CN106197426A - A kind of unmanned plane emergency communication paths planning method and system - Google Patents

Abstract

The present invention relates to a UAV emergency communication path planning method and system. The method includes the following steps: Step 1. Carry out three-dimensional grid division for the flying space of the UAV, obtain grid nodes, construct a grid diagram, and mark no The starting point and target point of the man-machine, as well as marking the radar distribution and terrain information; step 2. Construct a flight threat model according to the radar distribution and terrain information in the UAV flight space, and build a flight limit model at the same time; step 3. According to the flight threat The flight path is constructed by the model and the flight-limited model; Step 4. Optimizing the flight path through a chaotic genetic algorithm to determine the final flight path. Compared with the prior art, the present invention can safely avoid low-altitude obstacles and radar interference, has the shortest flight time and optimal path to reach the designated destination, and has better real-time and rapidity.

Description

A UAV emergency communication path planning method and system

technical field

The invention relates to the technical field of unmanned aerial vehicles, in particular to a method and system for planning emergency communication paths of unmanned aerial vehicles.

Background technique

UAVs are more and more widely used in emergency communications, homeland security, air police, emergency rescue and agricultural planting. UAVs are more popular because of their unique vertical take-off and landing and air hovering performance. In a complex and restricted space, UAVs can use its hovering, horizontal flying, inverted flying and other flying methods to shuttle between obstacles and complete tasks such as communication base station construction, terrain survey, scene shooting, emergency rescue and disaster relief. Different from vehicles and mobile robots on a two-dimensional plane, any collision is fatal to an unmanned helicopter, which directly leads to the failure of the scheduled mission. The ability to autonomously and quickly avoid danger and find the optimal trajectory is the key to its mission completion. Therefore, fast and real-time planning of the optimal path under the premise of ensuring safe flight has become an urgent problem to be solved in the UAV industry.

At present, the commonly used UAV route planning algorithms include linear and nonlinear planning methods, A* algorithm, ant colony algorithm, etc.

Linear and non-linear UAV path planning methods can comprehensively consider the safety and executability related to the path during UAV flight. The disadvantage is that these methods need to solve a series of constrained optimization problems, and the amount of calculation is large. , long calculation time, slow convergence speed and easy to be affected by local minimum and fall into local optimal solution, which is suitable for local track planning.

The A* algorithm has been widely used due to its fast speed. Its performance depends on the selection of the heuristic function, but it is difficult to obtain the optimal heuristic function in practice. Therefore, the A* algorithm generally can only search for one For a relatively optimal path, when the flight path of the UAV changes, the information obtained in the first search cannot be used, and the search can only be performed again, which requires a relatively long time to re-plan the path.

The ant colony algorithm is that ants randomly choose a direction to search when they are looking for food. When an ant finds a food source, it tells its companions to carry the food together in a certain way. When multiple ants cooperate to carry the food, the path they take becomes the shortest. path of. Ant colony algorithm is a global planning method that can consider multiple goals at the same time, but when the environment changes suddenly or real-time planning is required, the ant colony algorithm cannot guarantee that the path can be searched quickly within a limited time.

Contents of the invention

The technical problem to be solved by the present invention is to provide a UAV emergency communication path planning method and system, which can safely avoid the interference of low-altitude obstacles and radar, and reach the designated destination with the shortest flight time and optimal path. On the basis of this, the use of chaotic sequences to control the crossover and mutation in genetic operations avoids the blindness of completely random operations, and has better real-time and rapidity, and the obtained flight path is closer to the actual UAV optimal trajectory .

The technical solution of the present invention to solve the above-mentioned technical problems is as follows: a UAV emergency communication path planning method, comprising the following steps:

Step 1. Carry out three-dimensional grid division on the flying space of the UAV, obtain grid nodes, construct a grid diagram, mark the starting point and target point of the UAV on the grid diagram, and mark the flight of the UAV Radar distribution and terrain information in space;

Step 2. Construct a flight threat model according to the radar distribution and terrain information in the flight space of the UAV, and construct a flight limitation model according to the flight technical parameters of the UAV;

Step 3. Construct a flight route between the starting point and the target point of the drone according to the flight threat model and the flight limitation model;

Step 4. Optimizing the flight route through the chaotic genetic algorithm to determine the final flight route.

The beneficial effect of the present invention is: it can safely avoid the interference of low-altitude obstacles and radar, the flight time to reach the designated destination is the shortest and the path is optimal, and at the same time, on the basis of the original genetic algorithm, the chaotic sequence is used to control the crossover and variation in the genetic operation , to replace the original completely random crossover and mutation operations, to accurately determine whether to perform crossover or mutation operations and to determine the specific position of the crossover or mutation operations, avoiding the blindness of random operations, and making the obtained solutions highly accurate. The convergence speed is fast, and at the same time, the technical solution has better real-time and rapidity, so that when the UAV performs tasks such as emergency communication and air rescue, the searched track is closer to the actual optimal track of the UAV.

On the basis of the above technical solutions, the present invention can also be improved as follows.

Further, the flight threat model is specifically:

min W＝k ₁ J _threat +k ₂ J _fuel ;

Where J _threat is radar threat cost, k ₁ (k ₁ ∈ (0, 1)) is the weight of radar threat cost, J _fuel is fuel cost, k ₂ (k ₂ ∈ (0, 1)) is the weight of fuel cost .

The beneficial effect of adopting the above-mentioned further solution is: through the limited modeling of radar threat and fuel threat, it is ensured that the drone flies without threat, and the flight path of the drone is selected more accurately.

Further, the flight-limited model includes:

When the UAV encounters an obstacle, the distance r from which the UAV changes its attitude and flies straight is less than the distance S _min between the starting point of the UAV and the obstacle point, and r<S _min .

The beneficial effect of adopting the above-mentioned further solution is that by limiting the direct flight distance of the UAV when changing its attitude, it can ensure that the UAV can fly safely according to the set route.

Further, the flight-limited model includes:

The hovering time T _min of the UAV when an obstacle is detected is less than the total flight time t of the UAV, and T _min < t.

The beneficial effect of adopting the above further solution is that the drone can be guaranteed to fly smoothly according to the set route by limiting the hovering time when the drone detects an obstacle.

Further, the flight-limited model includes:

The altitude H of the flight route is greater than the minimum flight altitude H _min set for the drone flight, and smaller than the maximum flight altitude H _max set for the drone flight, and H _min <H<H _max .

The beneficial effect of adopting the above-mentioned further solution is that the unmanned aerial vehicle can be guaranteed to fly smoothly according to the flight route through the limitation of the height H of the flight route.

Further, the flight-limited model includes:

The distance L of the flight route is less than half of the flight limit distance of the UAV, and L<L _max /2.

Preferably, the flight limitation model includes: the UAV flies along the positive direction of the x coordinate axis, the coordinates of the current node are (x1, y1, z1), and the coordinates of the next node are (x2, y2, z2) , the maximum deflection angle Ф of the UAV in the horizontal direction and vertical direction satisfies:

$\left\{\begin{array}{c}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; arctan</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ <span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; arctan</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\end{array}\right.<span\; class="notranslate">\; ;</span>$

Among them, Ф _max is the limit angle of UAV horizontal deflection; у _max is the limit angle of UAV vertical deflection.

The beneficial effect of adopting the above-mentioned further solution is that the unmanned aerial vehicle can be guaranteed to fly smoothly according to the flight route through the limitation of the height H of the flight route.

Further, optimizing the flight route through the chaotic genetic algorithm specifically includes the following steps:

Step 4.1. Use the chaotic motion Logistic mapping equation x _k+1 =ux _k (1-x _k )(1) to randomly generate the initial chaotic sequence, where u is the control parameter. When u=4, the above formula is completely in a chaotic state , and it is traversed within [0,1];

Step 4.2. Map the chaotic variable X ^K generated once into a new chaotic variable X according to formula (2), and map the entire traversal interval [0,1] to the value interval [a,b] of the optimized variable at the same time;

$\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Step 4.3. Use the chaotic variable to perform chaotic search;

The random step size generated by the chaos generator disturbs the flight route within the range [a,b] of the drone’s mission, and the flight route is based on the deviation degree L _i of the target point and the initial point in real time and the unmanned Calculate the _fitness value f(X) of the road section individuals in the flight route, and sort the fitness values of the road section individuals from high to low, and eliminate 10% of the road section individuals with low fitness values, leaving the high fitness values 90% of individual road sections;

Step 4.4. Repeat steps 4.1 to 4.3 until the number of individuals in the road section reaches the set scale, and combine them into an initial population;

Step 4.5. Randomly select two paired individuals among the 90% individuals with high fitness in the initial population, and perform the crossover operation according to the chaotic crossover rule;

Step 4.6. The paired individuals are mutated according to the law of chaotic variation to obtain multiple mutated individuals, and the multiple mutated individuals are combined with the top 10% individuals in the initial population to form the second generation population;

Step 4.7. Delete one of the two identical individuals in the second-generation population, and eliminate the bottom 10% individuals in the second-generation population in the fitness ranking, and obtain 90% of individuals with higher fitness in the second-generation population;

Step 4.8. Perform chaotic genetic operations on 90% of the individuals with high fitness in the second generation population, generate multiple genetic individuals, and form a genetic population, eliminate 90% of the individuals with low fitness in the genetic population, and obtain the genetic population with high fitness 10% individual;

Step 4.9. Decode the 10% individuals with high fitness in the genetic population, and plan to obtain the final UAV flight path.

The beneficial effect of adopting the above further scheme is: add chaotic operation to the genetic algorithm, use chaotic sequence to control the crossover and mutation in the genetic operation, to replace the original crossover and mutation operation that is completely random under a certain probability, and avoid complete randomness. The blindness of operation; effectively prevent and overcome the inaccurate screening caused by the reduction of population diversity in evolution.

Further, the fitness value calculation is specifically as follows:

$\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\mathrm{<span\; class="notranslate">\; L</span>}\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; *</span>\frac{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ;</span>$

in is the penalty coefficient for the degree of deviation of the path from the line connecting the target point and the initial point;

Li is the distance from the i-th path point to the target point, and d0 is the length of the route segment, which is usually determined according to the navigation method of the UAV.

(Ni)*d0 is the distance along the direction of the initial point and the target line, l _i is the actual distance of the path segment; where is the security penalty coefficient;

di, min is the shortest distance between the i-th path segment and the threat, and C _i is the penalty coefficient that the deflection angle between the target point of the path and its previous point satisfies the constraint.

The beneficial effect of adopting the above-mentioned further solution is: to improve the accuracy of the fitness value calculation and ensure the optimal route selection.

Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: a UAV emergency communication path planning system, comprising:

The division module is used for performing three-dimensional grid division on the flying space of the UAV, obtaining grid nodes, constructing a grid diagram, marking the starting point and target point of the UAV on the grid diagram, and marking no one Radar distribution and terrain information in the flight space of the aircraft;

The modeling module is used to construct a flight threat model according to the radar distribution and terrain information in the flight space of the UAV, and to construct a flight limitation model according to the flight technical parameters of the UAV;

Constructing a route module for constructing a flight route between the starting point and the target point of the drone according to the flight threat model and the flight restriction model;

The screening module is used to optimize the flight route through the chaotic genetic algorithm to determine the final flight route.

The beneficial effect of the present invention is: it can safely avoid the interference of low-altitude obstacles and radar, the flight time to reach the designated destination is the shortest and the path is optimal, and at the same time, on the basis of the original genetic algorithm, the chaotic sequence is used to control the crossover and variation in the genetic operation , to replace the original completely random crossover and mutation operations, to accurately determine whether to perform crossover or mutation operations and to determine the specific position of the crossover or mutation operations, avoiding the blindness of completely random operations, and making the solution obtained with high accuracy , the convergence speed is fast, and at the same time, this technical solution has better real-time and rapidity, so that when the UAV performs tasks such as emergency communication and air rescue, the searched track is closer to the actual optimal track of the UAV .

Description of drawings

Fig. 1 is the flow chart of a kind of unmanned aerial vehicle emergency communication path planning method of the present invention;

Fig. 2 is a module block diagram of a UAV emergency communication path planning system according to the present invention.

In the accompanying drawings, the list of parts represented by each label is as follows:

1. Division module, 2. Modeling module, 3. Route building module, 4. Screening module.

detailed description

The principles and features of the present invention are described below in conjunction with the accompanying drawings, and the examples given are only used to explain the present invention, and are not intended to limit the scope of the present invention.

As shown in Figure 1, a UAV emergency communication path planning method includes the following steps:

Step 1. Carry out three-dimensional grid division on the flying space of the UAV, obtain grid nodes, construct a grid diagram, mark the starting point and target point of the UAV on the grid diagram, and mark the flight of the UAV Radar distribution and terrain information in space;

Step 2. Construct a flight threat model according to the radar distribution and terrain information in the flight space of the UAV, and construct a flight limitation model according to the flight technical parameters of the UAV;

Step 3. Construct a flight route between the starting point and the target point of the drone according to the flight threat model and the flight limitation model;

Step 4. Optimizing the flight route through the chaotic genetic algorithm to obtain the final flight route.

Preferably, the flight threat model is specifically:

min W＝k ₁ J _threat +k ₂ J _fuel ;

Where J _threat is radar threat cost, k ₁ (k ₁ ∈ (0, 1)) is the weight of radar threat cost, J _fuel is fuel cost, k ₂ (k ₂ ∈ (0, 1)) is the weight of fuel cost ; The radar threat cost is the limitation of the flying height of the UAV. If the flying height exceeds the set height, it is easy to be detected by the radar, and there is a risk of being tracked, which is what the UAV needs to avoid. The fuel cost is the fuel loaded by the UAV itself. The weight of the drone can meet the fuel demand for the round-trip flight of the UAV. If the flight distance is too long, it will easily cause the UAV to fail to return, which is what the UAV needs to avoid.

Preferably, the flight-limited model includes:

The direct flight distance r of the UAV when changing the attitude is smaller than the distance S _min between the starting point and the target point of the UAV, and r<S _min .

Preferably, the flight-limited model includes:

The hovering time T _min of the UAV when an obstacle is detected is less than the total flight time t of the UAV, and T _min < t.

Preferably, the flight-limited model includes:

The altitude H of the flight route is greater than the minimum flight altitude H _min set for the drone flight, and smaller than the maximum flight altitude H _max set for the drone flight, and H _min <H<H _max .

Preferably, the flight-limited model includes:

The distance L of the flight route is less than half of the flight limit distance of the UAV, and L<L _max /2.

Preferably, the flight limitation model includes: the UAV flies along the positive direction of the x coordinate axis, the coordinates of the current node are (x1, y1, z1), and the coordinates of the next node are (x2, y2, z2) , the maximum deflection angle Ф of the UAV in the horizontal direction and vertical direction satisfies:

$\left\{\begin{array}{c}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; arctan</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ <span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; arctan</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\end{array}\right.<span\; class="notranslate">\; ;</span>$

Among them, Ф _max is the limit angle of UAV horizontal deflection; у _max is the limit angle of UAV vertical deflection.

Preferably, optimizing the flight route through the chaotic genetic algorithm specifically includes the following steps:

Step 4.1. Use the chaotic motion Logistic mapping equation x _k+1 =ux _k (1-x _k )(1) to randomly generate the initial chaotic sequence, where u is the control parameter. When u=4, the above formula is completely in a chaotic state , and it is traversed within [0,1];

Step 4.2. Map the chaotic variable X ^K generated once into a new chaotic variable X according to formula (2), and map the entire traversal interval [0,1] to the value interval [a,b] of the optimized variable at the same time;

$\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Step 4.3. Use the chaotic variable to perform chaotic search;

The random step size generated by the chaos generator disturbs the flight route within the range [a,b] of the drone’s mission, and the flight route is based on the deviation degree L _i of the target point and the initial point in real time and the unmanned Calculate the _fitness value f(X) of the road section individuals in the flight route, and sort the fitness values of the road section individuals from high to low, and eliminate 10% of the road section individuals with low fitness values, leaving the high fitness values 90% of individual road sections;

Step 4.4. Repeat steps 4.1 to 4.3 until the number of individuals in the road section reaches the set scale, and combine them into an initial population;

Step 4.5. Randomly select two paired individuals among the 90% individuals with high fitness in the initial population, and perform the crossover operation according to the chaotic crossover rule;

Step 4.6. The paired individuals are mutated according to the law of chaotic variation to obtain multiple mutated individuals, and the multiple mutated individuals are combined with the top 10% individuals in the initial population to form the second generation population;

Step 4.7. Delete one of the two identical individuals in the second-generation population, and eliminate the bottom 10% individuals in the second-generation population in the fitness ranking, and obtain 90% of individuals with higher fitness in the second-generation population;

Step 4.8. Perform chaotic genetic operations on 90% of the individuals with high fitness in the second generation population, generate multiple genetic individuals, and form a genetic population, eliminate 90% of the individuals with low fitness in the genetic population, and obtain the genetic population with high fitness 10% individual;

Step 4.9. Decode the 10% individuals with high fitness in the genetic population, and plan to obtain the final UAV flight path.

The segment individual is a flight segment in the entire flight route, which is used as the gene in the chaotic genetic algorithm;

Chaotic intersection determination operation: selection interval When performing the crossover operation, the selected two pairs of individuals, according to x _k+1 = ux _k (1-x _k ), determine whether to perform crossover by the current value x _k1 of an independent chaotic sequence generated, if x _k1 ∈P _c then carry out crossover operation, otherwise, do not carry out crossover.

To determine the cross position, the line individual is used as a gene, and according to the length of the chromosome in the gene, it is divided into several segments, which can be several bits or 1 bit, and the interval (0,1) is also divided into several sub-intervals, among which Each subinterval corresponds to a segment in the chromosome. For individuals that need to perform crossover operations, according to x _k+1 = ux _k (1-x _k ), the subinterval to which the current value x _k2 of another independent chaotic sequence belongs To determine the position of the gene segment for crossover operation;

Exchange the corresponding gene segments of the two selected chromosomes to generate two new individuals, thus completing the chaotic crossover;

Chaos mutation operation: the determination of mutation is the same as the determination method of crossover, except that the variation interval P _m should generally be smaller than the intersection interval P _c , and only when the current value of the chaotic sequence x _k3 ∈ P _m , the mutation operation is performed;

For the determination of the mutated bit, divide the interval (0,1) into N equal parts, where N is the length of the chromosome, and then determine the mutated gene position according to the sub-interval where the current value x _k4 of the chaotic sequence is located;

For the paired individuals generated by chaotic crossover, the mutation operation is performed according to the law of chaotic mutation; x _i is the i-th individual before the mutation operation, which corresponds to a vector composed of N components, and x _i (j) is the j-th component , x′ _i (j) is the jth component of the _ith individual after mutation, σi (j) is the approximate variation scale of the jth component, and the chaotic variation form is as follows:

x' _i (j) = x _i (j) + σ _i (j) K _j (0, 1); where K (0, 1) is a sequence of changes in the law of chaos.

For the mutant individuals obtained through the above operations, multiple mutant individuals are combined with 10% individuals with high fitness values in the initial population to form a second-generation population. Other similar individuals perform chaotic mutation operations in a large range of probability to ensure the diversity of the group and avoid the algorithm from falling into a local minimum.

Preferably, the fitness value calculation is specifically:

$\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\mathrm{<span\; class="notranslate">\; L</span>}\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; *</span>\frac{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ;</span>$

in is the penalty coefficient for the degree of deviation of the path from the line connecting the target point and the initial point;

Li is the distance from the i-th path point to the target point, and d0 is the length of the route segment, which is usually determined according to the navigation method of the UAV.

(Ni)*d0 is the distance along the direction of the initial point and the target line, l _i is the actual distance of the path segment; where is the security penalty coefficient;

di, min is the shortest distance between the i-th path segment and the threat, and C _i is the penalty coefficient that the deflection angle between the target point of the path and its previous point satisfies the constraint.

As shown in Figure 2, a UAV emergency communication path planning system includes:

The division module 1 is used for performing three-dimensional grid division on the flying space of the drone, obtaining grid nodes, constructing a grid map, marking the starting point and the target point of the drone on the grid map, and marking the unmanned aerial vehicles. Radar distribution and terrain information in the man-machine flight space;

The modeling module 2 is used to construct a flight threat model according to the radar distribution and terrain information in the flight space of the UAV, and to construct a flight limitation model according to the flight technical parameters of the UAV;

Construct route module 3, be used for constructing flight route between the starting point and target point of unmanned aerial vehicle according to flight threat model and flight restriction model;

The screening module 4 is used to screen the flight route through the chaotic genetic algorithm to determine the flight route.

The specific implementation of chaotic genetic algorithm to optimize the flight route:

Initialize the system parameters: the initial value of x _k of the chaotic variable is 0.2; the initial value of the range a and b of the UAV flight task is a=0, b=2000m; the current longest flight distance is =2000m; The initial value of the number of carrier iteration k is 0, and the number of cycles i=1; the initial value x ₀ of the chaotic variable is substituted into x _k+1 = 4x _k (1-x _k ), and the iterative operation is performed, and the value obtained by the above formula is substituted into x _k+1 substitute Obtain the degree of deviation X between the target point and the initial point of the UAV flight, let k=k+1;

Calculate the fitness value of the flight path of the UAV;

Whether the flight path distance of the drone is greater than the 2000m predicted by the flight distance of the drone, if yes, assign the value of the flight distance of the drone to the maximum value of the flight distance prediction of the drone; if not, execute i=i+1 ;

Judging whether the number of cycles i is greater than 20 or whether the number of iterations k of the chaotic variable of a carrier is greater than the number of iterations 100, if so, calculate the fitness value of each path in the flight path, and sort the fitness values of the obtained paths from high to low, and for the low fitness value 10% of individuals are eliminated, leaving 90% of individuals with high fitness; if not, the initial value x ₀ of the chaotic variable is substituted into x _k+1 = 4x _k (1-x _k ) for iterative operation, and then the cycle is performed.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (10)

1. A UAV emergency communication path planning method, is characterized in that, comprises the following steps: Step 1. Carry out three-dimensional grid division on the flying space of the UAV, obtain grid nodes, construct a grid diagram, mark the starting point and target point of the UAV on the grid diagram, and mark the flight of the UAV Radar distribution and terrain information in space; Step 2. Construct a flight threat model according to the radar distribution and terrain information in the flight space of the UAV, and construct a flight limitation model according to the flight technical parameters of the UAV; Step 3. Construct a flight route between the starting point and the target point of the drone according to the flight threat model and the flight limitation model; Step 4. Optimizing the flight route through the chaotic genetic algorithm to determine the final flight route. 2. A kind of unmanned aerial vehicle emergency communication route planning method according to claim 1, is characterized in that, described flight threat model is specifically: min W=k ₁ J _threat +k ₂ J _fuel ; Where J _threat is radar threat cost, k ₁ (k ₁ ∈ (0, 1)) is the weight of radar threat cost, J _fuel is fuel cost, k ₂ (k ₂ ∈ (0, 1)) is the weight of fuel cost . 3. A kind of unmanned aerial vehicle emergency communication route planning method according to claim 1, is characterized in that, described flight limitation model comprises: When the UAV encounters an obstacle, the distance r from which the UAV changes its attitude and flies straight is less than the distance S _min between the starting point of the UAV and the obstacle point, and r<S _min . 4. A kind of unmanned aerial vehicle emergency communication path planning method according to claim 1, is characterized in that, described flight limitation model comprises: The hovering time t of the UAV when an obstacle is detected is greater than the shortest duration T _min of the UAV control response, and T _min <t. 5. A kind of unmanned aerial vehicle emergency communication path planning method according to claim 1, is characterized in that, described flight limitation model comprises: The altitude H of the flight route is greater than the minimum flight altitude H _min set for the drone flight, and smaller than the maximum flight altitude H _max set for the drone flight, and H _min <H<H _max . 6. A kind of unmanned aerial vehicle emergency communication path planning method according to claim 1, is characterized in that, described flight limitation model comprises: The distance L of the flight route is less than half of the flight limit distance of the UAV, and L<L _max /2. 7. A kind of unmanned aerial vehicle emergency communication path planning method according to claim 1, is characterized in that, described flight limitation model comprises: unmanned aerial vehicle flies along the positive direction of x coordinate axis, and the coordinate of current location node is ( x1, y1, z1), the coordinates of the next node are (x2, y2, z2), the maximum deflection angle Ф of the drone in the horizontal and vertical directions satisfies: $\left\{\begin{array}{c}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; arctan</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ <span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; arctan</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\end{array}\right.<span\; class="notranslate">\; ;</span>$ Among them, Ф _max is the limit angle of UAV horizontal deflection; у _max is the limit angle of UAV vertical deflection. 8. A kind of unmanned aerial vehicle emergency communication path planning method according to claim 1, is characterized in that, optimizes flight route by chaotic genetic algorithm and specifically comprises the following steps: Step 4.1. Use the chaotic motion Logistic mapping equation x _k+1 = ux _k (1-x _k ) (1) to randomly generate the initial chaotic sequence, where u is the control parameter. When u=4, the above formula is completely in a chaotic state , and it is traversed within [0,1]; Step 4.2. Map the chaotic variable X ^K generated once into a new chaotic variable X according to formula (2), and map the entire traversal interval [0,1] to the value interval [a,b] of the optimized variable at the same time; $\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$ Step 4.3. Use the chaotic variable to perform chaotic search; The random step size generated by the chaos generator disturbs the flight route within the range [a,b] of the drone’s mission, and the flight route is based on the deviation degree L _i of the target point and the initial point in real time and the unmanned Calculate the _fitness value f(X) of the road section individuals in the flight route, and sort the fitness values of the road section individuals from high to low, and eliminate 10% of the road section individuals with low fitness values, leaving the high fitness values 90% of individual road sections; Step 4.4. Repeat steps 4.1 to 4.3 until the number of individuals in the road section reaches the set scale, and combine them into an initial population; Step 4.5. Randomly select two paired individuals among the 90% individuals with high fitness in the initial population, and perform the crossover operation according to the chaotic crossover rule; Step 4.6. The paired individuals are mutated according to the law of chaotic variation to obtain multiple mutated individuals, and the multiple mutated individuals are combined with the top 10% individuals in the initial population to form the second generation population; Step 4.7. Delete one of the two identical individuals in the second-generation population, and eliminate the bottom 10% individuals in the second-generation population in the order of fitness, so as to obtain 90% of individuals with high fitness in the second-generation population; Step 4.8. Perform chaotic genetic operations on 90% of individuals with high fitness in the second generation population to generate multiple genetic individuals to form a genetic population, eliminate 90% of individuals with low fitness in the genetic population, and obtain high fitness in the genetic population 10% individual; Step 4.9. Decode the 10% individuals with high fitness in the genetic population, and plan to obtain the final UAV flight path. 9. A kind of unmanned aerial vehicle emergency communication route planning method according to claim 8, is characterized in that, fitness calculation is specifically: $\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\mathrm{<span\; class="notranslate">\; L</span>}\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; *</span>\frac{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ;</span>$ in is the penalty coefficient for the degree of deviation of the path from the line connecting the target point and the initial point; Li is the distance from the i-th path point to the target point, and d0 is the length of the route segment, which is usually determined according to the navigation method of the UAV; (Ni)*d0 is the distance along the direction of the initial point and the target line, l _i is the actual distance of the path segment; where is the security penalty coefficient; di, min is the shortest distance between the i-th path segment and the threat, and C _i is the penalty coefficient that the deflection angle between the target point of the path and its previous point satisfies the constraint. 10. A UAV emergency communication path planning system, characterized in that it comprises: The division module (1) is used for carrying out three-dimensional grid division to the flying space of the unmanned aerial vehicle, obtains the grid node, constructs the grid diagram, marks the starting point and the target point of the unmanned aerial vehicle on the described grid diagram, and Mark the radar distribution and terrain information in the UAV flight space; The modeling module (2) is used to construct a flight threat model according to the radar distribution and terrain information in the flight space of the UAV, and simultaneously constructs a flight limitation model according to the flight technical parameters of the UAV; Construct route module (3), be used for constructing flight route between the starting point and target point of unmanned aerial vehicle according to flight threat model and flight restriction model; The screening module (4) is used to screen the flight route through the chaotic genetic algorithm to determine the flight route.