CN115420289B - Unmanned ship route planning method based on particle swarm improvement artificial potential field method - Google Patents
Unmanned ship route planning method based on particle swarm improvement artificial potential field method Download PDFInfo
- Publication number
- CN115420289B CN115420289B CN202210988311.8A CN202210988311A CN115420289B CN 115420289 B CN115420289 B CN 115420289B CN 202210988311 A CN202210988311 A CN 202210988311A CN 115420289 B CN115420289 B CN 115420289B
- Authority
- CN
- China
- Prior art keywords
- particle
- unmanned ship
- potential field
- particle swarm
- artificial potential
- 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.)
- Active
Links
- 239000002245 particle Substances 0.000 title claims abstract description 150
- 238000000034 method Methods 0.000 title claims abstract description 93
- 230000006872 improvement Effects 0.000 title claims description 10
- 238000005457 optimization Methods 0.000 claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 230000006870 function Effects 0.000 claims description 32
- 238000004364 calculation method Methods 0.000 claims description 15
- 230000006978 adaptation Effects 0.000 claims description 10
- 101100412093 Schizosaccharomyces pombe (strain 972 / ATCC 24843) rec16 gene Proteins 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 244000144992 flock Species 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000004888 barrier function Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
- G01C21/203—Specially adapted for sailing ships
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses an unmanned ship route planning method based on a particle swarm improved artificial potential field method, which comprises the following steps: step 1, analyzing an optimization target of unmanned ship route planning, and designing an adaptability function in a particle swarm algorithm; step 2, initializing a particle swarm; step 3, planning a route by adopting an improved artificial potential field method according to the particle swarm parameters to obtain N routes; calculating an individual fitness value of each particle by adopting the fitness function; step 4, updating the particle swarm according to the individual fitness value; step 5, checking whether the global optimal value of the particle swarm meets an end condition or whether the iteration number is equal to the maximum particle swarm optimizing algebra, if so, entering a step 6, otherwise, returning to the step 3; and 6, outputting the finally optimized particle swarm parameters, and re-planning the route by using the parameters and adopting an improved artificial potential field method, outputting an optimal route, and completing the unmanned ship route planning.
Description
Technical Field
The invention relates to an unmanned ship route planning method, in particular to an unmanned ship route planning method based on a particle swarm improved artificial potential field method.
Background
When the unmanned aerial vehicle executes the task, the unmanned aerial vehicle can not avoid the occurrence of barriers on the route planned in advance, the unmanned aerial vehicle is required to automatically and locally adjust the route according to the barrier information, and the problem of autonomous collision avoidance of the unmanned aerial vehicle can be effectively solved under most conditions by adopting the existing improved artificial potential field method. However, when the center of the obstacle is not on the line connecting the unmanned ship and the specific position, the artificial potential field method using the improved potential field repulsive force function can also plan a path to finally reach the target point, but when the unmanned ship approaches the obstacle, the repulsive force is rapidly increased, so that a larger corner occurs in the planned path. Unmanned boats are typically under-actuated structures, which make the smoothness of the planned route noticeable during path planning, because if the path is not smooth, the unmanned boat cannot change a large heading angle in a short time, and may hit an obstacle. Meanwhile, frequent changes of the heading or changes of the larger heading in a short time are also waste of energy for the unmanned ship, and adverse effects are brought to the endurance of the unmanned ship. In addition, when the artificial potential field method is used for planning a path, in order to better avoid the obstacle, the influence range of the obstacle is often enlarged, and meanwhile, the unmanned ship can sail away from the obstacle as far as possible, so that the planned path is overlong, and the energy consumption is increased. Therefore, it is an important matter to investigate how to shorten the path length and reduce the corner size while effectively avoiding obstacles, and to improve the evading ability of the unmanned ship.
In the traditional artificial potential field method, the setting of the gravitational gain constant k f and the repulsive gain constant m f has no specific requirement, and a method of artificial setting is generally adopted in simulation, so that the value of the two constants has limitation, and if the two parameters are improperly set, the smoothness of a planned path is poor.
Disclosure of Invention
The invention aims to: the invention aims to solve the technical problem of providing an unmanned ship route planning method based on a particle swarm improved artificial potential field method aiming at the defects of the prior art.
In order to solve the technical problems, the invention discloses an unmanned ship route planning method based on a particle swarm improved artificial potential field method, which comprises the following steps:
step 1, analyzing an optimization target of unmanned ship route planning, and designing an adaptability function in a particle swarm algorithm;
Step 2, initializing a particle swarm, including: the population scale N, the position x i and the speed v i of each particle, the artificial potential field parameters, the particle group, namely the initial local optimal value and the global optimal value of the population, and the maximum particle group optimizing algebra G and the searching range are set; the artificial potential field parameters include: an attraction gain constant and a repulsion gain constant;
Step3, planning a route by adopting an improved artificial potential field method according to the particle swarm parameters to obtain N routes; calculating an individual fitness value of each particle by adopting the fitness function according to the navigation information of the unmanned ship;
Step 4, updating the particle swarm according to the calculated individual fitness value of each particle; updating the population of particles includes: updating individual extremum of the particles and population extremum of the particle swarm, and updating position and speed information of the particles;
wherein updating the individual extremum of the particles and the population extremum of the particle population comprises: taking the optimal fitness value in the history record of each particle as the current individual extremum of the particle, taking the historical optimal fitness value of all particles as the population extremum of the population, and recording the corresponding position information;
Updating position and velocity information of the particles, including: updating the position and speed information of each particle, and ensuring that the particles do not exceed the search limit according to the limit range;
step 5, checking whether the global optimal value of the particle swarm meets an end condition or whether the iteration number is equal to the maximum particle swarm optimization algebra G, if the condition is met, entering a step 6, otherwise, returning to the step 3;
And 6, outputting the finally optimized artificial potential field parameters, and re-planning the route by using the parameters and adopting an improved artificial potential field method, outputting an optimal route and finishing the unmanned ship route planning.
The unmanned ship route planning method based on the particle swarm improvement artificial potential field method adopts a particle swarm algorithm to optimize two parameters, namely an attraction gain constant and a repulsion gain constant in the improved artificial potential field method.
The fitness function described in step 1 of the present invention includes:
Constructing the fitness function by adopting the relative length f 1, the relative offset f 2 and the corner coefficient f 3 of the unmanned ship route; wherein the relative length f 1 represents the ratio of the sum of the distances between all the path points on the path to the unmanned ship bollard, the relative offset f 2 represents the ratio of the maximum value of the distances between the connecting lines from the starting point and the end point of all the path points in the repulsive force influence range to the unmanned ship bollard, and the rotation angle coefficient f 3 represents the maximum value of the rotation angles between the adjacent path points on the path;
let num denote unmanned ship path point quantity, L denote unmanned ship motion step length, L denote unmanned ship length, then the calculation formula of relative length f 1 of the course is:
The connection starting point and the target point form a straight line, and the expression of the straight line is set as follows:
Ax+By+C＝0
wherein A, B and C represent parameters of the linear equation, which are determined by coordinates of the starting point and the target point; x and y represent the abscissa and ordinate of the point on the line;
Let (x t,yt) denote the position coordinates of the unmanned ship at time t, the calculation formula of the relative offset f 2 is:
the calculation formula of the rotation angle coefficient f 3 is as follows:
Wherein (x t+1,yt+1) represents the position coordinate of the unmanned ship at the time t+1, and (x t-1,yt-1) represents the position coordinate of the unmanned ship at the time t-1;
The fitness function F is:
Where α is a weight of a relative length, β is a weight of a relative offset, and γ is a weight of a rotation angle coefficient, and satisfies α+β+γ=1.
The fitness function in the step 1 of the invention comprehensively considers the relative length, the relative offset and the corner coefficient as optimization targets, and if only one of the optimization targets is optimized, the weights of the other optimization targets are set to be zero.
The method for initializing the particle swarm in the step 2 of the invention comprises the following steps:
regarding each particle in the particle group as a vector, the position of the ith particle Flight speed in solution spaceThe expression is as follows:
wherein m represents the dimension of the target search space, and N represents the size of the particle swarm;
According to the method, two parameters of an attractive force gain constant k f and a repulsive force gain constant m f in an improved artificial potential field method are optimized by adopting a particle swarm algorithm, a target search space dimension m=2 is set, and:
the initialization setting of the particle swarm is completed.
The method for planning the route by adopting the improved artificial potential field method in the step 3 of the invention comprises the following steps:
Step 3-1, constructing a virtual potential field model, initializing parameters, and setting position information of a departure point of the unmanned ship, position information of an obstacle and position information of a target point;
Step 3-2, judging the position information of the obstacle influencing the unmanned ship, calculating repulsive force and attractive force according to an improved artificial potential field method, and calculating resultant force and resultant force direction, namely an included angle theta between the resultant force and an x-axis;
Step 3-3, the method for calculating the next position information of the unmanned ship comprises the following steps:
xt+1＝xt+l×cos(θ)
yt+1＝yt+l×sin(θ)
wherein l represents the advancing distance of the unmanned ship, and theta represents the included angle between the resultant force and the x axis;
step 3-4, judging whether the unmanned ship has reached the target point, and if the unmanned ship has not reached the target point, executing step 3-1; otherwise, ending the route planning.
The method for calculating the gravitation according to the improved artificial potential field method in the step 3-2 comprises the following steps:
Assuming that the unmanned ship is located at the position denoted as q= (x, y) T, the method of calculating the attraction force is as follows:
The force potential function is expressed as:
Wherein U att (q) represents the gravitational potential field, k f represents the gravitational gain constant, q goal represents the position of the target point, and ρ (q, q goal) represents the distance between the unmanned ship and the target point; the attraction force of the unmanned ship is a negative gradient of the attraction function, and the direction of the attraction force is pointed to a target point by the unmanned ship, which is expressed as:
Wherein F att (q) represents the attraction force applied by the unmanned ship, and gard () represents the gradient calculation symbol.
The method for calculating repulsive force according to the improved artificial potential field method in the step 3-2 comprises the following steps:
Wherein U rep (q) represents a repulsive force potential field, m f represents a repulsive force gain constant, q 0 represents the position of an obstacle, q represents the real-time position of the unmanned ship, q goal represents the position of a target point, ρ (q, q 0) represents the distance between the unmanned ship and the obstacle, and ρ 0 represents the influence range of the repulsive force generated by the obstacle; wherein, unmanned ship and the distance of target point represent:
after the repulsive force potential field function is improved, the repulsive force and the attractive force are no longer acted on the same straight line, and the repulsive force is decomposed and expressed as follows:
Wherein F rep1 (q) represents the component of the repulsive force applied by the unmanned ship along the direction of the attractive force, and the direction is from the obstacle to the unmanned ship; f rep2 (q) represents another component of the repulsive force experienced by the unmanned aerial vehicle, the direction being directed by the unmanned aerial vehicle to the target point;
the decomposition method of the two components is as follows:
where, |f rep1 | and |f rep2 | represent the modes of the two repulsive force components, respectively.
The updating method of the individual extremum and the group extremum in the step 4 comprises the following steps:
the optimal position experienced by the ith particle itself is noted as:
The individual extremum is p i _best;
The optimal positions experienced by the whole population are noted as:
the population extremum is g_best;
for each particle, comparing the adaptation value with an individual extremum p i _best, and if the adaptation value is better than p i _best, taking the position of the particle at the moment as the current optimal position;
For each particle, the adaptation value is compared with the global extremum g_best, and if the adaptation value is better than g_best, the position of the particle at the moment is taken as the current group optimal position.
The method for updating the position and speed information of the particles in the step 4 comprises the following steps:
the velocity update for each particle is expressed as:
vi(t+1)＝w×vi(t)+c1×rand()×[pi-best-xi(t)]+c2×rand()×[g-best-xi(t)]
Wherein rand () is a random number between (0, 1), x i (t) is the current position of the particle, w is the inertial weight, c 1 and c 2 are learning factors; the maximum limiting speed v max;vi (t+1) of the particle represents the speed of the particle at time t+1, and v i (t) represents the speed of the particle at time t;
The location update for each particle is expressed as:
xi(t+1)＝xi(t)+vi(t+1)
Wherein x i (t+1) represents the position of the particle at the next moment.
The beneficial effects are that:
compared with the prior art, the invention has the remarkable advantages that:
(1) Compared with the conventional method for manually adjusting simulation parameters to perform optimization, a large amount of simulation time can be saved, the search range and the search precision are enlarged, and the optimal solution is found.
(2) According to the invention, the weight of each optimization target in the fitness function can be flexibly adjusted according to the actual task requirement of the unmanned ship, so that the current most suitable collision prevention scheme is obtained.
(3) The particle swarm optimization is combined to optimize two parameters in the artificial potential field method, so that the problems of overlarge rotation angle and overlong path in the autonomous collision prevention process of the unmanned ship can be effectively solved, the collision prevention success rate can be effectively improved, and unnecessary energy consumption can be reduced.
Drawings
The foregoing and/or other advantages of the invention will become more apparent from the following detailed description of the invention when taken in conjunction with the accompanying drawings and detailed description.
FIG. 1 is a schematic flow chart of the method of the present invention.
Fig. 2 is a schematic flow chart of optimization parameters of a particle swarm algorithm.
Fig. 3 is a schematic illustration of the unmanned boat subjected to virtual forces.
Fig. 4 is a schematic diagram of unmanned boat stress analysis.
Fig. 5 is a schematic diagram of an artificial potential field method route planning flow.
Detailed Description
The invention will be further described with reference to the accompanying drawings and examples. As shown in fig. 1, an unmanned ship route planning method based on a particle swarm improvement artificial potential field method comprises the following steps:
Design of fitness function
The invention adopts three coefficients of the relative length f 1, the relative offset f 2 and the corner coefficient f 3 of the route to construct the fitness function, wherein the length coefficient f 1 represents the sum of the distances between all the separated path points on the route, the offset coefficient f 2 represents the maximum value of the distances between all the path points in the repulsive force influence range and the connecting line of the starting point and the ending point, and the corner coefficient f 3 represents the maximum value of the corners between the adjacent path points on the route.
Let n denote unmanned ship path point quantity, L denote unmanned ship motion step length, L denote unmanned ship length, then the calculation formula of the relative length f 1 of the course is:
connecting the starting point and the target point, and setting the expression of the straight line as follows:
Ax+By+C＝0 (2)
Let (x t,yt) denote the position coordinates of the unmanned ship at time t, the calculation formula of the relative offset f 2 is:
the calculation formula of the maximum rotation angle f 3 is:
in summary, the fitness function F can be obtained as:
Where α, β, γ are weights of the corresponding coefficients, and α+β+γ=1 is satisfied.
And (5) comprehensively considering optimization targets such as the route length, the route offset, the rotation angle and the like, and if only one of the optimization targets is optimized, taking the reciprocal of the coefficient as an fitness function.
(II) parameter optimization based on particle swarm optimization
(1) Particle swarm optimization flow
The searching process of the particle swarm algorithm is similar to the predation behavior of birds, particles represent birds, the searching range is the area where birds search for food, and a piece of food exists in the area, namely the global optimal solution. Birds in the flock initially scatter as individuals in this area, and some birds fly toward the food when they are closer to the food, and other flocks fly in the direction of the bird due to the collective behavior of the flock, searching for the vicinity of the bird until the piece of food is found. Particle swarm algorithm.
The key elements of the particle swarm algorithm mainly comprise an initial particle swarm, a fitness function, an individual extremum, a population extremum and the like, and each element is essential in the particle swarm algorithm process. The steps of the particle swarm algorithm are shown in FIG. 2.
The first step: randomly generating a particle population as a primary solution of the problem (it is necessary to ensure that the primary solution is randomly generated to ensure the diversity of individual genes);
And a second step of: designing a proper fitness function according to a specific problem to calculate the fitness value of an individual;
and a third step of: searching an individual extremum and a population extremum;
fourth step: updating the particle position and speed information;
fifth step: updating the individual extremum and the population extremum;
sixth step: judging whether the maximum particle swarm algebra is reached, if so, outputting an optimization result, and stopping the program; if not, returning to the second step to continue operation;
(2) Parameter optimization based on particle swarm optimization
As can be seen from equation (5), when the planned path length is longer, the maximum offset distance is larger, and the maximum rotation angle is larger, the value of the fitness function is smaller, so that we can judge the merits of the planned path through the fitness function value, and the larger value indicates the higher path quality.
The specific steps of parameter optimization are as follows:
1) Initializing a population of particles
The population of particles is initialized, including a population size N, a position x i and a velocity v i for each particle.
Each particle can be regarded as a vector, and the position of the ith particle and the flight speed in the solution space can be expressed as:
Wherein m is the dimension of the particle swarm target search space.
In the present invention, the optimization is to improve two parameters of the gravitational gain constant k f and the repulsive gain constant m f of the artificial potential field method, so m=2, and:
initializing population individual optimal values and population optimal values according to experience, and setting maximum particle swarm optimization algebra G and search range.
2) Individual fitness value calculation
According to the initialized particle swarm and parameters, the improved artificial potential field method is adopted to complete the route planning, N routes are obtained, according to the route point information, the adaptive value of each individual is calculated by adopting a formula (5), and the specific route planning method is shown in (III) the detailed description in the route planning part based on the improved artificial potential field algorithm.
3) Selecting individual extremum and population extremum
The optimal position experienced by each particle in the flight process is the optimal solution found by the particle itself, and the corresponding adaptation value is called an individual extremum and is the flight experience of the particle itself. The optimal position experienced by the whole population is the optimal solution found by the whole population at present, and the corresponding adaptation value is called global extremum and is the experience of the companion.
The optimal position experienced by the ith particle itself is noted as:
the individual extremum is p i _best.
The optimal positions experienced by the whole population are noted as:
The population extremum is g_best.
For each particle, its fitness value is compared to the individual extremum p i _best, and if p i _best is better than that, the position of the particle at that time is taken as the current best position.
For each particle, its fitness value is compared to the global extremum g_best, and if it is better than g_best, the position of the particle at that time is taken as the current best position of the population.
4) Updating position and velocity information for each particle
The location update of each particle can be expressed as:
Where rand () is a random number between (0, 1), x i (t) is the current position of the particle, w is the inertial weight, c 1 and c 2 are the learning factors. Maximum limiting velocity of particles v max(vmax > 0).
The location update of each particle can be expressed as:
xi(t+1)＝xi(t)+vi(t+1) (11)
5) Judging whether to finish optimization
Judging whether the maximum optimizing algebra G of the particle swarm is reached. If the maximum optimizing algebra of the particle swarm is reached, outputting k f and m f; if the maximum particle swarm algebra is not reached, the population X is updated, and the particle swarm operation is performed on the population again.
And (3) checking whether the optimal value meets the end condition or is equal to the maximum evolution algebra, stopping iteration and outputting the finally optimized artificial potential field parameter if the optimal value is equal to the maximum iteration number, then carrying out route planning again by adopting an improved artificial potential field method by utilizing the parameter, and outputting the optimal route, otherwise, continuing the step (3).
(III) improved artificial potential field algorithm-based route planning
The method has the core ideas that the working environment of the unmanned ship is abstracted into a virtual potential field, in the potential field model, a target point generates attractive force to the unmanned ship, an obstacle generates repulsive force to the unmanned ship, then the two potential fields are subjected to vector superposition, and finally the unmanned ship is guided to move by calculating the magnitude and the direction of resultant force, as shown in figure 3.
(1) Virtual force calculation
Assuming that the unmanned boat is located at a position denoted as q= (x, y) T, then
1) Gravitational field
The gravitational potential function can be expressed as:
Wherein: u att (q) denotes the gravitational potential field, k f denotes the gravitational gain constant, q goal denotes the position of the target point, and ρ (q, q goal) denotes the distance of the unmanned boat from the target point. The attraction force of the unmanned aerial vehicle is a negative gradient of the attraction function, and the direction of the attraction force is pointed to a target point by the unmanned aerial vehicle, so the attraction force can be expressed as:
2) Repulsive force field
Aiming at the problem of local minimum of the traditional artificial potential field method, scholars at home and abroad put forward various improvement methods for the repulsive potential field, the repulsive potential field adopted by the invention is as follows:
Where U rep (q) represents a repulsive force potential field, m f represents a repulsive force gain constant, q 0 represents a position of an obstacle, q represents a real-time position of the unmanned ship, q goal represents a position of a target point, ρ (q, q 0) represents a distance between the unmanned ship and the obstacle, and ρ 0 represents an influence range of the repulsive force generated by the obstacle. The distance between the unmanned ship and the target point can be expressed as:
After the repulsive function is improved, the repulsive force is no longer in the same line with the attractive force, so that the repulsive force needs to be decomposed, which is expressed as follows:
Wherein F rep1 (q) represents the component of the repulsive force applied by the unmanned ship along the direction of the attractive force, and the direction is from the obstacle to the unmanned ship; f rep2 (q) represents another component of the repulsive force experienced by the unmanned aerial vehicle, the direction being directed by the unmanned aerial vehicle to the target point.
The decomposition method of the two components is shown in fig. 4, and their calculation formulas are as follows:
where, |f rep1 | and |f rep2 | represent the modes of the two repulsive force components, respectively.
(2) Route planning based on artificial potential field method
The steps of the path planning algorithm for the unmanned ship by using the artificial potential field method are as shown in fig. 5:
① Firstly, constructing a proper virtual potential field model, initializing all parameters, and setting position information of an unmanned ship departure point, position information of an obstacle and position information of a target point.
② Then judging the information of the obstacle which can influence the unmanned ship, calculating the repulsive force or the repulsive force resultant force according to the repulsive force calculation formula, obtaining the attractive force by the same way, carrying out vector superposition on the two forces to obtain the resultant force born by the unmanned ship, and obtaining the resultant force direction, namely the included angle theta between the resultant force and the x-axis according to trigonometric function knowledge.
③ And then calculating the next position information of the unmanned ship according to the formula (18):
④ Before updating the next position of the unmanned ship each time, it is determined whether the unmanned ship has reached the target point. The judging method comprises the following steps: comparing the step length of the unmanned ship with the distance between the unmanned ship and the target point, if the step length is smaller than or equal to the distance between the unmanned ship and the target point, considering that the unmanned ship does not reach the target point yet, and continuing planning; otherwise, the unmanned ship is considered to reach the target point, and the search is stopped.
In a specific implementation, the application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and the computer program can run the application content of the unmanned ship route planning method based on the particle swarm improvement artificial potential field method and part or all of the steps in each embodiment when being executed by the data processing unit. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random-access memory (random access memory, RAM), or the like.
It will be apparent to those skilled in the art that the technical solutions in the embodiments of the present invention may be implemented by means of a computer program and its corresponding general hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied essentially or in the form of a computer program, i.e. a software product, which may be stored in a storage medium, and include several instructions to cause a device (which may be a personal computer, a server, a single-chip microcomputer MUU, or a network device, etc.) including a data processing unit to perform the methods described in the embodiments or some parts of the embodiments of the present invention.
The invention provides a thought and a method for unmanned ship route planning based on particle swarm improvement artificial potential field method, and the method and the way for realizing the technical scheme are numerous, the above is only a preferred embodiment of the invention, and it should be pointed out that a plurality of improvements and modifications can be made by those skilled in the art without departing from the principle of the invention, and the improvements and modifications are also considered as the protection scope of the invention. The components not explicitly described in this embodiment can be implemented by using the prior art.
Claims (4)
1. The unmanned ship route planning method based on the particle swarm improved artificial potential field method is characterized by comprising the following steps of:
step 1, analyzing an optimization target of unmanned ship route planning, and designing an adaptability function in a particle swarm algorithm;
Step 2, initializing a particle swarm, including: the population scale N, the position x i and the speed v i of each particle, the artificial potential field parameters, the particle group, namely the initial local optimal value and the global optimal value of the population, and the maximum particle group optimizing algebra G and the searching range are set; the artificial potential field parameters include: an attraction gain constant and a repulsion gain constant;
Step3, planning a route by adopting an improved artificial potential field method according to the particle swarm parameters to obtain N routes; calculating an individual fitness value of each particle by adopting the fitness function according to the navigation information of the unmanned ship;
Step 4, updating the particle swarm according to the calculated individual fitness value of each particle; updating the population of particles includes: updating individual extremum of the particles and population extremum of the particle swarm, and updating position and speed information of the particles;
wherein updating the individual extremum of the particles and the population extremum of the particle population comprises: taking the optimal fitness value in the history record of each particle as the current individual extremum of the particle, taking the historical optimal fitness value of all particles as the population extremum of the population, and recording the corresponding position information;
Updating position and velocity information of the particles, including: updating the position and speed information of each particle, and ensuring that the particles do not exceed the search limit according to the limit range;
step 5, checking whether the global optimal value of the particle swarm meets an end condition or whether the iteration number is equal to the maximum particle swarm optimization algebra G, if the condition is met, entering a step 6, otherwise, returning to the step 3;
step 6, outputting the finally optimized artificial potential field parameters, re-planning the route by using the parameters and adopting an improved artificial potential field method, and outputting an optimal route to finish the unmanned ship route planning;
the method adopts a particle swarm algorithm to optimize two parameters, namely an attractive force gain constant and a repulsive force gain constant in an improved artificial potential field method;
the method for initializing the particle swarm in the step 2 comprises the following steps:
regarding each particle in the particle group as a vector, the position of the ith particle And flying speed/>, in solution spaceThe expression is as follows:
wherein m represents the dimension of the target search space, i=1, 2, … … N, N being the particle swarm size;
According to the method, two parameters of an attractive force gain constant k f and a repulsive force gain constant m f in an improved artificial potential field method are optimized by adopting a particle swarm algorithm, a target search space dimension m=2 is set, and:
finishing initialization setting of the particle swarm;
the method for planning the route by adopting the improved artificial potential field method in the step 3 comprises the following steps:
Step 3-1, constructing a virtual potential field model, initializing parameters, and setting position information of a departure point of the unmanned ship, position information of an obstacle and position information of a target point;
Step 3-2, judging the position information of the obstacle influencing the unmanned ship, calculating repulsive force and attractive force according to an improved artificial potential field method, and calculating resultant force and resultant force direction, namely an included angle theta between the resultant force and an x-axis;
Step 3-3, the method for calculating the next position information of the unmanned ship comprises the following steps:
xt+1＝xt+l×cos(θ)
yt+1＝yt+l×sin(θ)
wherein l represents the advancing distance of the unmanned ship, and theta represents the included angle between the resultant force and the x axis;
Step 3-4, judging whether the unmanned ship has reached the target point, and if the unmanned ship has not reached the target point, executing step 3-1; otherwise, ending the route planning;
the updating method of the individual extremum and the group extremum in the step 4 comprises the following steps:
the optimal position experienced by the ith particle itself is noted as:
The individual extremum is p i _best;
The optimal positions experienced by the whole population are noted as:
the population extremum is g_best;
for each particle, comparing the adaptation value with an individual extremum p i _best, and if the adaptation value is better than p i _best, taking the position of the particle at the moment as the current optimal position;
Comparing the adaptation value of each particle with a global extremum g_best, and if the adaptation value is better than the global extremum g_best, taking the position of the particle at the moment as the current group optimal position;
the method for updating the position and speed information of the particles in the step 4 comprises the following steps:
the velocity update for each particle is expressed as:
vi(t+1)＝w×vi(t)+c1×rand()×[pi-best-xi(t)]+c2×rand()×[g_best-xi(t)]
Wherein rand () is a random number between (0, 1), x i (t) is the current position of the particle, w is the inertial weight, c 1 and c 2 are learning factors; the maximum limiting speed v max;vi (t+1) of the particle represents the speed of the particle at time t+1, and v i (t) represents the speed of the particle at time t;
The location update for each particle is expressed as:
xi(t+1)＝xi(t)+vi(t+1)
wherein x i (t+1) represents the position of the particle at the next moment;
the fitness function described in step1 includes:
Constructing the fitness function by adopting the relative length f 1, the relative offset f 2 and the corner coefficient f 3 of the unmanned ship route; wherein the relative length f 1 represents the ratio of the sum of the distances between all the path points on the path to the unmanned ship bollard, the relative offset f 2 represents the ratio of the maximum value of the distances between the connecting lines from the starting point and the end point of all the path points in the repulsive force influence range to the unmanned ship bollard, and the rotation angle coefficient f 3 represents the maximum value of the rotation angles between the adjacent path points on the path;
let num denote unmanned ship path point quantity, L denote unmanned ship motion step length, L denote unmanned ship length, then the calculation formula of relative length f 1 of the course is:
The connection starting point and the target point form a straight line, and the expression of the straight line is set as follows:
Ax+By+C＝0
wherein A, B and C represent parameters of the linear equation, which are determined by coordinates of the starting point and the target point; x and y represent the abscissa and ordinate of the point on the line;
Let (x t,yt) denote the position coordinates of the unmanned ship at time t, the calculation formula of the relative offset f 2 is:
the calculation formula of the rotation angle coefficient f 3 is as follows:
Wherein (x t+1,yt+1) represents the position coordinate of the unmanned ship at the time t+1, and (x t-1,yt-1) represents the position coordinate of the unmanned ship at the time t-1;
The fitness function F is:
Where α is a weight of a relative length, β is a weight of a relative offset, and γ is a weight of a rotation angle coefficient, and satisfies α+β+γ=1.
2. The unmanned ship route planning method based on the particle swarm improvement artificial potential field method according to claim 1, wherein the fitness function in step 1 comprehensively considers the relative length, the relative offset and the rotation angle coefficient as optimization targets, and if only one of the optimization targets is optimized, the weights of the rest of the optimization targets are set to zero.
3. The unmanned ship routing method based on the particle swarm improved artificial potential field method according to claim 2, wherein the method for calculating the attraction force according to the improved artificial potential field method in step 3-2 comprises the steps of:
assuming that the position of the unmanned ship is denoted as q= (x, y) T and the position of the target point is denoted as q goal＝(xgoal,ygoal)T, the method of calculating the attractive force is as follows:
The force potential function is expressed as:
Wherein U att (q) represents the gravitational potential field, k f represents the gravitational gain constant, and ρ (q, q goal) represents the distance between the unmanned boat and the target point; the attraction force of the unmanned ship is a negative gradient of the attraction function, and the direction of the attraction force is pointed to a target point by the unmanned ship, which is expressed as:
Wherein F att (q) represents the attraction force applied by the unmanned ship, and gard () represents the gradient calculation symbol.
4. A method of unmanned ship routing based on the particle swarm improved artificial potential field method according to claim 3, wherein the method of calculating repulsive force according to the improved artificial potential field method in step 3-2 comprises:
Wherein U rep (q) represents a repulsive force potential field, m f represents a repulsive force gain constant, q 0 represents the position of an obstacle, q represents the real-time position of the unmanned ship, q goal represents the position of a target point, ρ (q, q 0) represents the distance between the unmanned ship and the obstacle, and ρ 0 represents the influence range of the repulsive force generated by the obstacle; wherein, unmanned ship and the distance of target point represent:
after the repulsive force potential field function is improved, the repulsive force and the attractive force are no longer acted on the same straight line, and the repulsive force is decomposed and expressed as follows:
Wherein F rep1 (q) represents the component of the repulsive force applied by the unmanned ship along the direction of the attractive force, and the direction is from the obstacle to the unmanned ship; f rep2 (q) represents another component of the repulsive force experienced by the unmanned aerial vehicle, the direction being directed by the unmanned aerial vehicle to the target point;
the decomposition method of the two components is as follows:
where, |f rep1 | and |f rep2 | represent the modes of the two repulsive force components, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210988311.8A CN115420289B (en) | 2022-08-17 | 2022-08-17 | Unmanned ship route planning method based on particle swarm improvement artificial potential field method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210988311.8A CN115420289B (en) | 2022-08-17 | 2022-08-17 | Unmanned ship route planning method based on particle swarm improvement artificial potential field method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115420289A CN115420289A (en) | 2022-12-02 |
| CN115420289B true CN115420289B (en) | 2024-05-21 |
Family
ID=84198996
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210988311.8A Active CN115420289B (en) | 2022-08-17 | 2022-08-17 | Unmanned ship route planning method based on particle swarm improvement artificial potential field method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115420289B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117193348B (en) * | 2023-02-03 | 2024-05-31 | 河海大学 | Intelligent inspection method for dam reservoir area of bionic unmanned aerial vehicle group |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106705975A (en) * | 2017-01-11 | 2017-05-24 | 江苏科技大学 | Method for planning route of unmanned surface vehicle based on universal gravitation search |
| CN109933067A (en) * | 2019-03-12 | 2019-06-25 | 哈尔滨工程大学 | A kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm |
| CN112099501A (en) * | 2020-09-15 | 2020-12-18 | 哈尔滨工程大学 | Unmanned ship path planning method based on potential field parameter optimization |
| CN114545931A (en) * | 2022-01-27 | 2022-05-27 | 大连海事大学 | Water surface unmanned ship path planning method based on Bi-RRT algorithm guided by improved artificial potential field method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210287556A1 (en) * | 2020-03-13 | 2021-09-16 | Electronics And Telecommunications Research Institute | Method and apparatus for generating optimal path for an unmanned aerial vehicle |
-
2022
- 2022-08-17 CN CN202210988311.8A patent/CN115420289B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106705975A (en) * | 2017-01-11 | 2017-05-24 | 江苏科技大学 | Method for planning route of unmanned surface vehicle based on universal gravitation search |
| CN109933067A (en) * | 2019-03-12 | 2019-06-25 | 哈尔滨工程大学 | A kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm |
| CN112099501A (en) * | 2020-09-15 | 2020-12-18 | 哈尔滨工程大学 | Unmanned ship path planning method based on potential field parameter optimization |
| CN114545931A (en) * | 2022-01-27 | 2022-05-27 | 大连海事大学 | Water surface unmanned ship path planning method based on Bi-RRT algorithm guided by improved artificial potential field method |
Non-Patent Citations (2)
| Title |
|---|
| An improved artificial potential field algorithm with swerving force for USV path planning;Xiaowei Li et al.;《IEEE》;20211231;第1019-1024页 * |
| 基于量子粒子群优化算法的无人艇航线规划;金建海 等;《船舶力学》;20200331;第24卷(第3期);第352-361页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115420289A (en) | 2022-12-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108803321B (en) | Autonomous underwater vehicle track tracking control method based on deep reinforcement learning | |
| CN111781948A (en) | Unmanned aerial vehicle formation shape transformation and online dynamic obstacle avoidance method | |
| CN111399541B (en) | Unmanned aerial vehicle whole-region reconnaissance path planning method of unsupervised learning type neural network | |
| CN109871031B (en) | Trajectory planning method for fixed-wing unmanned aerial vehicle | |
| Yu et al. | Analytical entry guidance for coordinated flight with multiple no-fly-zone constraints | |
| CN113848974B (en) | Aircraft trajectory planning method and system based on deep reinforcement learning | |
| Gramajo et al. | An efficient energy constraint based UAV path planning for search and coverage | |
| CN115420289B (en) | Unmanned ship route planning method based on particle swarm improvement artificial potential field method | |
| CN111930141A (en) | Three-dimensional path visual tracking method for underwater robot | |
| Mousavian et al. | Identification-based robust motion control of an AUV: optimized by particle swarm optimization algorithm | |
| CN113093733B (en) | Sea-to-sea striking method for unmanned boat cluster | |
| CN115454115A (en) | Rotor unmanned aerial vehicle path planning method based on hybrid wolf-particle swarm algorithm | |
| CN109870906A (en) | A kind of high-speed rotor aircraft paths planning method based on BBO optimization Artificial Potential Field | |
| CN115903888A (en) | Rotor unmanned aerial vehicle autonomous path planning method based on longicorn swarm algorithm | |
| Wu et al. | Multi-phase trajectory optimization for an aerial-aquatic vehicle considering the influence of navigation error | |
| Yin et al. | A hybrid path planning algorithm based on simulated annealing particle swarm for the self-driving car | |
| CN117724524A (en) | Unmanned aerial vehicle route planning method based on improved spherical vector particle swarm algorithm | |
| CN114943168B (en) | Method and system for combining floating bridges on water | |
| CN114609925B (en) | Training method of underwater exploration strategy model and underwater exploration method of bionic machine fish | |
| CN112161626B (en) | High-flyability route planning method based on route tracking mapping network | |
| Jose et al. | Navigating the Ocean with DRL: Path following for marine vessels | |
| Hou et al. | AUV path planning with kinematic constraints in unknown environment using reinforcement learning | |
| Kadir et al. | Dynamic Path Planning of Unmanned Surface Vehicle Based On Genetic Algorithm With Sliding Curve Guidance System | |
| CN114020010B (en) | Unmanned aerial vehicle cluster space-based recovery method, unmanned aerial vehicle cluster space-based recovery system, electronic equipment and medium | |
| Guan et al. | Motorized buoy path following based on improved LOS algorithm and Aquila Optimizer algorithm |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |