic factor importance degree coefficient change in real time, randomly producing different pheromone concentration importance degree coefficients and heuristic factor importance degree coefficient values in the path searching process of each ant in each iteration, automatically comparing an optimal result when an ant colony is subjected to path searching, and realizing the algorithm as shown in the formula (1-9):
α＝rand×4+3
β＝rand×4+3 (1-9)
preferably, the S2 includes: let S be { S ═ S1，s2，s3，...，snThe ith unmanned aerial vehicle drives the state variable of each track point s in the three-dimensional space to be Pi(x, y, z) represents the position coordinates of the drone, and the set of tracks from the starting point to the target point is Ei＝{e1，e2，e3，...，emN (N is the number of drones), and m is the number of tracks in a single drone route;
if sa，sbA track e representing the ith unmanned planei，kTwo end points of, Li(sa，sb) Representing the edges of the two nodes, ji，a，bIndicating that the unmanned aerial vehicle has overflowed the edge Li(sa，sb) The track cost of; then
In the formula (2-1), Oi(Sa，Sb) At the cost of oil consumption, Hi(Sa，Sb) At a high cost, Di(Sa，Sb) At the cost of the threat, it is,
for each cost of the ambiguity factor, δiFor the fuzzy factor of the ith unmanned plane, the flight path planning problem can be described as shown in equation (2-2):
preferably, the constraints in S3 include drone performance constraints, external threat constraints, and multi-drone collaborative space-time constraints.
Preferably, the drone performance constraints include:
(1) maximum turning angle: ith drives unmanned aerial vehicle on track ei，kVector a for flight direction ofi，kIt is shown that the maximum turning angle from the kth track to the next track is α (α)<90) If the constraint condition is represented by the formula (2-3);
(2) maximum/minimum track length: set the maximum track length sum to LmaxThe minimum track length is LminIf the constraint condition is represented by the formula (2-4);
Li(sa，sb)≥Lmin (2-4)
(3) maximum/minimum flying height: setting the maximum flying height as HmaxMinimum flying height of HminIf the flight height of the ith unmanned aerial vehicle on the track point s is h, the constraint condition is shown in the formula (2-5);
Hmin≤h≤Hmax (2-5)
preferably, the external threat constraints include:
(1) radar threat: when the unmanned aerial vehicle is set in the detection range of the radar, the distance between the unmanned aerial vehicle and the radar is RlThe maximum detection range of the radar is RmaxThen the probability that the drone is detected by the radar can be roughly expressed as shown in equation (2-6);
(2) and (3) mountain threat: set for the height of unmanned aerial vehicle when the waypoint s is less than mountain height H, this moment apart from the horizontal distance of mountain edge be d, be greater than safe distance d apart from the distance of mountain edge when unmanned aerial vehiclesafeWhen the distance from the edge of the mountain land is less than the safety distance, the probability that the unmanned aerial vehicle hits the mountain is in inverse proportion to the horizontal distance d, and the probability formula (2-7) that the unmanned aerial vehicle hits the mountain is shown;
preferably, the multi-drone collaborative space-time constraint includes:
(1) and (3) collision restraint: the position of the ith unmanned plane at the moment t is set as (x)i，yi，zi) And the position of the jth unmanned plane at the time t is (x)j，yj，zj) If so, the constraint condition is expressed by the formula (2-8);
in the formula DminRepresents a minimum safe distance between drones;
(2) and (3) time constraint: the speed range of the unmanned plane is [ v ]min，vmax]And simultaneously the speed of the single unmanned aerial vehicle in the whole route keeps certain, and the total length of the route for driving the unmanned aerial vehicle is as follows:
the time when the piloting drone arrives at the target location:
the constraint formula is shown as (2-9);
max[T1，min，T2，min，...，Tn，min]≤Ti≤min[T1，max，T2，max，...，Tn，max]，i＝1，2，...，n (2-9)
preferably, the step S4 includes the following steps:
s4-1: establishing a random ground terrain space;
Z1＝500×[rand(60)-0.3] (3-1)
and equating the radar constraint and the mountain region constraint to a mountain region environment, wherein the equivalence process comprises the following steps:
in the formula (3-2), hiIndicates the height of each peak, (x)0i，y0i) Two-dimensional plane projection representing peak-peakCoordinate position, xtAnd ytRepresenting a slope measure for each peak;
s4-2: fusing a mountain environment and a random terrain environment, wherein the fusing mode is shown as a formula (3-3);
Z(x,y)＝max[Zi(x,y),Z2(x,y)] (3-3)
s4-3: and setting each external threat constraint in a three-dimensional terrain space, equivalently converting the external threats into a mountain environment, fusing the mountain environment with a random terrain environment, and constructing a three-dimensional digital map required by the flight path planning simulation.
Preferably, in the step S1-4, the distance between the next node and each threat center needs to be considered when calculating the heuristic factor, and the distance needs to be as far as possible from the threat center, and the calculation method of the heuristic factor threat track cost in the ant colony algorithm is shown in formula (3-4);
(x0i，y0i) Coordinates projected on the XY axis plane as the center of the threat.
The beneficial effects of the technical scheme are as follows:
(1) on the basis of the traditional ant colony algorithm, a wolf colony distribution principle is introduced for reference, a roulette mechanism is introduced, two operation modes of a random pheromone volatilization coefficient, a random pheromone and a heuristic factor importance degree coefficient are added on the basis, the performance of the ant colony algorithm is improved, the operation time is reduced, the phenomenon of falling into a local optimal solution is avoided, the unmanned aerial vehicle is re-planned in real time when the unmanned aerial vehicle faces emergent threats in a dynamic environment, and the intelligent degree of the unmanned aerial vehicle is improved;
(2) through simulation calculation of three ant colony algorithms with different improvement degrees, the feasibility of the improved ant colony algorithm in solving the unmanned aerial vehicle flight path planning problem and the superiority of the improved ant colony algorithm compared with the traditional ant colony algorithm are verified in a contrast mode, and finally the improved ant colony algorithm is applied to the simulation of the multi-unmanned aerial vehicle collaborative flight path planning, so that a proper flight path of the multi-unmanned aerial vehicle flight path planning is obtained.
Drawings
FIG. 1 is a schematic diagram of a flat random ground terrain environment according to the present invention;
FIG. 2 is a three-dimensional digital representation of the track planning of the present invention;
FIG. 3 is a schematic diagram of a three-dimensional terrain track two-dimensional contour terrain track simulation of a single unmanned aerial vehicle with a traditional ant colony algorithm according to the present invention;
FIG. 4 is a schematic diagram of a simulation of a three-dimensional terrain track and a two-dimensional contour terrain track of a single unmanned aerial vehicle according to the paper [17] with an improved ant colony algorithm;
FIG. 5 is a schematic diagram of a simulation of a three-dimensional terrain track and a two-dimensional contour terrain track of a single unmanned aerial vehicle according to the invention with an improved ant colony algorithm;
FIG. 6 is a schematic diagram of an improved ant colony algorithm three-dimensional multi-UAV collaborative planning track simulation according to the present invention;
fig. 7 is a schematic diagram of the improved ant colony algorithm three-dimensional multi-unmanned aerial vehicle collaborative planning two-dimensional equal-altitude terrain track simulation of the invention.
Detailed Description
The foregoing and other aspects, features and advantages of the invention are apparent from the following detailed description of the embodiments, which is to be read in connection with the accompanying drawings of fig. 1-7.
The invention has the following brief steps:
s1: determining the optimal flight path of the unmanned aerial vehicle based on an improved ant colony algorithm, wherein the optimal flight path comprises the properties of updating pheromones, randomness of ant path finding processes, introduction of a random pheromone volatilization coefficient mechanism, and real-time change of pheromone concentration importance coefficient and heuristic factor importance coefficient;
s2: introducing a correlation factor when planning the tracks of the multiple unmanned aerial vehicles and determining a track cost function based on the correlation factor;
s3: setting constraint conditions when planning the tracks of the multiple unmanned aerial vehicles;
s4: determining the flight environment of the unmanned aerial vehicle and establishing an environment model of the flight space of the unmanned aerial vehicle by using a grid method.
Before describing the present solution in detail with the above steps, a conventional ant colony algorithm distancing mechanism is first explained, as follows:
a probability type intelligent cluster algorithm for finding the optimal path in a map is designed, at the initial moment, m ants are randomly scattered on each node in the map, the pheromone concentration on each path is the same, and tauij(0)＝τ0Is the initial value of the pheromone. An ant k (k is 1, 2, 3.. and m) selects a node to go to next according to a random transfer rule, and the selection probability calculation method of the ant is as follows:
in the formula (1-1), τijConcentration of pheromones on edges (i, j) of two nodes, etaij＝1/dijIs a heuristic factor from the i node to the j node; a iskIs the set of nodes that the kth ant is allowed to access next.
In order not to let ants access nodes that have already been experienced, a tabu list is introduced in the ant system. After the time t, all ants wander to each map node, the length of the route traveled by each ant is calculated, the shortest path length is stored, the pheromone concentration [10] on each edge is updated, and the pheromone updating process is as follows.
(1) And (5) volatilizing the pheromone.
τij＝(1-ρ)τij (1-2)
In the formula (1-2), rho is the pheromone volatilization coefficient, and rho is more than 0 and less than or equal to 1.
(2) The ants release pheromone on the passing edge.
In the formula (1-3), Δ τij kThe pheromone released by the kth ant to the edge through which the kth ant passes is calculated by the following method:
as can be seen from the equations (1-4), the shorter the path length traveled by an ant, the more pheromones are added to the edge, and it is more likely that the ant will be selected by other ants in the subsequent iteration, the ant colony algorithm is more suitable for solving the small-scale TSP problem, but if the problem is scaled up, for example, when the multi-drone flight path planning problem in the three-dimensional environment is solved, the performance of the ant system will be seriously degraded, the iteration time is too long, the convergence speed is slow, or the ant system will be trapped in the local optimal solution or even the algorithm will be stagnated earlier.
The detailed steps and embodiments of the invention are as follows:
improvement of ant colony algorithm
Aiming at the problems of the traditional ant colony algorithm, the traditional ant colony algorithm is improved at present in order to improve the performance of the ant colony algorithm, reduce the operation time and avoid trapping in a local optimal solution;
(1) wolf pack allocation principle
The improved method aims at the problem that the convergence speed is low and the iteration time is long when the traditional ant colony algorithm is used for solving a large-scale problem, a wolf colony distribution principle is introduced to improve an pheromone updating mechanism so as to accelerate the convergence speed of the algorithm, namely, after each hunting is completed, a wolf colony can distribute better meat on a hunting object to strong wolfs which are remarkably contributed in the current hunting, and wolfs which are old and weak and do not contribute to a colony can be distributed to the rest parts, so that the whole wolf colony becomes stronger and is easier to succeed in the next hunting;
in the ant system, after each iteration, some ants find a shorter path, the pheromone released by the ants promotes the positive feedback optimization of the whole ant colony, but a part of ants find a poorer path, and the pheromone released by the part of ants slows down the convergence speed of the ant colony optimization. By using a wolf colony distribution mechanism, ants with short searched paths are classified into one class, the concentration of pheromones carried by the ants is increased, and ants with poor searched paths are classified into another class, so that the concentration of the pheromones carried by the ants is reduced;
the pheromone updating method under the mechanism is as follows:
in the formulae (1-6) and (1-7), L*And L**The optimal path and the worst path found by the ant colony in the iteration are shown, and the number of ants for finding the optimal path and the worst path is respectively shown by delta and omega.
(2) Rules of roulette
In the traditional ant colony algorithm, ants select the next node to go to by calculating the selection probability, and usually select the node with the highest probability, but the randomness of the ants in selecting the node cannot be ensured under the condition, so that the ant is easy to fall into local optimum, or the ant cannot find food due to the fact that the algorithm falls into a stagnation state;
by introducing a roulette mechanism, each node which the ant can possibly go to is taken as each sector of the roulette, the sector area represented by the node with high selection probability is larger, a random number between 0 and 1 is generated by rand (0,1) to replace a pointer, the pointer is rotated, the node represented by the area where the pointer stays is the node where the ant moves next step, the roulette mechanism can ensure that each node which the ant can possibly go to has the selected probability, the randomness in the ant routing process is ensured, and the algorithm is prevented from falling into a local optimal or stagnation state;
assuming that the node set which can be moved next by the current node is N, four nodes A, B, C and D are in the set, and the selection probabilities of the nodes are 0.1, 0.3, 0.4 and 0.2 respectively. If 0 is more than or equal to rand and less than 0.1, the ants move to the node A; if 0.1 is more than or equal to rand is less than 0.4, the ant will move to the node B; if rand is more than or equal to 0.4 and less than 0.8, the ants move to the node C; if rand is more than or equal to 0.8 and less than or equal to 1, the ants move to the node D, so that the randomness of the ants in selecting the nodes is ensured, and the phenomenon that local parts are easy to fall into an optimal state or an algorithm falls into a stagnation state is avoided.
(3) Volatile coefficient of random pheromone
In the traditional ant colony algorithm, when pheromone is updated in each iteration, the volatilization coefficient rho of the pheromone is fixed and unchanged, but in a natural environment, because the environment changes at any moment, factors such as weather, illumination and the like can influence the volatilization degree of the pheromone left by ants, for example, in windy days and rainy days, the pheromone on a path through which the ants pass can volatilize more quickly, and the pheromone can remain for a longer time in clear and calm weather, so that the advancing process of the ant colony in the natural environment when food is searched can be simulated more truly, meanwhile, the phenomenon that the pheromone of part of paths of the ant colony algorithm is too high to cause the algorithm to fall into local optimum and premature convergence due to unreasonable arrangement of the pheromone is avoided, and a random pheromone volatilization coefficient mechanism is introduced;
firstly, setting the upper limit and the lower limit of pheromone volatilization coefficients according to application cases of a large number of ant colony algorithms, and selecting the minimum rho of the pheromone volatilization coefficients in the textmin0.1, max ρmax0.6. At the beginning of each iteration, the pheromone volatility index is assigned a random value based on this range, as shown in equations (1-8):
ρ＝rand×0.5+1 (1-8)
(4) random pheromone concentration and factor importance coefficient
In the traditional ant colony algorithm, an pheromone concentration importance coefficient alpha and an elicitation factor beta importance coefficient are preset and are invariable in the whole iterative optimization process, the size of alpha influences the randomness of the algorithm in iteration and the simulation degree of ant colony path-seeking, the size of beta influences the path-seeking speed of the algorithm, the sizes of the alpha and the beta determine the optimization performance of the algorithm and whether the algorithm falls into local optimization or not, and the important influence is exerted on the ant colony algorithm;
in general, when the ant colony algorithm is applied to solve different problems, values of alpha and beta need to be tested for many times before a proper value can be determined, but when the problems are complex and have large scale, a simple test cannot determine the optimal values of alpha and beta, and a large number of tests waste too much time. Therefore, when the ant colony algorithm is improved, the alpha and beta real-time change property is given, different alpha and beta values are randomly produced in the path finding process of each ant in each iteration, the optimal result is automatically compared when the ant colony path is found, wherein the alpha is more than or equal to 3 and less than or equal to 7, the beta is more than or equal to 3 and less than or equal to 7 according to the experience of the prior ant colony algorithm, and the implementation method of the algorithm is shown as the formula (1-9);
α＝rand×4+3
β＝rand×4+3 (1-9)
mathematical model for planning flight paths of two or more unmanned aerial vehicles
(1) Track cost function determination based on correlation factors
The comprehensive cost indexes of the flight path planning mainly comprise oil consumption cost, threat cost, height cost and the like[16]The comprehensive cost target is determined by weighted summation of all the costs, wherein the threat costs include atmospheric threat cost, terrain threat cost, radar threat cost and the like, the goal of the track planning is to minimize the overall cost, in the unmanned aerial vehicle track planning research, the cost weights of the tracks are often determined according to experience, so that the method has greater subjectivity, when a plurality of schemes are subjected to optimal decision, the threats are assumed to be independent from each other, but threat factors are correlated with each other, the overall uniformity is embodied, the factors are correlated with each other and influence the system characteristics together, and the influence degree is difficult to determine, so that a comprehensive cost model is constructed from the perspective of a system, and simultaneously, in system decision, the information given by experts has ambiguity, and the ambiguity often contains the relevance between indexes, so that a novelty introduction relevance factor is created in a comprehensive cost model;
let s be { s ═ s1，s2，s3，...，snThe ith unmanned aerial vehicle drives the state variable of each track point s in the three-dimensional space to be Pi(x, y, z) represents the position coordinates of the drone. The set of tracks from the starting point to the target point is Ei＝{e1，e2，e3，...，emN (N is the number of drones), and m is the number of tracks in a single drone route; if sa，sbA track e representing the ith unmanned planei，kTwo end points of, Li(sa，sb) Representing the edges of the two nodes, ji，a，bIndicating that the unmanned aerial vehicle has overflowed the edge Li(sa，sb) The track cost of; then
In the formula (2-1), Oi(Sa，Sb) At the cost of oil consumption, Hi(Sa，Sb) At a high cost, Di(Sa，Sb) At the cost of the threat, it is,
for each cost of the ambiguity factor, δiFor the fuzzy factor of the ith unmanned plane, the flight path planning problem can be described as shown in equation (2-2):
factor deltaiAnd
the determination method, as shown in tables 1 and 2, is processed using the following rules:
TABLE 1
Computing rules
TABLE 2. deltaiComputing rules
(2) Setting of constraints
When the flight path planning of the multiple unmanned aerial vehicles is carried out, the constraint conditions are divided into three types, namely unmanned aerial vehicle performance constraint, external threat constraint and multi-unmanned aerial vehicle collaborative space-time constraint;
unmanned aerial vehicle performance constraint: the self performance of the unmanned aerial vehicle determines the conditions which should be mainly considered during the flight path planning, and the limit of the self performance determines that the unmanned aerial vehicle has a limited feasible solution when the unmanned aerial vehicle carries out route selection in the face of external threats.
The maximum turning angle: when the unmanned aerial vehicle turns, according to the physical property constraint of the unmanned aerial vehicle, a maximum turning angle exists, and the angle limits that the angle of the turning position in the planned flight path is not larger than the maximum turning angle. Ith drives unmanned aerial vehicle on track ei，kVector a for flight direction ofi，kIt is shown that the maximum turning angle from the kth track to the next track is α (α)<90) If the constraint condition is represented by the formula (2-3);
maximum/minimum track length: because of the fuel limitation, the maximum flight path length of the drone has an upper limit, and the drone must fly a minimum distance on a flight path to change the flight direction again. Set the maximum track length sum to LmaxThe minimum track length is LminIf the constraint condition is represented by the formula (2-4);
Li(sa，sb)≥Lmin (2-4)
third, highest/lowest flying height: the drone has a maximum flying height due to its own performance, while a minimum flying height limit is required to avoid collisions with terrain. Setting the maximum flying height as HmaxMinimum flying height of HminIf the flight height of the ith unmanned aerial vehicle on the track point s is h, the constraint condition is shown in the formula (2-5);
Hmin≤h≤Hmax (2-5)
external threat constraints: during actual navigation, the unmanned aerial vehicle can also be threatened by thunderstorm weather, air-to-air enemy places, air-to-air anti-air weapons and the like;
radar threat: assuming that the distance between the unmanned aerial vehicle and the radar is R when the unmanned aerial vehicle is in the detection range of the radarlThe maximum detection range of the radar is RmaxThe probability that the drone is detected by the radar can be roughly expressed as shown in equation (2-6).
Secondly, mountain region threat: assuming that the height of the unmanned aerial vehicle at the track point s is smaller than the height H of the mountain, the horizontal distance from the edge of the mountain is d. When the distance between the unmanned aerial vehicle and the edge of the mountain land is greater than the safety distance dsafeWhen the distance from the edge of the mountain land is smaller than the safety distance, the probability that the unmanned aerial vehicle hits the mountain is in inverse proportion to the horizontal distance d, and the probability formula (2-7) that the unmanned aerial vehicle hits the mountain is shown.
And (3) multi-unmanned aerial vehicle collaborative space-time constraint: when multiple unmanned aerial vehicles simultaneously execute the same task, each unmanned aerial vehicle needs to be ensuredThe unmanned aerial vehicles do not collide and can reach the target location within the same time node to execute the task, so that the success probability of the task is improved. In order to ensure that a plurality of unmanned aerial vehicles can fly safely without collision, a minimum safe distance D is required between the unmanned aerial vehiclesmin. Suppose that the position of the ith unmanned aerial vehicle at time t is (x)i，yi，zi) And the position of the jth unmanned plane at the time t is (x)j，yj，zj) If so, the constraint condition is expressed by the formula (2-8);
when many unmanned aerial vehicles carry out the same task in a coordinated manner, in order to improve the task completion rate, many unmanned aerial vehicles need to fly to the target location simultaneously. Suppose the speed range of the drone is vmin，vmax]Meanwhile, the speed of the single unmanned aerial vehicle in the whole route is kept constant, and the total length of the route for driving the ith unmanned aerial vehicle is
The time for the unmanned aerial vehicle to reach the target location is as follows:
the constraint is expressed by the formula (2-9);
max[T1，min，T2，min，...，Tn，min]≤Ti≤min[T1，max，T2，max，...，Tn，max]，i＝1，2，...，n (2-9)
three, establishing a space environment model
(1) Unmanned aerial vehicle flight environment model establishment
A three-dimensional digital map of a complex mountain environment is constructed by using a grid map method to serve as a three-dimensional planning space for flight path planning, so that the three-dimensional environment is simplified for the convenience of simulation calculation, and the radar and mountain constraint conditions are simultaneously equivalent to the complex mountain environment. The three-dimensional planning space is set to be a mountain environment with the length and width of 60km and the height of 5000 m. A random ground terrain space is first established. Random terrain height generation in MATLAB, as shown in fig. 1;
Z1＝500×[rand(60)-0.3] (3-1)
and equating the radar constraint and the mountain region constraint to a mountain region environment, wherein the equivalence process comprises the following steps:
in the formula (3-2), hiIndicates the height of each peak, (x)0i，y0i) Two-dimensional plane projection coordinate position, x, representing peak vertextAnd ytRepresenting the slope measure of each peak.
Finally, fusing the mountain environment and the random terrain environment, namely selecting the maximum height in the same coordinate point of the random terrain and the mountain environment as the height data of the fused three-dimensional map environment, wherein the fusion mode is shown as a formula (3-3);
Z(x，y)＝max[Z1(x，y)，Z2(x，y)] (3-3)
in the three-dimensional planning space, setting each external threat constraint as shown in table 3;
the external threat is equivalently converted into a mountain environment, and then the mountain environment is fused with a random terrain environment to construct a three-dimensional digital map required by the flight path planning simulation, as shown in fig. 2.
(2) Ant path-finding mode setting based on three-dimensional planning space
When the ant colony algorithm is used for carrying out the unmanned aerial vehicle track planning simulation calculation, the traditional ant routing mode is not suitable any more due to the particularity and complexity of the unmanned aerial vehicle carrying out the track planning in the three-dimensional planning space. Therefore, when the unmanned aerial vehicle flight path planning simulation calculation is carried out, a special ant path searching mode is firstly set.
For a three-dimensional topography of 60X 50, the space is cut in planes, each plane being a two-dimensional planar environment of 60X 50, by 60 equal divisions along the X axis. The ants start from the starting point, move from one point on a plane in the X-axis direction to the next plane, and finally move to the last plane where the end point is located, so that the whole path searching process is completed.
Meanwhile, due to the constraint of the performance of the unmanned aerial vehicle, when the node of the current plane is fixed, the reachable point of the next adjacent plane can only be selected within the constraint range. Thus setting the maximum moving distance Y of the ant in the Y-axis directionmaxMaximum distance Z of movement in the Z-axis direction of 2max＝3。
In addition, the heuristic factor calculation method based on which the selection probability is selected when the next mobile node is selected in the ant routing process is improved. In the conventional ant colony algorithm, the heuristic factor generally depends on the distance from the next node to the current node, however, the flight path planning focuses on the overall path length of the ant routing, and therefore, the calculation of the heuristic factor needs to comprehensively consider the distance from the next node to the current node and the distance from the next node to the end point. Meanwhile, the threat cost of the next node needs to be considered in the flight path planning, so that the distance between the next node and each threat center needs to be considered in the heuristic factor calculation, and the distance needs to be as far as possible from the threat center. The calculation method of the flight path threat cost of the heuristic factor in the ant colony algorithm is shown as a formula (3-4).
Wherein (x)0i，y0i) Coordinates projected on the XY axis plane as the center of the threat.
Simulation of multi-unmanned aerial vehicle collaborative flight path planning based on improved ant colony algorithm
(1) And in the planning space, selecting an unmanned aerial vehicle with the maximum flying height of 3000m, the farthest flying distance of 100km and the flying speed of 30-90 km/h for simulation calculation.
Firstly, the flight path planning simulation calculation of the single unmanned aerial vehicle is carried out, so that the feasibility of the ant colony algorithm is verified, and the advantages and disadvantages of the ant colony algorithm and the traditional ant colony algorithm in solving the flight path planning problem are compared and improved. And setting various initial parameters of the ant colony algorithm as shown in a table.
TABLE 4 Ant colony Algorithm initial parameter Table
Setting a starting point (1,10,30) and an end point (60,10,35) in a three-dimensional space, and respectively applying an improved ant colony algorithm and a route planning research [ J ] electronic measurement technology based on the improved ant colony algorithm in a traditional ant colony algorithm and a reference 17 (Liangkai, MaoJianlin. dynamic environment, 2020,43(01):57-62.) to carry out flight path planning simulation calculation, as shown in FIGS. 3-5, and simultaneously when a single unmanned aerial vehicle flight path is planned by simulation calculation, the comprehensive cost and distance cost of an optimal flight path are shown in Table 3;
TABLE 3 three ant colony algorithm simulation calculation track cost table
Compared with the traditional ant colony algorithm, the improved ant colony algorithm has the advantages that the cost of the flight path distance is reduced by 14.8% and the comprehensive cost of the flight path is reduced by 10.19% when the flight path planning of the unmanned aerial vehicle is solved.
(2) Multi-unmanned aerial vehicle collaborative track planning simulation
The simulation calculation comparison can verify that the improved ant colony algorithm has feasibility in solving the problem of single unmanned aerial vehicle flight path planning, and has better optimization capability than the traditional ant colony algorithm. When the multi-unmanned aerial vehicle collaborative flight path planning is carried out, a plurality of ant populations are added to respectively simulate a plurality of unmanned aerial vehicles to carry out tasks in a collaborative mode. The simulation results are shown in fig. 6 and 7;
in fig. 6 and 7, the cost of the track distance of the drone 1 is 78.62, the cost of the track is 72.59, the cost of the track distance of the drone 2 is 88.27, and the cost of the track is 75.06. Through simulation calculation, the improved ant colony algorithm has strong feasibility in solving the multi-unmanned aerial vehicle collaborative planning track, the simulation result simultaneously meets the external threat constraint condition and the multi-unmanned aerial vehicle collaborative space-time constraint condition, the safety of the multi-unmanned aerial vehicle collaborative task execution is guaranteed, and the success probability of the task execution is improved.
According to the scheme, on the basis of the traditional ant colony algorithm, a wolf colony distribution principle is introduced for reference, a roulette mechanism is introduced, and two operation modes of a random pheromone volatilization coefficient, a random pheromone and a heuristic factor importance degree coefficient are added on the basis;
through simulation calculation of three ant colony algorithms with different improvement degrees, the feasibility of the improved ant colony algorithm in solving the unmanned aerial vehicle flight path planning problem and the superiority of the improved ant colony algorithm compared with the traditional ant colony algorithm are verified in a contrast mode, and finally the improved ant colony algorithm is applied to the simulation of the multi-unmanned aerial vehicle collaborative flight path planning, so that a proper flight path of the multi-unmanned aerial vehicle flight path planning is obtained.
Claims (9)
1. The flight path planning method based on the improved ant colony algorithm and used for the cooperative work of the multiple unmanned aerial vehicles in the multi-constraint complex environment is characterized by comprising the following steps of: s1: determining the optimal flight path of the unmanned aerial vehicle based on an improved ant colony algorithm, wherein the optimal flight path comprises the properties of updating pheromones, randomness of ant path finding processes, introduction of a random pheromone volatilization coefficient mechanism, and real-time change of pheromone concentration importance coefficient and heuristic factor importance coefficient; s2: introducing a correlation factor when planning the tracks of the multiple unmanned aerial vehicles and determining a track cost function based on the correlation factor; s3: setting constraint conditions when planning the tracks of the multiple unmanned aerial vehicles; s4: determining the flight environment of the unmanned aerial vehicle and establishing an environment model of the flight space of the unmanned aerial vehicle by using a grid method.
2. The flight path planning method for cooperative work of multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 1, wherein the step of S1 is as follows:
s1-1: the ants with shorter found paths are classified into one class, and the concentration of pheromones carried by the ants is increased; meanwhile, ants with poor found paths are classified into another type, the concentration of pheromones carried by the ants is reduced, and the updating method for obtaining the pheromones based on the mechanism is as follows:
in the formulae (1-6) and (1-7), L*And L**Representing the optimal path and the worst path found by the ant colony in the iteration; delta and omega respectively represent the number of ants for finding the optimal path and the worst path;
s1-2: taking each node which the ant may go to as each sector of the wheel disc, selecting the sector area represented by the node with high probability to be larger, and simultaneously using rand (0,1) to generate a random number between 0 and 1 to replace a pointer, rotating the pointer, wherein the node represented by the area where the pointer stays is the node of the ant moving next step;
s1-3: a random pheromone volatilization coefficient mechanism is introduced, and when each iteration starts, a random value based on the range is given to the pheromone volatilization coefficient, as shown in the formula (1-8):
ρ＝rand×0.5+1 (1-8)
s1-4: endowing the property that the pheromone concentration importance degree coefficient and the heuristic factor importance degree coefficient change in real time, randomly producing different pheromone concentration importance degree coefficients and heuristic factor importance degree coefficient values in the path searching process of each ant in each iteration, automatically comparing an optimal result when an ant colony is subjected to path searching, and realizing the algorithm as shown in the formula (1-9):
α＝rand×4+3
β＝rand×4+3 (1-9)。
3. the flight path planning method for cooperative work of multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 1, wherein the S2 includes: let S be { S ═ S1，s2，s3，…，snThe ith unmanned aerial vehicle drives the state variable of each track point s in the three-dimensional space to be Pi(x, y, z) represents the position coordinates of the drone, and the set of tracks from the starting point to the target point is Ei＝{e1，e2，e3，...，emN (N is the number of drones), and m is the number of tracks in a single drone route;
if sa，sbA track e representing the ith unmanned planei，kTwo end points of, Li(sa，sb) Representing the edges of the two nodes, ji，a，bIndicating that the unmanned aerial vehicle has overflowed the edge Li(sa，sb) The track cost of; then
In the formula (2-1), Oi(Sa，Sb) At the cost of oil consumption, Hi(Sa，Sb) At a high cost, Di(Sa，Sb) At the cost of the threat, it is,
for each cost of the ambiguity factor, δiIs the fuzzy factor of the ith unmanned aerial vehicle, then trackThe planning problem can be described as shown in equation (2-2):
4. the flight path planning method for cooperative work of multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment as claimed in claim 1, wherein the constraint conditions in S3 include unmanned aerial vehicle performance constraint, external threat constraint, and multi-unmanned aerial vehicle cooperative space-time constraint.
5. The flight path planning method for cooperative work of multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 4, wherein the unmanned aerial vehicle performance constraint comprises:
(1) maximum turning angle: ith drives unmanned aerial vehicle on track ei，kVector a for flight direction ofi，kThe maximum turning angle from the kth track to the next track is alpha (alpha is less than 90), and the constraint condition is not represented by the formula (2-3);
(2) maximum/minimum track length: set the maximum track length sum to LmaxThe minimum track length is LminIf the constraint condition is represented by the formula (2-4);
Li(sa，sb)≥Lmin (2-4)
(3) maximum/minimum flying height: setting the maximum flying height as HmaxMinimum flying height of HminIf the flight height of the ith unmanned aerial vehicle on the track point s is h, the constraint condition is shown in the formula (2-5);
Hmin≤h≤Hmax (2-5)。
6. the method for planning the flight path of the cooperative work of the multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 4, wherein the external threat constraint comprises:
(1) radar threat: when the unmanned aerial vehicle is set in the detection range of the radar, the distance between the unmanned aerial vehicle and the radar is RlThe maximum detection range of the radar is RmaxThen the probability that the drone is detected by the radar can be roughly expressed as shown in equation (2-6);
(2) and (3) mountain threat: set for the height of unmanned aerial vehicle when the waypoint s is less than mountain height H, this moment apart from the horizontal distance of mountain edge be d, be greater than safe distance d apart from the distance of mountain edge when unmanned aerial vehiclesafeWhen the distance from the edge of the mountain land is less than the safety distance, the probability that the unmanned aerial vehicle hits the mountain is in inverse proportion to the horizontal distance d, and the probability formula (2-7) that the unmanned aerial vehicle hits the mountain is shown;
7. the flight path planning method for cooperative work of multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 4, wherein the cooperative space-time constraint of multiple unmanned aerial vehicles comprises:
(1) and (3) collision restraint: the position of the ith unmanned plane at the moment t is set as (x)i，yi，zi) And the position of the jth unmanned plane at the time t is (x)j，yj，zj) If so, the constraint condition is expressed by the formula (2-8);
in the formula DminRepresents a minimum safe distance between drones;
(2) and (3) time constraint: the speed range of the unmanned plane is [ v ]min，vmax]And simultaneously the speed of the single unmanned aerial vehicle in the whole route keeps certain, and the total length of the route for driving the unmanned aerial vehicle is as follows:
the time when the piloting drone arrives at the target location:
the constraint formula is shown as (2-9);
max[T1，min，T2，min，...，Tn，min]≤Ti≤min[T1，max，T2，max，...，Tn，max]，i＝1，2，...，n (2-9)。
8. the flight path planning method for cooperative work of multiple unmanned aerial vehicles based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 1, wherein the step of S4 is as follows:
s4-1: establishing a random ground terrain space;
Z1＝500×[rand(60)-0.3] (3-1)
and equating the radar constraint and the mountain region constraint to a mountain region environment, wherein the equivalence process comprises the following steps:
in the formula (3-2)，hiIndicates the height of each peak, (x)0i，y0i) Two-dimensional plane projection coordinate position, x, representing peak vertextAnd ytRepresenting a slope measure for each peak;
s4-2: fusing a mountain environment and a random terrain environment, wherein the fusing mode is shown as a formula (3-3);
Z(x，y)＝max[Z1(x，y)，Z2(x，y)] (3-3)
s4-3: and setting each external threat constraint in a three-dimensional terrain space, equivalently converting the external threats into a mountain environment, fusing the mountain environment with a random terrain environment, and constructing a three-dimensional digital map required by the flight path planning simulation.
9. The flight path planning method for multi-unmanned aerial vehicle collaborative work based on the improved ant colony algorithm in the multi-constraint complex environment according to claim 2, wherein in the step S1-4, the distance between the next node and each threat center needs to be considered when calculating the heuristic factor, and the distance needs to be as far as possible from the threat center, and the calculation method of the threat path cost of the heuristic factor in the ant colony algorithm is shown in the formula (3-4);
(x0i，y0i) Coordinates projected on the XY axis plane as the center of the threat.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110635113.9A CN113268087A (en) | 2021-06-04 | 2021-06-04 | Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110635113.9A CN113268087A (en) | 2021-06-04 | 2021-06-04 | Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113268087A true CN113268087A (en) | 2021-08-17 |
Family
ID=77234677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110635113.9A Withdrawn CN113268087A (en) | 2021-06-04 | 2021-06-04 | Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113268087A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113625767A (en) * | 2021-09-02 | 2021-11-09 | 大连海事大学 | Fixed-wing unmanned aerial vehicle cluster collaborative path planning method based on preferred pheromone gray wolf algorithm |
| CN113703483A (en) * | 2021-08-31 | 2021-11-26 | 湖南苍树航天科技有限公司 | Multi-UAV collaborative trajectory planning method and system, equipment and storage medium |
| CN114397911A (en) * | 2022-01-18 | 2022-04-26 | 中科视拓（北京）科技有限公司 | Unmanned aerial vehicle cluster confrontation decision-making method based on multiple intelligent agents |
| CN114863075A (en) * | 2022-07-05 | 2022-08-05 | 深圳市新天泽消防工程有限公司 | Fire-fighting evacuation path planning method, device and equipment based on multiple sensors |
-
2021
- 2021-06-04 CN CN202110635113.9A patent/CN113268087A/en not_active Withdrawn
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113703483A (en) * | 2021-08-31 | 2021-11-26 | 湖南苍树航天科技有限公司 | Multi-UAV collaborative trajectory planning method and system, equipment and storage medium |
| CN113703483B (en) * | 2021-08-31 | 2024-04-09 | 湖南苍树航天科技有限公司 | Multi-UAV collaborative trajectory planning method, system, equipment and storage medium |
| CN113625767A (en) * | 2021-09-02 | 2021-11-09 | 大连海事大学 | Fixed-wing unmanned aerial vehicle cluster collaborative path planning method based on preferred pheromone gray wolf algorithm |
| CN114397911A (en) * | 2022-01-18 | 2022-04-26 | 中科视拓（北京）科技有限公司 | Unmanned aerial vehicle cluster confrontation decision-making method based on multiple intelligent agents |
| CN114397911B (en) * | 2022-01-18 | 2024-04-09 | 中科视拓（北京）科技有限公司 | Unmanned aerial vehicle cluster countermeasure decision-making method based on multiple intelligent agents |
| CN114863075A (en) * | 2022-07-05 | 2022-08-05 | 深圳市新天泽消防工程有限公司 | Fire-fighting evacuation path planning method, device and equipment based on multiple sensors |
| CN114863075B (en) * | 2022-07-05 | 2022-10-14 | 深圳市新天泽消防工程有限公司 | Fire-fighting evacuation path planning method, device and equipment based on multiple sensors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113268087A (en) | Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment | |
| CN109631900B (en) | Unmanned aerial vehicle three-dimensional flight path multi-target particle swarm global planning method | |
| CN110488859B (en) | Unmanned aerial vehicle route planning method based on improved Q-learning algorithm | |
| CN110926477B (en) | Unmanned aerial vehicle route planning and obstacle avoidance method | |
| Ali et al. | Cooperative path planning of multiple UAVs by using max–min ant colony optimization along with cauchy mutant operator | |
| CN112082552A (en) | Unmanned aerial vehicle flight path planning method based on improved hybrid particle swarm optimization algorithm | |
| CN112000131A (en) | Unmanned aerial vehicle cluster path planning method and system based on artificial potential field method | |
| CN103267528A (en) | Multi-unmanned aerial vehicle cooperative area search method under non-flight zone limitation | |
| CN112230678A (en) | Three-dimensional unmanned aerial vehicle path planning method and planning system based on particle swarm optimization | |
| CN111813144B (en) | Multi-unmanned aerial vehicle collaborative route planning method based on improved flocks of sheep algorithm | |
| CN109885082B (en) | Unmanned aerial vehicle track planning method based on task driving | |
| CN112666981B (en) | Unmanned aerial vehicle cluster dynamic route planning method based on dynamic group learning of original pigeon group | |
| Wang | Path planning of mobile robot based on a* algorithm | |
| CN115657721A (en) | Space environment unmanned aerial vehicle trajectory planning method based on improved ant colony algorithm | |
| CN110530373A (en) | A kind of robot path planning method, controller and system | |
| CN110471419A (en) | A kind of multirobot collaboration paths planning method based on differential evolution fireworks algorithm | |
| Ding et al. | Improved GWO algorithm for UAV path planning on crop pest monitoring | |
| Wang et al. | UAV online path planning based on improved genetic algorithm | |
| Wu et al. | Ucav path planning based on improved chaotic particle swarm optimization | |
| CN116954239B (en) | Unmanned aerial vehicle track planning method and system based on improved particle swarm optimization | |
| Gu et al. | Robot path planning of improved adaptive Ant Colony System Algorithm based on Dijkstra | |
| CN116400737B (en) | Safety path planning system based on ant colony algorithm | |
| CN116518982B (en) | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method | |
| CN116880561A (en) | Optimization method and system based on improved particle swarm unmanned aerial vehicle path planning safety enhancement | |
| Lei et al. | A new machine learning framework for air combat intelligent virtual opponent |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210817 |
|
| WW01 | Invention patent application withdrawn after publication |