CN106845716B - Navigation error constraint-based water surface unmanned ship local hierarchical path planning method - Google Patents
Navigation error constraint-based water surface unmanned ship local hierarchical path planning method Download PDFInfo
- Publication number
- CN106845716B CN106845716B CN201710055413.3A CN201710055413A CN106845716B CN 106845716 B CN106845716 B CN 106845716B CN 201710055413 A CN201710055413 A CN 201710055413A CN 106845716 B CN106845716 B CN 106845716B
- Authority
- CN
- China
- Prior art keywords
- point
- gene
- chromosome
- unmanned surface
- navigation
- 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
- 238000000034 method Methods 0.000 title claims abstract description 65
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 230000003068 static effect Effects 0.000 claims abstract description 21
- 210000000349 chromosome Anatomy 0.000 claims description 81
- 108090000623 proteins and genes Proteins 0.000 claims description 72
- 101150076211 TH gene Proteins 0.000 claims description 25
- 230000008859 change Effects 0.000 claims description 19
- 230000035772 mutation Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 13
- 101150084750 1 gene Proteins 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 7
- QZXCCPZJCKEPSA-UHFFFAOYSA-N chlorfenac Chemical compound OC(=O)CC1=C(Cl)C=CC(Cl)=C1Cl QZXCCPZJCKEPSA-UHFFFAOYSA-N 0.000 claims description 6
- LFULEKSKNZEWOE-UHFFFAOYSA-N propanil Chemical compound CCC(=O)NC1=CC=C(Cl)C(Cl)=C1 LFULEKSKNZEWOE-UHFFFAOYSA-N 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000012614 Monte-Carlo sampling Methods 0.000 claims description 3
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000006978 adaptation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 2
- 230000000977 initiatory effect Effects 0.000 claims 1
- 230000002068 genetic effect Effects 0.000 abstract description 8
- 230000002411 adverse Effects 0.000 abstract description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 230000007613 environmental effect Effects 0.000 description 4
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 235000001968 nicotinic acid Nutrition 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Operations Research (AREA)
- Biomedical Technology (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a navigation error constraint-based method for planning local layered paths of unmanned surface vehicles, which mainly comprises the following steps: local static path planning based on navigation error constraint and combined with genetic algorithm; and local dynamic path planning fusing marine rules. The invention adopts a layering thought to divide the local path planning into two layers of static path planning and dynamic path planning, thereby solving the problem of avoiding static obstacles and the problem of avoiding dynamic obstacles; the static path planning based on the navigation error constraint and combined with the genetic algorithm reduces the adverse effect of the navigation error on the path selection and improves the safety of the planned path; dynamic path planning of the maritime affair rules is fused, self dynamic constraints of the unmanned surface vehicle are considered, and feasibility of path planning is improved.
Description
Technical Field
The invention relates to a navigation error constraint-based water surface unmanned ship local hierarchical path planning method, and belongs to the technical field of water surface unmanned ship path planning.
Background
With the rapid development of science and technology, the marine intelligent transportation is an important component of the science and technology strategy of China, and is mainly used for realizing the automation of ship navigation and the intellectualization of marine traffic management. Therefore, the surface unmanned ship is required to be used as an individual ship of the marine traffic system, and the research on the sailing path is necessary. The path planning is an important research field of the unmanned surface vehicle and is a premise that the unmanned surface vehicle can safely complete a cruise task. In recent years, unmanned surface vehicle path planning has become a research hotspot for offshore intellectualization.
The path planning methods are various, are suitable for various different situations due to advantages and disadvantages of the path planning methods, and can be roughly divided into a traditional algorithm, a graph method, an intelligent bionics algorithm and other algorithms. The traditional algorithm comprises an artificial potential field method, a fuzzy logic algorithm, a simulated annealing algorithm and the like; the graphic method includes a visual diagram space method, a grid method, a voronoi diagram method and the like; the intelligent bionics algorithm comprises an ant colony algorithm, a neural network algorithm, a genetic algorithm, a particle swarm algorithm and the like; other algorithms are the a-algorithm, Dijkstra algorithm, Floyd algorithm, etc.
At present, most path plans are only divided into two layers of global path plans and local path plans, the layering problem of the local path plans is not considered, and the method has certain limitations. In addition, most of static path planning assumes that the unmanned surface vehicle can accurately track a planned path on the premise of accurate navigation information. However, an error exists in an actual navigation system, which will cause a safe area between the unmanned surface vehicle and an actual obstacle to decrease, and cause a free path from an initial point to a target point to be absent. Even if a free path still exists, it is likely to give the surface unmanned boat unpredictable risks such as collision. Therefore, in the path planning process, the influence of the navigation error is a non-negligible problem, and the method has important significance for the practical engineering application of the unmanned surface vehicle.
At present, there are also some research reports related to the present invention, 1, research on a layered danger avoidance method for an unmanned surface vehicle under complex sea conditions, the university of harbin engineering, 2014, in which although three layered layers are considered, influence of a navigation error on path planning is not considered. 2. A navigation error space-based unmanned underwater vehicle path planning method comprises war institute reports 2014, 35(8):1243-1250, wherein navigation errors are considered in the document, but global path planning research is only carried out on an underwater vehicle, and a genetic algorithm is not combined.
Disclosure of Invention
The technical problem is as follows: the invention provides a navigation error constraint-based water surface unmanned ship local hierarchical path planning method for reducing adverse effects of navigation errors on path selection.
The technical scheme is as follows: the invention discloses a navigation error constraint-based method for planning local hierarchical paths of unmanned surface vehicles, which comprises the following steps:
step 1) discretizing an environment into two-dimensional basic unit grids by adopting a rectangular coordinate system and a grid method according to static obstacle information obtained by an operation task, an electronic chart and a sensor, and initializing a starting point position Start and a target point position Goal by carrying out environment modeling on marks of the grids;
step 2) obtaining M chromosomes according to the following modes: randomly selecting a free grid points on the grid map except the Start point position Start and the target point position Goal by adopting a random method, dividing the Start point and the target point into a +1 sections, and then connecting the Start point, the target point and the a free grid points by adopting an A-star algorithm to form a feasible initial solution, namely obtaining an effective chromosome;
taking the M chromosomes as an initial population, namely a first generation population, wherein M is the scale of the first generation population;
step 3) determining the navigation error of each gene point of each chromosome, which specifically comprises the following steps: using the desired navigation position mu at the kth gene point in each chromosomekAnd the expected navigation position mu at the k +1 gene pointk+1D (μ) of the distance between the twok,μk+1) And rate of change of navigation error ΔPerforming a multiplication operation, and adding the navigation error of the kth gene pointkTo obtain the navigation error of the k +1 gene point of each chromosomek+1(ii) a Then according to the error of Beidou navigationBDIf the navigation error of the (k +1) th gene pointk+1>BDThen will beBDAssignment of valuek+1(ii) a Otherwise, the navigation error of the (k +1) th gene pointk+1Is the original value;
step 4) obtaining the navigation error of each gene point of each chromosome according to the step 3), adopting a Gaussian model as a navigation error model, and distributing density functions according to the navigation positions of the navigation error modelCalculating the path cost L(s) of each chromosome by adopting a Monte Carlo sampling method0,sg) Wherein the minimum value of all chromosome path costs is the optimal solution of the generation population;
step 5) taking the dangerous area with the obstacle as a collision area A, integrating the navigation error probability density function in the collision area A according to the navigation error model, solving the collision probability of each gene point of each chromosome, and obtaining a set of collision probabilities { P { of all gene points from the starting point to the target point of each chromosome, namely from the 0 th gene point to the g th gene pointc(X0),Pc(X1)…Pc(Xg)};
Step 6) according to the collision probability set of all gene points of each chromosome obtained in the step 5), aiming at each chromosome, searching the collision probability set { P } of the chromosomec(X0),Pc(X1)…Pc(Xg) The maximum value of the collision probability in (1), which is then taken as the collision probability P of the chromosomec(X);
Step 7) according to the path cost of each chromosome obtained in the step 4) and the collision probability P of each chromosome obtained in the step 6)c(X) calculating fitness of each chromosome in the following manner;
judging the collision probability P of each chromosome obtained in the step 6)c(X) is less than or equal to the set value, and if so, calculating the fitness E of each chromosome according to the following formula:
otherwise the fitness E of each chromosome is calculated according to the following formula:
wherein m is a weight value set value>>L(s0,sg);
Adding fitness of all chromosomes to calculate the average value as the average fitness value of the generation population;
step 8) assuming that the nth generation of the population generation is x (n), and if the number of the population generation is x (n)<Set value nsettingEntering step 9); otherwise, processing is as follows:
if at the n- (n) thsetting-1) successive n generations of population to current nth generation populationsettingIn the generation evolution, the optimal solution of the population is not changed, and the variation of the population average adaptation value of two adjacent generations does not exceed 1%, then the chromosome with the optimal solution in the current nth generation is taken as an optimal collision-free path to be solved, a series of path node sequences from a starting point to a target point are obtained according to the coordinate values of gene points forming the chromosome, and the step 10) is carried out after the path node sequences are taken as local route point sequences, otherwise, the step 9) is carried out;
step 9) according to the crossing rate pcI.e. probability values of set crossover operations, randomly selecting pcPerforming crossover operation on M chromosomes, randomly exchanging gene points of crossed individuals to form new chromosomes, and replacing p with maximum fitnesscM chromosomes, and then the variation rate pmI.e. probability values of the mutation operations set, randomly selecting pmPerforming mutation operation on M chromosomes, performing real value mutation on randomly selected gene points of the individuals subjected to mutation, and replacing p with maximum fitnessmM chromosomes; taking the population obtained by the cross mutation operation as a new population, namely, the next generation (n +1 generation) population generation is x (n +1) ═ x (n) +1, and returning to the step 3);
step 10) according to the series of local route sequence points obtained in the step 8), taking a first local route point as a dynamic planning starting point and taking a second local route point as a dynamic planning target point;
step 11) obtaining dynamic obstacle position information and speed information in the sailing process of the water surface unmanned ship according to a sensor, solving the nearest meeting distance DCPA between the water surface unmanned ship and an obstacle and the time TCPA of reaching a nearest meeting point, and then weighting the DCPA and the TCPA according to the following formula to obtain the collision risk rho:
ρ=(ωDDCPA)2+(ωTTCPA)2
wherein, ω isD、ωTRespectively are weighted values;
step 12) judging whether the collision risk rho obtained in the step 11) is smaller than or equal to an empirical value, if so, adjusting the course angle or the speed of the unmanned surface vehicle according to an obstacle avoidance strategy fused with maritime regulations and dynamic constraint conditions of the unmanned surface vehicle, otherwise, keeping the original course angle and speed of the unmanned surface vehicle unchanged;
step 13) according to the change amount △α of the heading angle and the change amount △ V of the speed obtained in the step 12)USVAnd the set planning time △ T, and the current position (X, Y) of the unmanned surface vehicle after planning a △ T time is calculated as:
in the formula, X, Y represents the abscissa and ordinate of the unmanned surface vehicle in a rectangular coordinate system after planning a △ T time, Xprev、YprevRespectively carrying out α V and V on the horizontal coordinate value and the vertical coordinate value of the unmanned surface vehicle in the rectangular coordinate system before planningUSVRespectively representing the course angle and the speed of the unmanned surface vehicle.
Judging whether the current position of the unmanned surface vehicle is a current dynamic planning target point, if so, entering a step 14), and if not, returning to the step 11);
and 14) judging whether the current position of the unmanned surface vehicle is the target point position Goal or not according to the target point position Goal obtained in the step 1), if so, ending the process of the method, otherwise, taking the current dynamic planning target point as a next dynamic planning starting point, taking a next local route point as a next dynamic planning target point, and returning to the step 11).
Further, in the method of the present invention, the specific solving process for solving the path cost of each chromosome in the step 4) in combination with the navigation error model includes:
the navigation error model is:
Xk~N(μk,Σk)
in the formula,desired navigation position, X, for unmanned surface vehiclek=(xk,yk) For the navigation position of the unmanned surface vehicle at the kth gene point, sigmakCovariance matrix of kth gene point.
The navigation position distribution density function of the navigation error modelComprises the following steps:
the covariance matrix ΣkDetermined according to the following formula:
in the formula,kthe navigation error of the kth gene point; sigmakStandard deviation for the kth gene point;
the path cost L(s) of each chromosome0,sg) The specific calculation process comprises the following steps:
state s at the kth Gene Point according to the 3 σ criterion in the probabilistic modelkIn the navigation uncertainty region of (a), i.e. with the desired navigation position mu at the kth gene sitekIn the 3 sigma confidence region with the center as the center, Monte Carlo is adoptedN sample points are sampled by the method, and the state s of the kth gene point is obtainedkOf navigation uncertainty region of (a) a set of samples Θk:
State s at the k +1 th Gene sitek+1The navigation uncertain region is sampled by N sample points to obtain the state s of the (k +1) th gene pointk+1Of navigation uncertainty region of (a) a set of samples Θk+1:
According to the sampling set theta of the kth gene pointkSampling set theta with k +1 gene pointk+1Calculating Euclidean distances between N pairs of sampling points from the kth gene point to the (k +1) th gene point, summing the Euclidean distances between the N pairs of sampling points, and averaging to obtain the transfer cost L(s) from the kth gene point to the (k +1) th gene pointk,sk+1):
The state s at the starting Gene site was calculated according to the following formula0Status to target Gene site sgPer chromosome path cost L(s)0,sg):
Further, in the method of the present invention, the obstacle avoidance policy of the fusion maritime affair rule in step 12) is:
in the formula: e.g. of the typei+fi1, wherein eiAnd fiIs twoTaking 0 or 1 as a binary variable, wherein ξ is a positive real number and is infinite, and when the unmanned surface vehicle meets the right side of the obstacle in a crossing way or in a front way, avoiding the obstacle by the right side, and at the moment, ei=1,fiWhen the unmanned surface vehicle and the left side of the obstacle meet or trace the plane, the obstacle is avoided by the left side, and at the moment ei=0,fi1 and △ V is the velocity vector V of the unmanned surface vehicle relative to the obstacleUOSize;the minimum value of the direction change quantity of the relative speed △ V is determined on the premise that the unmanned surface vehicle can escape from the collision region when the unmanned surface vehicle turns left;when the unmanned surface vehicle turns to the right, the maximum value of the direction change quantity of the relative speed △ V is obtained on the premise that the unmanned surface vehicle can escape from the collision area;is a velocity vector V of the unmanned surface vehicle relative to the barrierUOVelocity vector V of unmanned surface vehicleUSVAngle of (△ V)USV△α is the change of the course angle of the unmanned surface vehicle;
in the step 12), the course angle and the speed adjustment of the unmanned surface vehicle are obtained by adopting a mixed integer linear programming method according to the collision avoidance strategy and the dynamic constraint of the unmanned surface vehicle.
Further, in the method of the present invention, in the step 11), the starboard target ω is determinedD=5、ωT0.5 for port target ωD=5、ωT=1。
Further, in the method of the present invention, in the step 12), the dynamic constraint conditions of the unmanned surface vehicle include maximum speed, maximum acceleration, and minimum turning radius. In the method, the path planning is mainly used for realizing obstacle avoidance, and the obstacle can be divided into a dynamic form and a static form, wherein the dynamic form and the static form are used as the former (… bureau)Partial static path planning) primarily avoid static obstacles, the latter (… partial dynamic path planning) primarily avoid dynamic obstacles. However, the two are not independent, a series of local waypoints are obtained after the static planning of the former and can be used as a starting point and a target point of the dynamic planning of the latter. Suppose that static planning obtains a series of location points S including a Start point Start0And a local waypoint sequence S of the position points Sn of the target point Goal0、S1、S2… Sn, when performing dynamic planning, first use S0As a starting point for dynamic planning, S1As a dynamic planning target point, after the dynamic path planning is carried out to reach the dynamic planning target point, S is added1As a starting point for dynamic planning, S2And as a dynamic planning target point, performing dynamic planning, and repeating continuously until the target point Sn is reached.
Has the advantages that: compared with the prior art, the invention has the following advantages:
1) compared with the prior art, the invention adopts the thought of partial layers, the partial path planning is divided into two layers of static path planning and dynamic path planning, the static partial path planning takes the shortest path as an optimization target, the shortest collision-free path can be obtained, and the dynamic partial path planning takes the motion area escaping from the barrier as soon as possible as the optimization target, so that the unmanned surface vehicle is far away from the motion barrier to ensure the safety of real-time navigation. In addition, in terms of algorithm implementation, static obstacle information obtained by the electronic chart and the sensor is transmitted to the local static path planning, repeated planning on the static obstacles is not needed in real time in the dynamic local planning, and the path planning time and the memory occupation are reduced.
2) Compared with the prior art, the invention adopts an improved genetic path planning method based on navigation error constraint, converts non-accurate navigation information into unknown environmental information in an environmental model, fuses the unknown environmental information into path planning, does not take accurate navigation information as an assumption premise, fully considers the influence of navigation error in the environmental model, reduces the adverse effect of the navigation error on path selection, and improves the safety of the planned path. Compared with traditional search methods such as enumeration, heuristic method, etc., the improved genetic algorithm has good global search capability, can search out all collision-free paths meeting constraint conditions from the path planning space rapidly, and can not get into local optimal solution; the search starts from a population formed by all feasible paths, has potential parallelism, can conveniently perform distributed computation, and accelerates the solving speed; when the input scale is large, the optimal or approximately optimal solution can be solved effectively, which is different from the situation of no solution in the traditional method.
Under the VC simulation condition, the invention carries out simulation experiment on the local static path planning based on the genetic algorithm by considering the influence of the navigation error:
setting genetic algorithm parameters: group size M50, crossover rate pc0.8, rate of variation pm=0.1;
Initial value of navigation error:0=0.1m;
△ navigation error change rate=0.05;
The results obtained by the method of the invention are respectively shown in FIG. 3. The result shows that under the condition of considering the influence of navigation errors, the method can select to bypass to a free area without selecting a narrow area in the path planning process, and the navigation safety and feasibility of the unmanned surface ship are ensured at the cost of sacrificing the path length.
Drawings
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a schematic diagram of maritime rule conflict scenario definitions and avoidance measures to be taken in different meeting scenarios according to the present invention;
FIG. 3 is a schematic diagram of a navigation error based surface unmanned vehicle path planning of the present invention;
FIG. 4 is a comparison graph of the local static path planning of the unmanned surface vehicle, with or without the influence of navigation errors.
Fig. 5 is a block diagram of the present invention.
Detailed Description
The invention is further described with reference to the following examples and the accompanying drawings.
The invention discloses a navigation error constraint-based method for planning local hierarchical paths of unmanned surface vehicles, which comprises the following steps:
step 1) discretizing an environment into two-dimensional basic unit grids by adopting a rectangular coordinate system and a grid method according to static obstacle information obtained by an operation task, an electronic chart and a sensor, and initializing a starting point position Start and a target point position Goal by carrying out environment modeling on marks of the grids; the area with obstacles is marked as 1, namely, an obstacle grid, and the area without obstacles is marked as 0, namely, a free grid; the rectangular coordinate system takes the initial point as the origin, and x, y and z point to east, north and sky of the initial point respectively;
step 2) obtaining M chromosomes according to the following modes: randomly selecting a free grid points on the grid map except the Start point position Start and the target point position Goal by adopting a random method, dividing the Start point and the target point into a +1 sections, and then connecting the Start point, the target point and the a free grid points by adopting an A-star algorithm to form a feasible initial solution, namely obtaining an effective chromosome;
taking the M chromosomes as an initial population, namely a first generation population, wherein M is the scale of the first generation population;
step 3) determining the navigation error of each gene point of each chromosome, which specifically comprises the following steps: using the desired navigation position mu at the kth gene point in each chromosomekAnd the expected navigation position mu at the k +1 gene pointk+1D (μ) of the distance between the twok,μk+1) And rate of change of navigation error ΔPerforming a multiplication operation, and adding the navigation error of the kth gene pointkTo obtain the navigation error of the k +1 gene point of each chromosomek+1:
f:k+1=k+△D(μk,μk+1)
Then according to the error of Beidou navigationBDIf the navigation error of the (k +1) th gene pointk+1>BDThen will beBDAssignment of valuek+1(ii) a Otherwise, the navigation error of the (k +1) th gene pointk+1Is the original value;
step 4) obtaining the navigation error of each gene point of each chromosome according to the step 3), adopting a Gaussian model as a navigation error model, and distributing density functions according to the navigation positions of the navigation error modelCalculating the path cost L(s) of each chromosome by adopting a Monte Carlo sampling method0,sg) Wherein the minimum value of all chromosome path costs is the optimal solution of the generation population;
4.1) the navigation error model is:
Xk~N(μk,Σk)
in the formula,desired navigation position, X, for unmanned surface vehiclek=(xk,yk) For the navigation position of the unmanned surface vehicle at the kth gene point, sigmakCovariance matrix of kth gene point.
The navigation position distribution density function of the navigation error modelComprises the following steps:
the covariance matrix ΣkDetermined according to the following formula:
in the formula,kthe navigation error of the kth gene point; sigmakStandard deviation for the kth gene point;
4.2) Path cost per chromosome L(s)0,sg) The specific calculation process comprises the following steps:
state s at the kth Gene Point according to the 3 σ criterion in the probabilistic modelkIn the navigation uncertainty region of (a), i.e. with the desired navigation position mu at the kth gene sitekIn a 3 sigma confidence region which is the circle center, adopting a Monte Carlo method to sample N sample points and obtaining the state s of the kth gene pointkOf navigation uncertainty region of (a) a set of samples Θk:
State s at the k +1 th Gene sitek+1The navigation uncertain region is sampled by N sample points to obtain the state s of the (k +1) th gene pointk+1Of navigation uncertainty region of (a) a set of samples Θk+1:
According to the sampling set theta of the kth gene pointkSampling set theta with k +1 gene pointk+1Calculating Euclidean distances between N pairs of sampling points from the kth gene point to the (k +1) th gene point, summing the Euclidean distances between the N pairs of sampling points, and averaging to obtain the transfer cost L(s) from the kth gene point to the (k +1) th gene pointk,sk+1):
The state s at the starting Gene site was calculated according to the following formula0Status to target Gene site sgPer chromosome path cost L(s)0,sg):
And 5) taking the dangerous area with the obstacle as a collision area A, integrating the navigation error probability density function in the collision area A according to the navigation error model, and solving the collision probability of each gene point of each chromosome:
in the formula, Pc(Xk) The collision probability at the kth gene point of each chromosome;
the set of collision probabilities { P) at all gene points from the 0 th gene point (i.e., the origin) to the g th gene point (i.e., the target point) of each chromosome is obtained according to the above formulac(X0),Pc(X1)…Pc(Xg)};
Step 6) according to the collision probability set of all gene points of each chromosome obtained in the step 5), aiming at each chromosome, searching the collision probability set { P } of the chromosomec(X0),Pc(X1)…Pc(Xg) The maximum value of the collision probability in (1), which is then taken as the collision probability P of the chromosomec(X):
Pc(X)=max{Pc(X0),Pc(X1)...Pc(Xg)}
Step 7) according to the path cost of each chromosome obtained in the step 4) and the collision probability P of each chromosome obtained in the step 6)c(X) calculating fitness of each chromosome in the following manner;
judging the collision probability P of each chromosome obtained in the step 6)c(X) is less than or equal to the set value, and if so, calculating the fitness E of each chromosome according to the following formula:
otherwise the fitness E of each chromosome is calculated according to the following formula:
wherein m is a weight value set value>>L(s0,sg);
Adding fitness of all chromosomes to calculate the average value as the average fitness value of the generation population;
step 8) assuming that the nth generation of the population generation is x (n), and if the number of the population generation is x (n)<Set value nsettingEntering step 9); otherwise, processing is as follows:
if at the n- (n) thsetting-1) successive n generations of population to current nth generation populationsettingIn the generation evolution, the optimal solution of the population is not changed, and the variation of the population average adaptation value of two adjacent generations does not exceed 1%, then the chromosome with the optimal solution in the current nth generation is taken as an optimal collision-free path to be solved, a series of path node sequences from a starting point to a target point are obtained according to the coordinate values of gene points forming the chromosome, and the step 10) is carried out after the path node sequences are taken as local route point sequences, otherwise, the step 9) is carried out;
step 9) according to the crossing rate pcI.e. probability values of set crossover operations, randomly selecting pcPerforming crossover operation on M chromosomes, randomly exchanging gene points of crossed individuals to form new chromosomes, and replacing p with maximum fitnesscM chromosomes, and then the variation rate pmI.e. probability values of the mutation operations set, randomly selecting pmPerforming mutation operation on M chromosomes, performing real value mutation on randomly selected gene points of the individuals subjected to mutation, and replacing p with maximum fitnessmM chromosomes; taking the population obtained by the cross mutation operation as a new population, namely, the next generation (n +1 generation) population generation is x (n +1) ═ x (n) +1, and returning to the step 3);
step 10) according to the series of local route sequence points obtained in the step 8), taking a first local route point as a dynamic planning starting point and taking a second local route point as a dynamic planning target point;
step 11) obtaining dynamic obstacle position information and speed information in the sailing process of the water surface unmanned ship according to a sensor, solving the nearest meeting distance DCPA between the water surface unmanned ship and an obstacle and the time TCPA of reaching a nearest meeting point, and then weighting the DCPA and the TCPA according to the following formula to obtain the collision risk rho:
ρ=(ωDDCPA)2+(ωTTCPA)2
wherein, ω isD、ωTRespectively are weighted values; in general, for starboard targets ωD=5、ωT0.5 for port target ωD=5、ωT=1;
Step 12) judging whether the collision risk rho obtained in the step 11) is smaller than or equal to an empirical value, if so, adjusting the course angle or the speed of the unmanned surface vehicle according to an obstacle avoidance strategy fused with maritime regulations and dynamic constraint conditions of the unmanned surface vehicle, otherwise, keeping the original course angle and speed of the unmanned surface vehicle unchanged;
the specific process of adjusting the course angle or the speed of the unmanned surface vehicle according to the obstacle avoidance strategy fused with the maritime regulations comprises the following steps: step 12.1) taking the navigation direction of the unmanned surface vehicle as a reference, solving the course difference delta theta between the unmanned surface vehicle and the obstacle, and then judging the type of the meeting situation according to the conflict situation definition of the international maritime regulations:
if the delta theta belongs to [0 ], 45 DEG ] and [315 DEG ], 360 DEG ], the situation is overtaking;
if delta theta is belonged to [165 degrees and 195 degrees ], the situation is a positive meeting situation;
if delta theta belongs to (45 degrees and 165 degrees), the right side intersection meeting situation is obtained;
if delta theta is belonged to (195 degrees and 315 degrees), the left side intersection meeting situation is obtained;
step 12.2) according to the position vector X of the unmanned surface vehicle at the current water surfaceUSVAnd velocity vector VUSVPosition vector X of obstacleObsAnd velocity vector VObsAnd combining a polar coordinate system, taking the north direction as the polar axis direction, taking the anticlockwise direction as the positive direction, and obtaining the collision avoidance model of the unmanned surface boat by a speed obstacle avoidance method as follows:
wherein △ V is the velocity vector V of the unmanned surface vehicle relative to the obstacleUOSize;the minimum value of the direction change quantity of the relative speed △ V is determined on the premise that the unmanned surface vehicle can escape from the collision region when the unmanned surface vehicle turns left;when the unmanned surface vehicle turns to the right, the maximum value of the direction change quantity of the relative speed △ V is obtained on the premise that the unmanned surface vehicle can escape from the collision area;is a velocity vector V of the unmanned surface vehicle relative to the barrierUOVelocity vector V of unmanned surface vehicleUSVAngle of (△ V)USV△α is the change of the course angle of the unmanned surface vehicle;
step 12.3) obtaining an anti-collision strategy according to the meeting situation type obtained in the step 12.1) and the obstacle avoidance model obtained in the step 12.3):
in the formula: e.g. of the typei+fi1, wherein eiAnd fiTaking 0 or 1 as binary variable, ξ as positive real number and infinity, and avoiding the obstacle by the right side when the unmanned surface vehicle meets the right side of the obstacle in a crossing way or in a front meeting way, wherein e is the momenti=1,fiWhen the unmanned surface vehicle and the left side of the obstacle meet or trace the plane, the obstacle is avoided by the left side, and at the moment ei=0,fi=1;
And 12.4) obtaining the course angle and the speed adjustment quantity of the unmanned surface vehicle by adopting a mixed integer linear programming method according to the collision avoidance strategy obtained in the step 12.3) and the dynamic constraints of the unmanned surface vehicle, such as the maximum speed, the maximum acceleration, the minimum turning radius and the like.
Step 13) automatically setting a planning time △ T according to the performance and the actual demand of the unmanned surface vehicle, and obtaining a course angle change △α and a speed change △ V according to the step 12)USVAnd the set planning time △ T, and the current position (X, Y) of the unmanned surface vehicle after planning a △ T time is calculated as:
in the formula, X, Y represents the abscissa and ordinate of the unmanned surface vehicle in a rectangular coordinate system after planning a △ T time, Xprev、YprevRespectively carrying out α V and V on the horizontal coordinate value and the vertical coordinate value of the unmanned surface vehicle in the rectangular coordinate system before planningUSVRespectively representing the course angle and the speed of the unmanned surface vehicle.
Judging whether the current position of the unmanned surface vehicle is a current dynamic planning target point, if so, entering a step 14), and if not, returning to the step 11);
and 14) judging whether the current position of the unmanned surface vehicle is the target point position Goal or not according to the target point position Goal obtained in the step 1), if so, ending the process of the method, otherwise, taking the current dynamic planning target point as a next dynamic planning starting point, taking a next local route point as a next dynamic planning target point, and returning to the step 11).
The above examples are only preferred embodiments of the present invention, it should be noted that: it will be apparent to those skilled in the art that various modifications and equivalents can be made without departing from the spirit of the invention, and it is intended that all such modifications and equivalents fall within the scope of the invention as defined in the claims.
Claims (5)
1. A navigation error constraint-based method for planning local hierarchical paths of unmanned surface vehicles is characterized by comprising the following steps:
step 1) discretizing an environment into two-dimensional basic unit grids by adopting a rectangular coordinate system and a grid method according to static obstacle information obtained by an operation task, an electronic chart and a sensor, and initializing a starting point position Start and a target point position Goal by carrying out environment modeling on marks of the grids;
step 2) repeating the steps for M times according to the following modes to obtain M chromosomes: randomly selecting a free grid points on the grid map except the Start point position Start and the target point position Goal by adopting a random method, dividing the Start point and the target point into a +1 sections, and then connecting the Start point, the target point and the a free grid points by adopting an A-star algorithm to form a feasible initial solution, namely obtaining an effective chromosome;
taking the M chromosomes as an initial population, namely a first generation population, wherein M is the scale of the first generation population;
step 3) determining the navigation error of each gene point of each chromosome, which specifically comprises the following steps: using the desired navigation position mu at the kth gene point in each chromosomekAnd the expected navigation position mu at the k +1 gene pointk+1D (μ) of the distance between the twok,μk+1) And rate of change of navigation error ΔPerforming a multiplication operation, and adding the navigation error of the kth gene pointkTo obtain the navigation error of the k +1 gene point of each chromosomek+1(ii) a Then according to the error of Beidou navigationBDIf the navigation error of the (k +1) th gene pointk+1>BDThen will beBDAssignment of valuek+1(ii) a Otherwise, the navigation error of the (k +1) th gene pointk+1Is the original value;
step 4) obtaining the navigation error of each gene point of each chromosome according to the step 3), adopting a Gaussian model as a navigation error model, and distributing density functions according to the navigation positions of the navigation error modelCalculating the path cost L(s) of each chromosome by adopting a Monte Carlo sampling method0,sg) Wherein the minimum of all chromosome path costs is the optimal solution for the generation population, XkIs the k geneNavigation position of unmanned surface vehicle at point, s0Is the state at the site of the initiating gene, sgBeing the state at the site of the target gene, sigmakA covariance matrix of the kth gene point;
step 5) taking the dangerous area with the obstacle as a collision area A, integrating the navigation error probability density function in the collision area A according to the navigation error model, solving the collision probability of each gene point of each chromosome, and obtaining a set of collision probabilities { P { of all gene points from the starting point to the target point of each chromosome, namely from the 0 th gene point to the g th gene pointc(X0),Pc(X1),…,Pc(Xg)};
Step 6) according to the collision probability set of all gene points of each chromosome obtained in the step 5), aiming at each chromosome, searching the collision probability set { P } of the chromosomec(X0),Pc(X1),…,Pc(Xg) The maximum value of the collision probability in (1), which is then taken as the collision probability P of the chromosomec(X);
Step 7) according to the path cost of each chromosome obtained in the step 4) and the collision probability P of each chromosome obtained in the step 6)c(X) calculating fitness of each chromosome in the following manner;
judging the collision probability P of each chromosome obtained in the step 6)c(X) is less than or equal to the set value, and if so, calculating the fitness E of each chromosome according to the following formula:
otherwise the fitness E of each chromosome is calculated according to the following formula:
wherein m is a weight value set value>>L(s0,sg);
Adding fitness of all chromosomes to calculate the average value as the average fitness value of the generation population;
step 8) assuming that the nth generation of the population generation is x (n), and if the number of the population generation is x (n)<Set value nsettingEntering step 9); otherwise, processing is as follows:
if at the n- (n) thsetting-1) successive n generations of population to current nth generation populationsettingIn the generation evolution, the optimal solution of the population is not changed, and the variation of the population average adaptation value of two adjacent generations does not exceed 1%, then the chromosome with the optimal solution in the current nth generation is taken as an optimal collision-free path to be solved, a series of path node sequences from a starting point to a target point are obtained according to the coordinate values of gene points forming the chromosome, and the step 10) is carried out after the path node sequences are taken as local route point sequences, otherwise, the step 9) is carried out;
step 9) according to the crossing rate pcI.e. probability values of set crossover operations, randomly selecting pcPerforming crossover operation on M chromosomes, randomly exchanging gene points of crossed individuals to form new chromosomes, and replacing p with maximum fitnesscM chromosomes, and then the variation rate pmI.e. probability values of the mutation operations set, randomly selecting pmPerforming mutation operation on M chromosomes, performing real value mutation on randomly selected gene points of the individuals subjected to mutation, and replacing p with maximum fitnessmM chromosomes; taking the population obtained by the cross mutation operation as a new population, namely, the next generation (n +1 generation) population generation is x (n +1) ═ x (n) +1, and returning to the step 3);
step 10) according to the series of local waypoints obtained in the step 8), taking the first local waypoint as a dynamic planning starting point and taking the second local waypoint as a dynamic planning target point;
step 11) obtaining dynamic obstacle position information and speed information in the sailing process of the water surface unmanned ship according to a sensor, solving the nearest meeting distance DCPA between the water surface unmanned ship and an obstacle and the time TCPA of reaching a nearest meeting point, and then weighting the DCPA and the TCPA according to the following formula to obtain the collision risk rho:
ρ=(ωDDCPA)2+(ωTTCPA)2
wherein, ω isD、ωTRespectively are weighted values;
step 12) judging whether the collision risk rho obtained in the step 11) is smaller than or equal to an empirical value, if so, fusing an obstacle avoidance strategy of a maritime rule and a dynamic constraint condition of the unmanned surface vehicle, and adjusting the course angle or the speed of the unmanned surface vehicle, otherwise, keeping the original course angle and the original speed of the unmanned surface vehicle unchanged;
step 13) according to the change amount △α of the heading angle and the change amount △ V of the speed obtained in the step 12)USVAnd the set planning time △ T, and the current position (X, Y) of the unmanned surface vehicle after planning a △ T time is calculated as:
in the formula, X, Y represents the abscissa and ordinate of the unmanned surface vehicle in a rectangular coordinate system after planning a △ T time, Xprev、YprevRespectively carrying out α V and V on the horizontal coordinate value and the vertical coordinate value of the unmanned surface vehicle in the rectangular coordinate system before planningUSVRespectively representing the course angle and the speed vector of the unmanned surface vehicle;
judging whether the current position of the unmanned surface vehicle is a current dynamic planning target point, if so, entering a step 14), and if not, returning to the step 11);
and 14) judging whether the current position of the unmanned surface vehicle is the target point position Goal or not according to the target point position Goal obtained in the step 1), if so, ending the process of the method, otherwise, taking the current dynamic planning target point as a next dynamic planning starting point, taking a next local route point as a next dynamic planning target point, and returning to the step 11).
2. The navigation error constraint-based method for planning the local hierarchical path of the unmanned surface vehicle according to claim 1, wherein in the step 4), the navigation error model is as follows:
Xk~N(μk,Σk)
in the formula,for the desired navigation position, X, at the kth gene site in each chromosomek=(xk,yk) For the navigation position of the unmanned surface vehicle at the kth gene point, sigmakA covariance matrix of the kth gene point;
the navigation position distribution density function of the navigation error modelComprises the following steps:
the covariance matrix ΣkDetermined according to the following formula:
in the formula,kthe navigation error of the kth gene point; sigmakStandard deviation for the kth gene point;
the path cost L(s) of each chromosome0,sg) The specific calculation process comprises the following steps:
state s at the kth Gene Point according to the 3 σ criterion in the probabilistic modelkIn the navigation uncertainty region of (a), i.e. with the desired navigation position mu at the kth gene sitekIn a 3 sigma confidence region which is the circle center, adopting a Monte Carlo method to sample N sample points and obtaining the state s of the kth gene pointkOf navigation uncertainty region of (a) a set of samples Θk:
State s at the k +1 th Gene sitek+1The navigation uncertain region is sampled by N sample points to obtain the state s of the (k +1) th gene pointk+1Of navigation uncertainty region of (a) a set of samples Θk+1:
According to the sampling set theta of the kth gene pointkSampling set theta with k +1 gene pointk+1Calculating Euclidean distances between N pairs of sampling points from the kth gene point to the (k +1) th gene point, summing the Euclidean distances between the N pairs of sampling points, and averaging to obtain the transfer cost L(s) from the kth gene point to the (k +1) th gene pointk,sk+1):
The state s at the starting Gene site was calculated according to the following formula0Status to target Gene site sgPer chromosome path cost L(s)0,sg):
In the above formula D (X)k,i,Xk+1,i) Represents the Euclidean distance between the ith pair of sampling points from the kth gene point to the (k +1) th gene point.
3. The navigation error constraint-based local hierarchical path planning method for unmanned surface vehicle according to claim 1, wherein the obstacle avoidance strategy fusing the maritime rules in step 12) is as follows:
in the formula:ei+fi1, wherein eiAnd fiTaking 0 or 1 as binary variable, ξ as positive real number and infinity, and avoiding the obstacle by the right side when the unmanned surface vehicle meets the right side of the obstacle in a crossing way or in a front meeting way, wherein e is the momenti=1,fiWhen the unmanned surface vehicle and the left side of the obstacle meet or trace the plane, the obstacle is avoided by the left side, and at the moment ei=0,fi1 is ═ 1; delta V is the velocity vector V of the unmanned surface vehicle relative to the obstacleUOSize;the minimum value of the direction change quantity of the relative speed delta V is determined on the premise that the unmanned surface vehicle can escape from the collision region when the unmanned surface vehicle turns left;when the unmanned surface vehicle turns to the right, the maximum value of the direction change quantity of the relative speed delta V is obtained on the premise that the unmanned surface vehicle can escape from the collision area;is a velocity vector V of the unmanned surface vehicle relative to the barrierUOVelocity vector V of unmanned surface vehicleUSVThe angle of (d); Δ VUSVDelta α is the change of the heading angle of the unmanned surface vehicle;
in the step 12), the course angle and the speed adjustment of the unmanned surface vehicle are obtained by adopting a mixed integer linear programming method according to the collision avoidance strategy and the dynamic constraint of the unmanned surface vehicle.
4. The navigation error constraint-based water surface unmanned ship local hierarchical path planning method according to claim 1, 2 or 3, wherein in the step 11), omega is used for starboard targetsD=5、ωT0.5 for port target ωD=5、ωT=1。
5. The navigation error constraint-based method for planning the local hierarchical path of the unmanned surface vehicle according to claim 1, 2 or 3, wherein the dynamic constraint conditions of the unmanned surface vehicle in the step 12) comprise maximum speed, maximum acceleration and minimum turning radius.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710055413.3A CN106845716B (en) | 2017-01-25 | 2017-01-25 | Navigation error constraint-based water surface unmanned ship local hierarchical path planning method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710055413.3A CN106845716B (en) | 2017-01-25 | 2017-01-25 | Navigation error constraint-based water surface unmanned ship local hierarchical path planning method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106845716A CN106845716A (en) | 2017-06-13 |
CN106845716B true CN106845716B (en) | 2020-09-08 |
Family
ID=59120620
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710055413.3A Active CN106845716B (en) | 2017-01-25 | 2017-01-25 | Navigation error constraint-based water surface unmanned ship local hierarchical path planning method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106845716B (en) |
Families Citing this family (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107422736B (en) * | 2017-08-03 | 2020-03-13 | 大连海事大学 | Unmanned ship autonomous return control method |
CN107329477B (en) * | 2017-08-14 | 2020-05-15 | 河海大学常州校区 | Unmanned ship navigation and automatic driving equipment and method thereof |
CN108416419B (en) * | 2018-01-31 | 2021-07-30 | 重庆邮电大学 | WLAN indoor target intrusion detection method based on multivariate signal characteristics |
CN108445879B (en) * | 2018-03-12 | 2021-02-23 | 上海大学 | Unmanned ship obstacle avoidance method based on collision danger prediction area |
CN108564202B (en) * | 2018-03-18 | 2022-03-18 | 哈尔滨工程大学 | Unmanned ship route optimization method based on environment forecast information |
CN109032131B (en) * | 2018-07-05 | 2021-07-27 | 东南大学 | Dynamic overtaking obstacle avoidance method applied to unmanned automobile |
CN109240288B (en) * | 2018-08-31 | 2021-08-10 | 武汉理工大学 | Unmanned ship collision avoidance path planning method based on track unit under condition of obstacle |
CN109298708B (en) * | 2018-08-31 | 2021-08-17 | 中船重工鹏力(南京)大气海洋信息系统有限公司 | Unmanned ship autonomous obstacle avoidance method integrating radar and photoelectric information |
CN109413665B (en) * | 2018-10-31 | 2022-01-18 | 中国船舶工业系统工程研究院 | Cooperative networking method for unmanned surface vehicle |
CN109375625B (en) * | 2018-11-12 | 2021-06-01 | 智慧航海(青岛)科技有限公司 | Intelligent ship path planning method based on rapid search genetic algorithm |
CN109445445A (en) * | 2018-12-28 | 2019-03-08 | 珠海云洲智能科技有限公司 | A kind of more ship cooperative control systems |
CN109799820B (en) * | 2019-01-22 | 2020-12-22 | 智慧航海(青岛)科技有限公司 | Unmanned ship local path planning method based on comparative random road map method |
CN109916419A (en) * | 2019-03-12 | 2019-06-21 | 哈尔滨工程大学 | A kind of hybrid genetic algorithm unmanned boat real-time route planing method of object-oriented |
CN109931943B (en) * | 2019-03-25 | 2020-09-01 | 智慧航海(青岛)科技有限公司 | Unmanned ship global path planning method and electronic equipment |
CN109960262B (en) * | 2019-03-25 | 2020-05-19 | 华中科技大学 | Unmanned ship dynamic obstacle avoidance method and system based on geometric method |
CN109992894B (en) * | 2019-04-03 | 2021-11-23 | 哈尔滨工程大学 | Unmanned ship local environment modeling method considering perception information error |
CN110146087B (en) * | 2019-06-14 | 2022-11-01 | 哈尔滨工程大学 | Ship path planning method based on dynamic planning idea |
CN110345939B (en) * | 2019-07-02 | 2021-03-19 | 山东科技大学 | Indoor positioning method integrating fuzzy logic judgment and map information |
CN110362089A (en) * | 2019-08-02 | 2019-10-22 | 大连海事大学 | A method of the unmanned boat independent navigation based on deeply study and genetic algorithm |
CN110516877A (en) * | 2019-08-28 | 2019-11-29 | 青岛蓝海未来海洋科技有限责任公司 | A kind of unmanned boat paths planning method and system based on the forward and reverse driving linear variation parameter's genetic algorithm of data |
CN110849370A (en) * | 2019-11-14 | 2020-02-28 | 中国船舶重工集团公司第七0七研究所 | Dynamic route planning method based on unmanned surface vehicle |
CN110906936B (en) * | 2019-12-18 | 2022-11-18 | 哈尔滨工程大学 | Underwater robot path planning method |
CN113009922B (en) * | 2021-04-23 | 2024-03-26 | 元通智能技术(南京)有限公司 | Scheduling management method for robot walking path |
CN114088094A (en) * | 2021-09-27 | 2022-02-25 | 华中光电技术研究所(中国船舶重工集团公司第七一七研究所) | Intelligent route planning method and system for unmanned ship |
CN115930973B (en) * | 2023-02-08 | 2023-07-14 | 中国民航大学 | Unmanned aerial vehicle route planning method and device |
CN116661501B (en) * | 2023-07-24 | 2023-10-10 | 北京航空航天大学 | Unmanned aerial vehicle cluster high dynamic environment obstacle avoidance and moving platform landing combined planning method |
CN116859956B (en) * | 2023-09-04 | 2023-12-08 | 中船(北京)智能装备科技有限公司 | Unmanned ship navigation route determining method, unmanned ship navigation route determining device and unmanned ship navigation route determining equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000172664A (en) * | 1998-10-02 | 2000-06-23 | Yoshinori Haseyama | Optimum route and optimum circulation route searching method |
CN104050329A (en) * | 2014-06-25 | 2014-09-17 | 哈尔滨工程大学 | Method for detecting degree of risk of ship collision |
CN105807769A (en) * | 2016-03-09 | 2016-07-27 | 哈尔滨工程大学 | Unmanned underwater vehicle IVFH (intelligent vector field histogram) collision avoidance method |
CN105867383A (en) * | 2016-05-16 | 2016-08-17 | 哈尔滨工程大学 | Automatic collision preventing control method of USV |
-
2017
- 2017-01-25 CN CN201710055413.3A patent/CN106845716B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000172664A (en) * | 1998-10-02 | 2000-06-23 | Yoshinori Haseyama | Optimum route and optimum circulation route searching method |
CN104050329A (en) * | 2014-06-25 | 2014-09-17 | 哈尔滨工程大学 | Method for detecting degree of risk of ship collision |
CN105807769A (en) * | 2016-03-09 | 2016-07-27 | 哈尔滨工程大学 | Unmanned underwater vehicle IVFH (intelligent vector field histogram) collision avoidance method |
CN105867383A (en) * | 2016-05-16 | 2016-08-17 | 哈尔滨工程大学 | Automatic collision preventing control method of USV |
Non-Patent Citations (6)
Title |
---|
一种基于导航误差空间的无人水下航行器路径规划方法;严浙平等;《兵工学报》;20140831;第35卷(第08期);第1243-1250页 * |
基于海事规则的水面无人艇动态障碍规避方法;杜开君等;《航海工程》;20150630;第44卷(第03期);第119-124页 * |
基于进化遗传算法的无人艇避碰系统;刘佳男;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20160215(第02期);第C036-152页 * |
实时避碰的无人水面机器人在线路径规划方法;冷静等;《智能系统学报》;20150630;第10卷(第03期);第343-348页 * |
无人水面艇路径规划;苏金涛;《指挥控制与仿真》;20151231;第37卷(第06期);第36-40页 * |
智能车辆组合定位与路径导航技术研究;杨易;《中国博士学位论文全文数据库 工程科技Ⅱ辑》;20071115(第05期);第C035-2页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106845716A (en) | 2017-06-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106845716B (en) | Navigation error constraint-based water surface unmanned ship local hierarchical path planning method | |
Zinchenko et al. | Automatic collision avoidance system with many targets, including maneuvering ones | |
CN107504972B (en) | A kind of aircraft's flight track method and device for planning based on dove group's algorithm | |
CN109933067B (en) | Unmanned ship collision avoidance method based on genetic algorithm and particle swarm algorithm | |
Zhao et al. | A real-time collision avoidance learning system for Unmanned Surface Vessels | |
CN106970648A (en) | Unmanned plane multi-goal path plans combined method for searching under the environment of city low latitude | |
CN110362089A (en) | A method of the unmanned boat independent navigation based on deeply study and genetic algorithm | |
CN109460045B (en) | Improved ant colony optimization-based collision avoidance planning method for USV under dynamic obstacle online perception | |
CN110906935B (en) | Unmanned ship path planning method | |
Zheng et al. | A Decision‐Making Method for Ship Collision Avoidance Based on Improved Cultural Particle Swarm | |
Lazarowska | Ant colony optimization based navigational decision support system | |
Blaich et al. | Fast grid based collision avoidance for vessels using A∗ search algorithm | |
Wang et al. | Cooperative collision avoidance for unmanned surface vehicles based on improved genetic algorithm | |
Yao et al. | A hierarchical architecture using biased min-consensus for USV path planning | |
Guan et al. | Autonomous collision avoidance of unmanned surface vehicles based on improved A-star and dynamic window approach algorithms | |
Du et al. | Trajectory-cell based method for the unmanned surface vehicle motion planning | |
CN113538973B (en) | Automatic ship collision avoidance method based on improved particle swarm optimization | |
CN109765890B (en) | Multi-USV group collaborative collision avoidance planning method based on genetic algorithm | |
CN115143970B (en) | Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation | |
Zyczkowski et al. | Collision risk-informed weather routing for sailboats | |
CN113985899A (en) | Underwater robot global path planning method based on interval multi-objective optimization | |
CN113032896A (en) | Collision avoidance aid decision-making method based on ship driver preference | |
CN109556609A (en) | A kind of collision prevention method and device based on artificial intelligence | |
Wu et al. | Multi-vessels collision avoidance strategy for autonomous surface vehicles based on genetic algorithm in congested port environment | |
Liu et al. | PE-A* algorithm for ship route planning based on field theory |
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 |