CN105867383A - Automatic collision preventing control method of USV - Google Patents

Abstract

The invention relates to the technical field of maritime searches and rescue and surveys, in particular to an automatic collision preventing control method of a USV, and aims at providing the automatic collision preventing control method of the USV. By means of mutual cooperation among all subsystems of a whole collision preventing system, the unmanned surface vehicle (USV) can automatically search for obstacles and implement collision preventing strategies according to distribution of the obstacles when the USV executes tasks such as rescue and surveys. The method comprises the following steps that 1, environmental image information is collected; 2, the distance between the obstacles and the USV is measured; 3, a GPS measures relative position information of the USV with other USVs, and a tachometer measures the movement velocity of the USV and other USVs; 4, an obstacle judgment system processes received information; 5, the sea clutter situation is transmitted to a genetic algorithm controller; 6, the genetic algorithm controller generates the collision preventing strategies. The automatic collision preventing control method of the USV is suitable for the technical field of the maritime searches and rescue and surveys.

Description

Autonomous collision avoidance control method for USV

Technical Field

The invention relates to the technical field of marine search and rescue and survey, in particular to a method for controlling autonomous collision avoidance of a universal serial bus (USV).

Background

An Unmanned Surface Vehicle (USV) is a small water surface platform which integrates autonomous planning, autonomous navigation and autonomous completion of tasks such as environment sensing, target detection and the like. The marine environment is complicated changeable, and if want to keep unmanned ship intelligence navigation need rely on inside information and the perfect integration and interaction of external environment. An important prerequisite for such fusion and interaction is that unmanned boats can perform autonomous collision avoidance and can complete a specified mission in a complex marine environment. As an important sign of unmanned ship intellectualization, the autonomous collision avoidance technology of the unmanned ship not only reflects the intelligent level of the maritime unmanned ship to a certain extent, but also is an important research content in the key technical field of the unmanned ship. At present, most of researches on collision avoidance of unmanned boats are focused on collision avoidance algorithms, and the mutual cooperation and cooperation among subsystems of the whole collision avoidance system are rarely involved.

Disclosure of Invention

The invention aims to provide a method for controlling autonomous collision avoidance of a USV (Universal Serial bus), which can realize that an unmanned ship can autonomously search obstacles when executing tasks such as search and rescue, exploration and the like through mutual cooperation among subsystems of a whole collision avoidance system, implements a collision avoidance strategy according to the distribution condition of the obstacles, and ensures the safety of the unmanned ship during autonomous navigation.

In order to realize the method for controlling the autonomous collision avoidance of the USV, an autonomous collision avoidance control device of the USV is adopted, and the method comprises the following steps: the system comprises an obstacle detection device, a positioning system and a genetic algorithm controller, wherein the obstacle detection device comprises a camera, an image identification system, a distance meter, a velocimeter and an obstacle judgment system; the positioning system comprises a GPS positioning system and a communication system; the genetic algorithm controller comprises a wind, wave and water flow detection system, a controller and an execution mechanism.

The main factors that generally affect the collision avoidance effect of the unmanned boat are: course of unmanned shipSpeed v of flight₀(ii) a Course of the obstacleSpeed v of flight_T(ii) a Distance D between unmanned ship and barrier_T(ii) a The course intersection angle of the unmanned ship and the obstacle isThe true orientation of the obstacle relative to the unmanned surface is θ_T(ii) a The relative speed of the two ships is v_R(ii) a The relative course of the two ships isThe minimum passing distance when the unmanned ship meets the obstacle is DCPA; the time when the unmanned boat and the obstacle reach the nearest meeting point is TCPA.

A method for controlling autonomous collision avoidance of a USV comprises the following steps:

the camera shoots an environment image within a 360-degree range taking an unmanned boat as a center, and transmits image information to an image recognition system;

secondly, the image recognition system processes the image information and then transmits the suspected obstacle information to the obstacle judgment system, and meanwhile, the range finder measures the distance between the unmanned ship and an obstacle within 500 meters of the unmanned ship and transmits the information to the obstacle judgment system;

thirdly, a GPS positioning system positions the current position of the unmanned ship in real time, a communication system establishes contact with other unmanned ships to obtain relative position information of the unmanned ship and other unmanned ships, a velocimeter measures the movement speed of the unmanned ship and other unmanned ships, the unmanned ships are equivalent to obstacles, and the information is transmitted to an obstacle judgment system;

fourthly, the barrier judgment system processes all the received information and then transmits the barrier data to the genetic algorithm controller;

fifthly, monitoring the sea surface interference condition of the sea area where the unmanned ship is located by a wind, wave and water flow monitoring system in real time, and transmitting data to a genetic algorithm controller;

and sixthly, the genetic algorithm controller works out a collision prevention strategy according to the barrier data and carries out safe collision prevention through an execution mechanism.

The invention has the following beneficial effects:

1. according to the collision avoidance system, the subsystems of the whole collision avoidance system are matched and cooperated with each other, so that the unmanned ship can independently search for the obstacles when performing tasks such as search and rescue, exploration and the like, a collision avoidance strategy is implemented according to the distribution condition of the obstacles, and the safety of the unmanned ship during autonomous navigation is ensured;

2. according to the unmanned ship collision avoidance system, all collision avoidance subsystems of the unmanned ship are reasonably combined together, a series of processes from searching for a nearby navigation environment, collecting and analyzing barrier information to making collision avoidance strategies and avoiding barriers of the unmanned ship are completed, the intellectualization of the unmanned ship is effectively improved, and the workload of operators is reduced;

3. when the unmanned ship executes tasks such as search and rescue, survey and the like, static and dynamic obstacles in the advancing process can be safely prevented from colliding according to the method provided by the invention.

Drawings

FIG. 1 is a block diagram of a USV autonomous collision avoidance control apparatus;

FIG. 2 is a diagram of USV versus obstacle parameters;

FIG. 3 is a USV barrier collision risk resolution flow chart;

FIG. 4 is a flow chart of a genetic algorithm;

fig. 5 is a schematic view of an unmanned boat avoiding obstacles.

Detailed Description

In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention is further described in detail with reference to fig. 1 to 4 and a detailed embodiment thereof, wherein fig. 1 is a schematic top view of a mounting configuration of a USV sensor, fig. 2 is a schematic relative parameter diagram of a USV and an obstacle, and fig. 3 is a flow chart of USV obstacle collision risk calculation.

In a first specific embodiment, a method for autonomous collision avoidance control of an USV according to this embodiment includes the following steps:

the camera shoots an environment image within a 360-degree range taking an unmanned boat as a center, and transmits image information to an image recognition system;

secondly, the image recognition system processes the image information and then transmits the suspected obstacle information to the obstacle judgment system, and meanwhile, the range finder measures the distance between the unmanned ship and an obstacle within 500 meters of the unmanned ship and transmits the information to the obstacle judgment system;

thirdly, a GPS positioning system positions the current position of the unmanned ship in real time, a communication system establishes contact with other unmanned ships to obtain relative position information of the unmanned ship and other unmanned ships, a velocimeter measures the movement speed of the unmanned ship and other unmanned ships, the unmanned ships are equivalent to obstacles, and the information is transmitted to an obstacle judgment system;

fourthly, the barrier judgment system processes all the received information and then transmits the barrier data to the genetic algorithm controller;

fifthly, monitoring the sea surface interference condition of the sea area where the unmanned ship is located by a wind, wave and water flow monitoring system in real time, and transmitting data to a genetic algorithm controller;

and sixthly, the genetic algorithm controller works out a collision prevention strategy according to the barrier data and carries out safe collision prevention through an execution mechanism.

The embodiment has the following beneficial effects:

1. in the embodiment, the subsystems of the whole collision avoidance system are matched with each other, so that the unmanned ship can independently search for the obstacles when performing tasks such as search and rescue, exploration and the like, a collision avoidance strategy is implemented according to the distribution condition of the obstacles, and the safety of the unmanned ship during autonomous navigation is ensured;

2. according to the implementation mode, all collision avoidance subsystems of the unmanned ship are reasonably combined together, a series of processes from searching for a nearby navigation environment, collecting and analyzing barrier information to making collision avoidance strategies and avoiding barriers of the unmanned ship are completed, the intellectualization of the unmanned ship is effectively improved, and the workload of operators is reduced;

3. when the unmanned ship executes tasks such as search and rescue, survey and the like, static and dynamic obstacles in the advancing process can be safely prevented from being collided according to the method of the embodiment.

In a second embodiment, which is a further description of the method for autonomous collision avoidance control of an USV according to the first embodiment, in the first step, the capture period of the captured environmental image is 0.05s, and each capture period is 10 periods, which takes 0.5 s.

In a third embodiment, the third embodiment is a further description of the method for autonomous collision avoidance control of an USV according to the first or second embodiment, and a measurement period of the distance meter in the second step is 0.5 s.

In a fourth embodiment, the present embodiment is further directed to the method for autonomous collision avoidance control of an USV according to any one of the first to third embodiments, and a communication cycle of the communication system in the third step is 0.5 s.

In a fifth embodiment, the present embodiment is further described with respect to the method for autonomous collision avoidance control of an USV according to any one of the first to fourth embodiments, where in the third step, the position information obtained by the GPS positioning system and the communication system is: the geographic coordinate of the unmanned ship is O (x) measured by a GPS positioning system_O,y_O) The communication system obtains the geographic coordinate of the obstacle as T (x)_T,y_T) Wherein the x-axis is north-south and the y-axis is east-west.

In a sixth embodiment, the present embodiment is a further description of the method for autonomous collision avoidance control of an USV according to any one of the first to fifth embodiments, where the motion speed measured by the velocimeter in the third step is: the motion velocity vector of the unmanned ship is v (v)_Ox,v_Oy) Velocity vector of movement of obstacleIs v (v)_Tx,v_Ty)。

In a seventh embodiment, the present embodiment is a further description of the method for autonomous collision avoidance control of an USV according to any one of the first to sixth embodiments, and the information processing in the fourth step includes: and (3) solving the course of the unmanned ship and the barrier according to the information obtained in the previous three steps:

course of the unmanned ship:

wherein:

the relative distance between the unmanned boat and the obstacle is calculated according to the geographic coordinates of the unmanned boat and the obstacle as follows:

$D_T=\sqrt{{(x_T-x_O)}^2+{(y_T-y_O)}^2}$

the true orientation of the obstacle relative to the unmanned surface is θ_T：

${\mathrm{\θ}}_T=arctan\frac{x_T-x_O}{y_T-y_O}+{\mathrm{\α}}_2$

The true orientation of the unmanned vehicle relative to the obstacle is theta₀：

${\mathrm{\θ}}_0=arctan\frac{x_O-x_T}{y_O-y_T}+{\mathrm{\α}}_2$

Wherein,

the phase orientation of the obstacle is α_T：

The relative velocity components of the obstacle relative to the unmanned vehicle in the x, y axes are:

$\left\{\begin{array}{c}{v}_{Rx}=v_{Tx}-v_{Ox}\\ v_{Ry}=v_{Ty}-v_{Oy}\end{array}\right.$

relative heading of the obstacle with respect to the unmanned boat:

wherein:

calculating the nearest meeting distance DCPA between the unmanned ship and the target ship:

calculate the time to reach the meeting closest point TCPA:

wherein v is_RThe relative speed of the two ships is large or small;

resolving the space collision risk u_dtComprises the following steps:

$u_{dt}=\left\{\begin{array}{cc}1& \left|DCPA\right|<d_1\\ \frac{1}{2}-\frac{1}{2}sin\&lsqb;\frac{\mathrm{\&pi;}}{d_2-d_1}-\frac{DCPA(d_2+d_1)}{2}\&rsqb;& d_1<\left|DCPA\right|\&le;d_2\\ 0& d_2<\left|DCPA\right|\end{array}\right.$

in the formula d₁＝1.5ρ(θ_T)，d₂＝2d₁；

ρ(θ_T) Obtained from the following formula

Solving the time collision risk u_tTComprises the following steps:

when TCPA > 0:

$u_{tT}=\left\{\begin{array}{cc}1& TCPA\&le;1\\ {\left(\frac{t_2-TCPA}{t_2-t_1}\right)}^2& t_1\&le;TCPA\&le;t_2\\ 0& TCPA>t_2\end{array}\right.$

when TCPA is less than or equal to 0

$u_{tT}=\left\{\begin{array}{cc}1& TCPA\&le;1\\ {\left(\frac{t_2+TCPA}{t_2-t_1}\right)}^2& t_1\&le;TCPA\&le;t_2\\ 0& TCPA>t_2\end{array}\right.$

Wherein

Obtaining the comprehensive collision risk degree u of the unmanned ship and the barrier according to the time collision risk degree and the space collision risk degree_t：

$u_t=u_{dt}\&CirclePlus;u_{tT}$

u_tThe value is divided into three cases, wherein, if u_dtWhen equal to 0, u_t＝0；

If u_dt≠0，u_tTWhen equal to 0, u_t＝0；

If u_dt≠0，u_tTWhen not equal to 0, u_t＝max(u_dt，u_tT)。

In an eighth embodiment, the present embodiment is further directed to the method for autonomous collision avoidance control of an USV according to any one of the first to seventh embodiments, and the process of the genetic algorithm controller making the collision avoidance policy according to the obstacle data in the fifth step is as follows: as shown in figure 4 of the drawings,

1. encoding the population:

each chromosome represents an initial path of the unmanned ship, in the population initialization process, a population with the number of individuals of n is automatically generated according to the initial position and the target position of the unmanned ship, all feasible paths from the starting point to the end point of the unmanned ship are coded, and each path is an individual in the population;

2. solving a fitness function:

the calculation of the fitness function considers the length of the path, the smoothness of the path and the safety of the path; the following two principles are generally followed when converting the objective function into the fitness function:

(1) the fitness value is non-negative;

(2) the change direction of the objective function in the optimization process is consistent with the change direction of the fitness function in the population evolution process.

For the problem of selecting the collision avoidance path of the unmanned ship, a fitness function with a mapping relation with a target function can be established through the following formula:

F_(x)＝C-f(x)

in the formula, F_(x)Is a fitness function; c is an adjustable parameter whose value is such that the fitness function F_(x)Constantly greater than or equal to 0; f (x) is an objective function of the optimization problem.

To ensure F_(x)Not less than 0, and therefore, the fitness function is established as follows:

$F_{\left(x\right)}=\left\{\begin{array}{cc}{C}_{max}-f\left(x\right)& f\left(x\right)<C_{max}\\ 0& f\left(x\right)\&GreaterEqual;C_{max}\end{array}\right.$

in the formula C_maxIs an adjustable parameter, C_maxThe maximum theoretically possible value of the objective function f (x) can be taken; the selection of the objective function f (x) takes into account the length of the path, the smoothness of the path, and the security of the path.

3. Basic genetic manipulation

The basic genetic operation comprises selection, crossing and mutation, and comprises the following specific operation steps:

(1) selection operation

To select the operationSelecting individuals with high fitness from the population, and eliminating the individuals with low fitness; probability P of individual i being selected_iProportional to its fitness value; adopting a roulette model, comprising the following steps:

1) calculating fitness of each chromosome F_i，1≤i≤n；

2) Integrating fitness values of all chromosomes, and recording intermediate integrated values S_i：S_iS₁＝F₁，S_i＝S_i-1+F_i；

3) Generating a random number x, 0 ≦ x ≦ S_n；

4) Selecting chromosomes if S_i-1＜x≤S_iIf i is more than or equal to 1 and less than or equal to n, the ith chromosome is selected to enter the next generation population;

5) repeating the steps 3) and 4) until enough chromosomes are obtained.

(2) Crossover operation

Crossover operation

Exchanging partial genes of two paired chromosomes according to a certain mode according to the cross probability so as to form two new individuals, wherein the cross rate is 0.5-0.95.

(3) Mutation operation

Mutation operation

And replacing some gene values in the individual codes with other gene values according to the mutation probability so as to form a new individual, wherein the mutation rate is generally set to be between 0 and 0.5, and the gene mutation is carried out with small probability.

In order to verify the beneficial effects of the invention, the following simulation experiments are carried out:

as shown in fig. 5, a total of 5 obstacles are set in the experiment, the position coordinates are (40,30), (80,80), (90,130), (120 ), (150,160), and the radius of the obstacle is 11,11,6,11, 11. The starting point of the unmanned boat is (10,10), and the coordinates of the end point are (190 ).

Wherein the steps of basic genetic manipulation:

1. initialization: mainly setting evolution parameters, setting a maximum evolution algebra and randomly generating n individuals as an initial population P (0);

2. individual evaluation: calculating the fitness of each individual in the population P (t) by a certain large method, wherein t represents an algebra;

3. selecting: acting a selection operator on the population;

4. and (3) crossing: applying a crossover operator to the population;

5. mutation: acting mutation operators on the groups;

6. algorithm termination conditions are as follows: two types can be set, wherein one type is that if the evolution algebra reaches the maximum value, the individual with the maximum fitness obtained in the evolution process is used as the optimal solution output, and the calculation is stopped; and the other method is to set an error, if the error of a certain individual in the population reaches the requirement, the individual with the optimal fitness is output as the optimal approximate solution, and the calculation is terminated.

The experiment realizes the effect of the invention, and when the unmanned ship runs to the target point, the unmanned ship can safely and automatically avoid collision when encountering obstacles on the way through the mutual cooperation of the modules, and can smoothly reach the preset target point.

Claims (8)

1. A method for controlling autonomous collision avoidance of a USV (Universal Serial bus) is realized by adopting an autonomous collision avoidance control device of the USV, and the autonomous collision avoidance control device of the USV comprises the following steps: the system comprises an obstacle detection device, a positioning system and a genetic algorithm controller, wherein the obstacle detection device comprises a camera, an image identification system, a distance meter, a velocimeter and an obstacle judgment system; the positioning system comprises a GPS positioning system and a communication system; the genetic algorithm controller comprises a wind, wave and water flow detection system, a controller and an execution mechanism;

the method is characterized by comprising the following steps of:

the camera shoots an environment image within a 360-degree range taking an unmanned boat as a center, and transmits image information to an image recognition system;

secondly, the image recognition system processes the image information and then transmits the suspected obstacle information to the obstacle judgment system, and meanwhile, the range finder measures the distance between the unmanned ship and an obstacle within 500 meters of the unmanned ship and transmits the information to the obstacle judgment system;

thirdly, a GPS positioning system positions the current position of the unmanned ship in real time, a communication system establishes contact with other unmanned ships to obtain relative position information of the unmanned ship and other unmanned ships, a velocimeter measures the movement speed of the unmanned ship and other unmanned ships, the unmanned ships are equivalent to obstacles, and the information is transmitted to an obstacle judgment system;

fourthly, the barrier judgment system processes all the received information and then transmits the barrier data to the genetic algorithm controller;

fifthly, monitoring the sea surface interference condition of the sea area where the unmanned ship is located by a wind, wave and water flow monitoring system in real time, and transmitting data to a genetic algorithm controller;

and sixthly, the genetic algorithm controller works out a collision prevention strategy according to the barrier data and carries out safe collision prevention through an execution mechanism.

2. The method for autonomous collision avoidance control of an USV according to claim 1, wherein in the first step, the capturing period of the captured environment image is 0.05s, and each capturing period is 10 periods, which takes 0.5 s.

3. The method for autonomous collision avoidance control of an USV according to claim 1 or 2, wherein in step two, the measuring period of the distance meter is 0.5 s.

4. The method as claimed in claim 3, wherein the communication period of the communication system in step three is 0.5 s.

5. The method for autonomous collision avoidance control of an USV according to claim 4, wherein the position information obtained by the GPS positioning system and the communication system in step three is: the geographic coordinate of the unmanned ship is O (x) measured by a GPS positioning system_O,y_O) The communication system obtains the geographic coordinate of the obstacle as T (x)_T,y_T) Wherein the x-axis is north-south and the y-axis is east-west.

6. The method for autonomous collision avoidance control of a USV according to claim 5, wherein the moving speed measured by the speedometer in step three is: the motion velocity vector of the unmanned ship is v (v)_Ox,v_Oy) The obstacle motion velocity vector is v (v)_Tx,v_Ty)。

7. The method for autonomous collision avoidance control of an USV according to claim 6, wherein the specific process of the information processing in step four is as follows: and (3) solving the course of the unmanned ship and the barrier according to the information obtained in the previous three steps:

course of the unmanned ship:

wherein:

the relative distance between the unmanned boat and the obstacle is calculated according to the geographic coordinates of the unmanned boat and the obstacle as follows:

$D_T=\sqrt{{(x_T-x_O)}^2+{(y_T-y_O)}^2}$

the true orientation of the obstacle relative to the unmanned surface is θ_T：

${\mathrm{\&theta;}}_T=\arctan \frac{x_T-x_O}{y_T-y_O}+{\mathrm{\&alpha;}}_2$

The true orientation of the unmanned vehicle relative to the obstacle is theta₀：

${\mathrm{\&theta;}}_0=\arctan \frac{x_O-x_T}{y_O-y_T}+{\mathrm{\&alpha;}}_2$

Wherein,

the phase orientation of the obstacle is α_T：

The relative velocity components of the obstacle relative to the unmanned vehicle in the x, y axes are:

$\left\{\begin{array}{c}{v}_{Rx}=v_{Tx}-v_{Ox}\\ v_{Ry}=v_{Ty}-v_{Oy}\end{array}\right.$

relative heading of the obstacle with respect to the unmanned boat:

wherein:

calculating the nearest meeting distance DCPA between the unmanned ship and the target ship:

calculate the time to reach the meeting closest point TCPA:

wherein v is_RThe relative speed of the two ships is large or small;

resolving the space collision risk u_dtComprises the following steps:

$u_{dt}=\left\{\begin{array}{cc}1& \left|DCPA\right|<d_1\\ \frac{1}{2}-\frac{1}{2}sin\&lsqb;\frac{\mathrm{\&pi;}}{d_2-d_1}-\frac{DCPA(d_2+d_1)}{2}\&rsqb;& d_1<\left|DCPA\right|\&le;d_2\\ 0& d_2<\left|DCPA\right|\end{array}\right.$

in the formula d₁＝1.5ρ(θ_T)，d₂＝2d₁；

ρ(θ_T) Obtained from the following formula

Solving the time collision risk u_tTComprises the following steps:

when TCPA > 0:

$u_{tT}=\left\{\begin{array}{cc}1& TCPA\&le;1\\ {\left(\frac{t_2-TCPA}{t_2-t_1}\right)}^2& t_1\&le;TCPA\&le;t_2\\ 0& TCPA>t_2\end{array}\right.$

when TCPA is less than or equal to 0

$u_{tT}=\left\{\begin{array}{cc}1& TCPA\&le;1\\ {\left(\frac{t_2+TCPA}{t_2-t_1}\right)}^2& t_1\&le;TCPA\&le;t_2\\ 0& TCPA>t_2\end{array}\right.$

Wherein

Obtaining the comprehensive collision risk degree u of the unmanned ship and the barrier according to the time collision risk degree and the space collision risk degree_t：

$u_t=u_{dt}\&CirclePlus;u_{tT}$

u_tThe value is divided into three cases,wherein, if u_dtWhen equal to 0, u_t＝0；

If u_dt≠0，u_tTWhen equal to 0, u_t＝0；

If u_dt≠0，u_tTWhen not equal to 0, u_t＝max(u_dt，u_tT)。

8. The method for autonomous collision avoidance control of an USV according to claim 7, wherein the process of the genetic algorithm controller to make a collision avoidance strategy according to the obstacle data in the fifth step is:

1. encoding the population:

each chromosome represents an initial path of the unmanned ship, in the population initialization process, a population with the number of individuals of n is automatically generated according to the initial position and the target position of the unmanned ship, all feasible paths from the starting point to the end point of the unmanned ship are coded, and each path is an individual in the population;

2. solving a fitness function:

the calculation of the fitness function considers the length of the path, the smoothness of the path and the safety of the path; the following two principles are followed when converting the objective function into the fitness function:

(1) the fitness value is non-negative;

(2) the change direction of the target function in the optimization process is consistent with the change direction of the fitness function in the population evolution process;

for the problem of selecting the collision avoidance path of the unmanned ship, a fitness function with a mapping relation with a target function is established through the following formula:

F_(x)＝C-f(x)

in the formula, F_(x)Is a fitness function; c is an adjustable parameter whose value is such that the fitness function F_(x)Constantly greater than or equal to 0; (x) is an objective function of the optimization problem;

to ensure F_(x)Not less than 0, the fitness function is established as follows:

$F_{\left(x\right)}=\left\{\begin{array}{cc}{C}_{max}-f\left(x\right)& f\left(x\right)<C_{max}\\ 0& f\left(x\right)\&GreaterEqual;C_{max}\end{array}\right.$

in the formula C_maxIs an adjustable parameter, C_maxTaking the maximum value theoretically possible for the objective function f (x); the length of the path, the smoothness of the path and the safety of the path are considered in the selection of the target function f (x);

3. basic genetic manipulation

The basic genetic operation comprises selection, crossing and mutation, and comprises the following specific operation steps:

(1) selection operation

Selecting individuals with high fitness from the population, and eliminating individuals with low fitness; probability P of individual i being selected_iProportional to its fitness value; adopting a roulette model, comprising the following steps:

1) calculating fitness of each chromosome F_i，1≤i≤n；

2) Integrating fitness values of all chromosomes, and recording intermediate integrated values S_i：S_iS₁＝F₁，S_i＝S_i-1+F_i；

3) Generating a random number x, 0 ≦ x ≦ S_n；

4) Selecting chromosomes if S_i-1＜x≤S_iIf i is more than or equal to 1 and less than or equal to n, the ith chromosome is selected to enter the next generation population;

5) repeating steps 3) and 4) until enough chromosomes are obtained;

(2) crossover operation

Crossover operation

Exchanging partial genes of two paired chromosomes according to a certain mode according to the cross probability so as to form two new individuals, wherein the cross rate is 0.5-0.95;

(3) mutation operation

Mutation operation

And replacing some gene values in the individual codes with other gene values according to the mutation probability so as to form a new individual, wherein the mutation rate is set to be between 0 and 0.5, and the gene mutation is carried out with small probability.