CN109933067A - A kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm - Google Patents
A kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm Download PDFInfo
- Publication number
- CN109933067A CN109933067A CN201910185335.8A CN201910185335A CN109933067A CN 109933067 A CN109933067 A CN 109933067A CN 201910185335 A CN201910185335 A CN 201910185335A CN 109933067 A CN109933067 A CN 109933067A
- Authority
- CN
- China
- Prior art keywords
- path
- unmanned boat
- collision prevention
- barrier
- genetic algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The present invention relates to maritime affairs intelligent transport technology unmanned boat collision prevention fields, and in particular to a kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm.The research of relevant parameter and collision prevention constraint rule in unmanned boat collision prevention path planning is carried out first, then the path planning that the unmanned surface vehicle based on genetic algorithm evades static-obstacle thing is carried out, finally carry out the dynamic collision prevention path planning combined based on genetic algorithm with particle swarm algorithm, path optimization is completed, feasible collision prevention path is exported and restores navigation；Relative to traditional unmanned boat collision prevention technology, the present invention can obtain optimal path planning, accurately avoid collision, it is ensured that unmanned boat arrives safe and sound target point.
Description
Technical field
The present invention relates to maritime affairs intelligent transport technology unmanned boat collision prevention fields, and in particular to one kind is based on genetic algorithm and grain
The unmanned boat collision prevention method of swarm optimization.
Background technique
Unmanned boat (USV) is a kind of collection contexture by self, and autonomous navigation independently completes environment sensing, the functions such as target acquisition
The small-size water surface motion platform being integrated, it has also become explore the essential equipment of marine resources.
Since path planning and collision prevention are then the key points of unmanned boat autonomous navigation.Therefore enter waters in unmanned boat to hold
Before row task, the hydrological data according to known sea area is needed, cooks up a whole navigation path.Since marine environment is multiple
Miscellaneous changeable, can not predict will appear any situation during navigation, for example meet with wind and waves, passing ships etc..Unmanned boat
Peripheral situation is needed to constantly detect, environmental information, accurate quickly adjustment operational configuration are obtained, avoiding obstacles are needed according to task
It asks and reaches specified target point, execution task is simultaneously restored navigation.
The collision prevention technology of unmanned boat reflects the height of maritime affairs unmanned boat intelligent level to a certain extent, is unmanned boat
One of important research content of key technology area.Application No. is the patents of CN201610942213.5, and one kind is based on improvement ant
The unmanned boat collision prevention method of group's algorithm plans that the method has one to the safety lanes of unmanned boat using ant group algorithm is improved
It settles finally sex-limited, and is not calculated respectively for the path planning of evading of static-obstacle thing and dynamic barrier.
Summary of the invention
The unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm that the purpose of the present invention is to provide a kind of, with
It to optimal path planning, accurately avoids collision, it is ensured that unmanned boat arrives safe and sound target point.
The embodiment of the present invention provides a kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm, comprising:
Step 1: it the research of relevant parameter and collision prevention constraint rule in unmanned boat collision prevention path planning: is navigated according to unmanned boat
Unmanned boat position, speed and Obstacle Position, speed involved in capable mathematical model parameter obtain relatively fast between the two
Degree, relative position, relative bearing kinematic parameter, by summarize unmanned boat and barrier front avoidance, overtake avoidance, right string intersects
Intersect the different collision situations of four kinds of avoidance with left string, obtain unmanned boat Collision Avoidance At Sea constraint rule and space collision danger level with
Time collision risk；
Step 2: the unmanned surface vehicle based on genetic algorithm evades the path planning of static-obstacle thing: passing through above-mentioned nothing
People's ship navigates by water water environment relevant parameter information, initializes to the relevant parameter of genetic algorithm, generates the most primary vaccination in path
Group, and enter the iterative cycles of genetic algorithm, calculate fitness function；According to obtained fitness function value, pass through " wheel disc
Gambling " method selects follow-on chromosome and is intersected, made a variation, repairing operation optimization population, and iteration obtains to advise after the completion
Keep away the optimal path of static-obstacle thing；
Step 3: the dynamic collision prevention path planning combined based on genetic algorithm with particle swarm algorithm: real-time monitoring ocean
Environment whether there is dynamic barrier, if there is dynamic barrier, judge unmanned boat between dynamic barrier at a distance from,
Risk-Degree of Collision is calculated, collision is if possible generated, judges whether the motion state of dynamic barrier can be surveyed；If can survey,
Using the conventional collision prevention in genetic algorithm；If can not survey, evading for dynamic barrier is carried out using particle swarm algorithm；Finally
Path optimization is completed, feasible collision prevention path is exported and restores navigation；
The step 1, comprising:
Relevant parameter in unmanned boat collision prevention path planning and collision prevention constraint rule are studied: according to unmanned boat navigation
Unmanned boat position, speed and Obstacle Position, speed involved in mathematical model parameter, obtain between the two relative velocity,
Relative position, relative bearing kinematic parameter, by summarize unmanned boat and barrier front avoidance, overtake avoidance, right string intersects and
Left string intersects four kinds of avoidance different collision situations, obtain unmanned boat Collision Avoidance At Sea constraint rule and space collision danger level and when
Between Risk-Degree of Collision；
Wherein, DCPA is enabled to represent unmanned boat and the most short distance of meeting of barrier, TCPA represents most short Encounter Time, the sky
Between Risk-Degree of Collision calculation method are as follows:
In above formula,RT、αTIndicate unmanned boat and barrier it is opposite away from
From, the course in relative velocity direction and barrier,
In above formula, θTIndicate the relative angle of unmanned boat and barrier course, d2=2 × d1；
The calculation method of the time collision risk are as follows:
In above formula,And the u as TCPA > 0tTSection 2 takes negative sign, otherwise takes
Positive sign；
It analyzes and researches to the calculated result of DCPA and TCPA, obtains and crash between unmanned boat and mobile object
The comprehensive evaluation index of possibility:
The step 2, comprising:
Unmanned surface vehicle based on genetic algorithm evades the path planning of static-obstacle thing: being navigated by water by above-mentioned unmanned boat
Water environment relevant parameter information, initializes the relevant parameter of genetic algorithm, generates the initial population in path, and enters
The iterative cycles of genetic algorithm calculate fitness function；According to obtained fitness function value, selected by " roulette " method
Follow-on chromosome is intersected, is made a variation, repairing operation optimization population, after the completion of iteration obtains that static-obstacle thing can be evaded
Optimal path；
Wherein, the initial population in the path are as follows:
Original movement road is generated according to the relevant information of the demand of task and the initial position of unmanned boat and target point
Diameter, that is, every chromosome:
(xi1, yi1)→(xi2, yi2)→…→(xili, yili)
In above formula, i (1=0,1,2...n) represents a paths of unmanned boat navigation, and n indicates initial population number, and will
Chromosome coding at type real, wherein if between unmanned boat initial position and target point and be not present obstructing objects,
One smooth shortest path of the original path obtained, if there are barriers between unmanned boat initial position and target point
Body, this can be iterated update on the basis of original path and generate a plurality of new path, can be by its point for each path
For multistage part, calculating analysis is carried out to the paths difference section part respectively, obtains the relevant information of the paths, used
Delta indicates the length value of the division section of any paths, kδThe scale parameter value positive as one, dnumIndicate the paths
The sum for the section being divided into:
In above formula, (xmax,ymax) and (xmin,ymin) be artificial window coordinate range, s (xs,ys) and e (xe,ye) it is nothing
The initial position and target point of people's ship；
After the computation partition of path number of segment above, the method initialized by heuristic genetic group is carried out
Initialization of population generates corresponding dye gene, if the i-th paths are expressed as P by a total of n item of initial original pathi,
If dnum=3, then appoint at random in jth (j=1,2,3) section and takes two point pi(2j-1)And pi(2j), then in the range of the two points
Interior any generation pointThis abscissa is limited inIn range, at the beginning of ordinate is confined to path
Beginning position s (xs,ys) and target point e (xe,ye) between, i.e.,Then according to priority by pi(2j-1),
pi(2j)It is connected, ultimately generates path Pi:
Obtain feasible path collection
Wherein, the fitness function are as follows:
It is Value (P by individual adaptation degree function setup*)=min [f1(P),f2(P),f3(P)]:
(1)The length of delegated path, the path length of i-th chromosome are as follows:
In above formula, miIt is PiInfeasible path number, C in path1For a suitable positive number；
(2)The slickness for indicating path, when the path of chromosome divides gene loci
Number is greater than 2, the P of unmanned surface vehicleiPath average corner value:
Wherein, aij(j=2 ..., liIt -1) is pi(j-1)pijWith pijpi(j+1)Between angle (0≤aij≤π),miAnd kiIt is aij
In be not less thanNumber, i.e., if some turning is not less thanWhen, then punishment calculating, C are carried out to target value2It is one suitable
Positive number, work as liWhen=2, path PiFor the line of initial point to target point, Turning (Pi)=mi×C2；
(3)The safety for indicating path, if path PiIt is feasible, danger
(di)=1/di, wherein di > 0 indicates minimum value of the navigation route of unmanned boat apart from static-obstacle thing；If path P i is
It is infeasible, danger (Pi)=mi×C3, miIt is less than safety for the distance between the route segment of path individual and barrier
The quantity of distance, C3It is then a positive number appropriate；
The step 3, comprising:
The dynamic collision prevention path planning combined based on genetic algorithm with particle swarm algorithm: around real-time monitoring marine environment
With the presence or absence of dynamic barrier, if there is dynamic barrier, judge unmanned boat between dynamic barrier at a distance from, calculate and touch
Danger level is hit, collision is if possible generated, judges whether the motion state of dynamic barrier can be surveyed；If can survey, using heredity
Conventional collision prevention in algorithm；If can not survey, evading for dynamic barrier is carried out using particle swarm algorithm；Finally complete path
Optimization, exports feasible collision prevention path and restores navigation；
Wherein, the dynamic collision prevention method combined based on genetic algorithm with particle swarm algorithm are as follows:
Δ v is resolved into Δ voWith Δ vr,
Assuming that the speed v of dynamic mobile object in a relatively short period of timeobsAnd direction change is little, can not consider, i.e.,
dvobsβ=0=0, d
In above formula,For vUSV、vobsBetween angle, Δ γ must satisfy the limitation of following condition:
By the multidate information of relevant equipment monitoring ambient ocean environment, if it find that dynamic barrier, judgement is dynamic
Whether the motion state of state barrier can be surveyed；It, can since genetic algorithm is adapted to solve complicated optimization problem if can survey
The globally optimal solution of optimization problem is found out, then carries out the dynamic collision prevention based on genetic algorithm；If can not survey, since population is calculated
Method has the quickish speed for approaching optimal solution, can effectively optimize to the parameter of system；It then carries out based on particle
The dynamic collision prevention of group's algorithm；
Collision prevention based on genetic algorithm can regard the objective optimisation problems under a many condition as:
f(ΔvUSV, Δ α) and it is fitness function in genetic algorithm, the optimal solution found out both will not be with known quiescent state obstacle
Object collision, it is also necessary to which the constraint for meeting unmanned boat Rules of Navigation calculates it according to the information of unmanned boat and dynamic barrier
The distance between, DCPA and TCPA value carries out collision prevention processing if the danger that can be collided, and collision prevention processing can lead to
Change navigation direction is crossed to realize:
R in above formulaTFor unmanned boat between dynamic barrier at a distance from；θTFor the relative bearing of dynamic barrier；β is opposite
The corner of the line of motion；For the speed of a ship or plane of dynamic barrier and unmanned surface vehicle ratio；Δ C is the angle of evacuation；
Steering angle Δ C is acquired by iterative method:
The motion state that surrounding dynamic barrier is monitored by equipment moment such as relevant AIS radars predicts that its movement is walked
To the location of current unmanned boat is then set as new initial position, is obtained by the distance of route segment of the length of safeD
One sub- target point, by the iteration optimization of genetic algorithm, neighbouring barrier arrival specific item punctuate can be evaded by cooking up one
Then path is designated as new initial position with the specific item again, repeated the above process, successfully arrive at specified target point and complete to appoint
Business；
During the collision prevention of unpredictable dynamic barrier motion state, the advance each time of unmanned boat will have pre-
Sentence, using polar coordinates, describes the position of unmanned boat and dynamic barrier, the length of unmanned boat is r, and mobile radius is ρ, current institute
It is set to (A in place1,B1), next step target point is (A2,B2), the position of dynamic barrier at this time is (C1,D1)；When Barrier is indicated in the range of unmanned boat safety moving, is carried out at this time based on grain
Swarm optimization collision prevention processing；
Inertia weight maximum value is set as ωmax, minimum value ωmin, Studying factors are respectively C1, C2, group D, maximum
The number of iterations is Dmax.Using current point as origin, ρ is that radius makees polar coordinates, carries out n equal portions.M particle generates m × n at random
Population and position X and speed V, current optimal location is setForm initial population t0；Make
PbestiRepresent the optimal value that i-th of particle search arrives, GbestiThe optimal value that entire group search arrives is represented, calculates initial kind
The fitness value of group, to global optimum GbestiIt updates, wherein fitness function includes path length and degree of safety, shortest path
The objective function of electrical path length:
In above formula, (xji,yji) be path j on i point coordinate, (xji-1,yji-1) it is (x on the j of pathji,yji) a point
Coordinate, at this point, next coordinate (x of unmanned boat path planning predictionjn,yjn), it needs this point and target point (g1,g2) even
It picks up to calculate path, entire path length function are as follows:
Y=Yfit1+Yfit2
Degree of safety function:
In above formula, xkFor the position coordinates of barrier；
In summary, fitness function: F=A × Y+B × OB (i), wherein A, the weighted factor of two function of B, for greater than etc.
In zero any real number, for complete collision prevention, B < A, according to speed, position iterative formula updates, and generates new population t1, grain
Subgroup speed iterative formula are as follows:
Vi,j(t+1)=ω Vi,j(t)+c1r1(Pi,j(t)-xi,j(t))+c2r2(Pg,j(t)-xi,j(t))
ωk+1=ωmin+(ωk-ωmin)×((Kmax-k)/Kmax)n
In above formula, r1, r2Random real number greater than 0 less than 1, Pi,jIt is the optimal location that particle i is searched out so far, Pg,i
It is global optimum position, ωkFor the resulting value of current iteration, initial value ωmax, k expression current iteration number；KmaxIt represents
Maximum number of iterations；
Calculate new population t1Fitness value, if better than generating new population t if previous generation2, it is otherwise converted into rectangular co-ordinate,
Loop iteration generates new population, until finding optimal collision prevention path；
The beneficial effects of the present invention are:
1., for the unmanned boat navigated by water in marine environment complicated and changeable, being cooked up invention introduces genetic algorithm
One optimal path is reached specified target point by mission requirements to save various resources；
2. the present invention makes improvements while introducing genetic algorithm, in order to balance the influence of marine stormy waves stream, will
Deletion, reparation and smoothing operator is added, unmanned boat is enabled to cook up optimal motion path under genetic algorithm；
3. present invention introduces the algorithm that genetic algorithm and particle swarm algorithm combine, for dynamic emergent in ocean
Barrier, unmanned boat can quick and precisely make a response, and cook up optimal collision prevention path, it is ensured that unmanned boat arrives safe and sound target
Point.
Detailed description of the invention
Fig. 1 is unmanned boat path planning collision prevention flow chart of the present invention；
Fig. 2 is unmanned boat static state global path planning flow chart of the present invention；
Fig. 3 is barrier of the present invention to path distance schematic diagram；
Fig. 4 is Obstacle avoidance model schematic diagram of the present invention；
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing to the present invention
It is described further:
The technical scheme of the present invention is realized as follows:
1, relevant parameter in unmanned boat path planning and collision prevention and collision prevention constraint rule are studied:
The mathematical model parameter of unmanned boat movement is related to the position of unmanned boat, the position of speed and barrier, speed.By
This two relative velocity drawn, relative position, relative bearing kinematic parameter.Sum up unmanned boat and barrier front avoidance,
Overtake avoidance, right string intersects and left string intersects four kinds of different collision situations of avoidance.Unmanned boat Collision Avoidance At Sea constraint rule is drawn,
And from time and two, space angle, draw the concept of danger level, wherein DCPA represent unmanned boat and barrier most it is short meet away from
From TCPA represents most short Encounter Time.
(1) space collision danger level
In formulaRT、αTIndicate relative distance, phase of the unmanned boat with barrier
To the course of directional velocity and barrier；
Wherein θTIndicate the relative angle of unmanned boat and barrier course, d2=2 × d1。
(2) time collision risk
WhereinAnd the u as TCPA > 0tTSection 2 takes negative sign, instead
Take positive sign.
It analyzes and researches to the calculated result of DCPA and TCPA, obtains and crash between unmanned boat and mobile object
The comprehensive evaluation index of possibility:
2, the unmanned surface vehicle path planning based on genetic algorithm
The algorithmic procedure of unmanned boat path planning is to obtain unmanned boat first to navigate by water water environment parameter, predominantly barrier
Information, seakeeping etc..Then define the fitness function of population at individual, setting genetic algorithm the number of iterations, chromosome quantitative,
The probability for intersecting and making a variation carries out the global path that genetic manipulation obtains unmanned boat navigation finally by selection algorithm.
Original fortune is generated first, in accordance with the relevant information of the demand of task and the initial position of unmanned boat and target point
Dynamic path, that is, every chromosome:
(xi1,yi1)→(xi2,yi2)→…→(xili,yili)
Wherein i (1=0,1,2...n) represents a paths of unmanned boat navigation, and n indicates initial population number, and will dye
Colour solid is encoded into type real.
Wherein if between unmanned boat initial position and target point and be not present obstructing objects, it is concluded that original path
One smooth shortest path.If there are obstructing objects between unmanned boat initial position and target point, this can be on original road
It is iterated update on the basis of diameter and generates a plurality of new path.It can be classified as multistage part for each path, respectively
Calculating analysis is carried out to the paths difference section part, obtains the relevant information of the paths.Any paths are indicated with delta
Division section length value, kδThe scale parameter value positive as one, dnumIndicate the sum for the section that the paths are divided into.
Wherein, (xmax,ymax) and (xmin,ymin) be artificial window coordinate range, s (xs,ys) and e (xe,ye) for nobody
The initial position and target point of ship.
After the computation partition of path number of segment above, the method initialized by heuristic genetic group is carried out
Initialization of population generates corresponding dye gene.If the i-th paths are expressed as P by a total of n item of initial original pathi,
If dnum=3, then appoint at random in jth (j=1,2,3) section and takes two point pi(2j-1)And pi(2j-1), then in the model of the two points
Enclose interior any generation pointThis puts horizontal seat mark and is limited inIn range, ordinate is confined to path
Initial position s (xs,ys) and target point e (xe,ye) between, i.e.,Then according to priority by pi(2j-1),
pi(2j)It is connected, ultimately generates path Pi:
Obtain feasible path collection
It is Value (P by individual adaptation degree function setup after initial population is set*)=min [f1(P),f2(P),f3
(P)]。
Wherein:
(1)The length of delegated path.The path length of i-th chromosome is
Wherein, miIt is PiInfeasible path number, C in path1For a suitable positive number.
(2)Indicate the slickness in path.When the path of chromosome divides gene loci
Number is greater than 2, the P of unmanned surface vehicleiPath average corner value:
Wherein, aij(j=2 ..., li-1) it is pi(j-1)pijWith pijpi(j+1)Between angle (0≤aij≤π),miAnd kiIt is aij
In be not less than pi/2 number, i.e., if some turning be not less than pi/2 when, punishment calculating, C are carried out to target value2It is one
Suitable positive number, works as liWhen=2, path PiFor the line of initial point to target point, Turning (Pi)=mi × C2
(3)Indicate the safety in path
If path PiIt is feasible, danger (di)=1/di, wherein diThe navigation route distance of the expression unmanned boat of > 0
The minimum value of static-obstacle thing；If path PiIt is infeasible, danger (Pi)=mi × C3, mi is the road of path individual
The distance between diameter section and barrier are less than the quantity of safe distance, C3It is then a positive number appropriate.According to fitness function
Value selects follow-on chromosome by proportional algorithm and the genetic manipulations such as is intersected, makes a variation, repairs, to optimize population.Most
Afterwards, by iteration optimization, the global path of a clear is cooked up.
3, the dynamic collision prevention combined based on genetic algorithm with particle swarm algorithm is carried out, key step is as follows:
As shown in 4 Obstacle avoidance model schematic diagram of attached drawing, Δ v is resolved into Δ voWith Δ vr。
Assuming that the speed v of dynamic mobile object in a relatively short period of timeobsAnd direction change is little, can not consider, i.e.,
dvObsβ=0=0, d
WhereinFor vUSV、vobsBetween angle.Δ γ must satisfy following condition limitation:
Pass through the multidate information of relevant equipment monitoring ambient ocean environment.If it find that dynamic barrier, judgement is dynamic
Whether the motion state of state barrier can be surveyed；It, can since genetic algorithm is adapted to solve complicated optimization problem if can survey
The globally optimal solution of optimization problem is found out, then carries out the dynamic collision prevention based on genetic algorithm；If can not survey, since population is calculated
Method has the quickish speed for approaching optimal solution, can effectively optimize to the parameter of system；It then carries out based on particle
The dynamic collision prevention of group's algorithm.
Collision prevention based on genetic algorithm can regard the objective optimisation problems under a many condition as:
f(ΔvUSV, Δ α) and it is fitness function in genetic algorithm, the optimal solution found out both will not be with known quiescent state obstacle
Object collision, it is also necessary to meet the constraint of unmanned boat Rules of Navigation.According to the information of unmanned boat and dynamic barrier, it is calculated
The distance between, DCPA, TCPA value, if the danger that can be collided, carries out collision prevention processing.Collision prevention processing can pass through
Change navigation direction to realize.
R in formulaTFor unmanned boat between dynamic barrier at a distance from；r
θTFor the relative bearing of dynamic barrier；
β is the corner of relative movement line；
For the speed of a ship or plane of dynamic barrier and unmanned surface vehicle ratio；
Δ C is the angle of evacuation.
Steering angle Δ C is acquired by iterative method:
The motion state that surrounding dynamic barrier is monitored by equipment moment such as relevant AIS radars predicts that its movement is walked
To the location of current unmanned boat is then set as new initial position, is obtained by the distance of route segment of the length of safeD
One sub- target point, by the iteration optimization of genetic algorithm, neighbouring barrier arrival specific item punctuate can be evaded by cooking up one
Then path is designated as new initial position with the specific item again, repeated the above process, successfully arrive at specified target point and complete to appoint
Business.
During the collision prevention of unpredictable dynamic barrier motion state, the advance each time of unmanned boat will have pre-
Sentence.Using polar coordinates, the position of unmanned boat and dynamic barrier is described.The length of unmanned boat is r, and mobile radius is ρ, current institute
It is set to (A in place1,B1), next step target point is (A2,B2), the position of dynamic barrier at this time is (C1,D1)；WhenBarrier is indicated in the range of unmanned boat safety moving, needs to carry out at this time
It is handled based on particle swarm algorithm collision prevention.
Inertia weight maximum value is set as ωmax, minimum value ωmin, Studying factors are respectively C1, C2, group D, maximum
The number of iterations is Dmax.Using current point as origin, ρ is that radius makees polar coordinates, carries out n equal portions.M particle generates m × n's at random
Population and position X and speed V, are arranged current optimal location Pi=(pi1,p2...piD), form initial population t0.Make
PbestiRepresent the optimal value that i-th of particle search arrives, GbestiRepresent the optimal value that entire group search arrives.Calculate initial kind
The fitness value of group, to global optimum GbestiIt updates.
Wherein fitness function includes path length and degree of safety.
The objective function of shortest path length:
Wherein (xji,yji) be path j on i point coordinate, (xji-1,yji-1) it is (x on the j of pathji,yji) a point sit
Mark.At this point, next coordinate (x of unmanned boat path planning predictionjn,yjn), it needs this point and target point (g1,g2) connection
Get up and calculates path, entire path length function are as follows:
Y=Yfit1+Yfit2
Degree of safety function:
Wherein xkFor the position coordinates of barrier.
In summary, fitness function: F=A × Y+B × OB (i)
The weighted factor of two function of wherein A, B, for any real number more than or equal to zero, for complete collision prevention, B < A.
According to speed, position iterative formula updates, and generates new population t1。
Particle group velocity iterative formula are as follows:
Vi,j(t+1)=ω Vi,j(t)+c1r1(Pi,j(t)-xi,j(t))+c2r2(Pg,j(t)-xi,j(t))
ωk+1=ωmin+(ωk-ωmin)×((Kmax-k)/Kmax)n
Wherein r1, r2Random real number greater than 0 less than 1, Pi,jIt is the optimal location that particle i is searched out so far, Pg,iIt is complete
Office's optimal location.ωkFor the resulting value of current iteration, initial value ωmax；K indicates current iteration number；KmaxRepresent maximum
The number of iterations
Calculate new population t1Fitness value, if better than generating new population t if previous generation2, otherwise it is converted into rectangular co-ordinate.
Loop iteration generates new population, until finding optimal collision prevention path.
Claims (4)
1. a kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm characterized by comprising
Step 1: the research of relevant parameter and collision prevention constraint rule in unmanned boat collision prevention path planning: according to unmanned boat navigation
Unmanned boat position, speed and Obstacle Position, speed involved in mathematical model parameter, obtain between the two relative velocity,
Relative position, relative bearing kinematic parameter, by summarize unmanned boat and barrier front avoidance, overtake avoidance, right string intersects and
Left string intersects four kinds of avoidance different collision situations, obtain unmanned boat Collision Avoidance At Sea constraint rule and space collision danger level and when
Between Risk-Degree of Collision；
Step 2: the unmanned surface vehicle based on genetic algorithm evades the path planning of static-obstacle thing: passing through above-mentioned unmanned boat
Water environment relevant parameter information is navigated by water, the relevant parameter of genetic algorithm is initialized, the initial population in path is generated, and
Into the iterative cycles of genetic algorithm, fitness function is calculated；According to obtained fitness function value, pass through " roulette " method
It selects follow-on chromosome intersected, made a variation, repairing operation optimization population, after the completion of iteration obtains that static barrier can be evaded
Hinder the optimal path of object；
Step 3: the dynamic collision prevention path planning combined based on genetic algorithm with particle swarm algorithm: real-time monitoring marine environment
Around whether there is dynamic barrier, if there is dynamic barrier, judge unmanned boat between dynamic barrier at a distance from, calculate
Risk-Degree of Collision out if possible generates collision, judges whether the motion state of dynamic barrier can be surveyed；If can survey, use
Conventional collision prevention in genetic algorithm；If can not survey, evading for dynamic barrier is carried out using particle swarm algorithm；It finally completes
Path optimization exports feasible collision prevention path and restores navigation.
2. a kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm according to claim 1, feature
It is, the step 1, comprising:
Relevant parameter in unmanned boat collision prevention path planning and collision prevention constraint rule are studied: the mathematics navigated by water according to unmanned boat
Unmanned boat position, speed and Obstacle Position, speed involved in model parameter obtain relative velocity, opposite between the two
Position, relative bearing kinematic parameter, by summarizing unmanned boat and barrier front avoidance, overtaking that avoidance, right string intersects and Zuo Xian
Intersect four kinds of avoidance different collision situations, obtains unmanned boat Collision Avoidance At Sea constraint rule and space collision danger level and the time touches
Hit danger level；
Wherein, DCPA is enabled to represent unmanned boat and the most short distance of meeting of barrier, TCPA represents most short Encounter Time, and the space is touched
Hit the calculation method of danger level are as follows:
In above formula,RT、αTThe relative distance of expression unmanned boat and barrier,
The course in relative velocity direction and barrier,
In above formula, θTIndicate the relative angle of unmanned boat and barrier course, d2=2 × d1；
The calculation method of the time collision risk are as follows:
In above formula,And the u as TCPA > 0tTSection 2 takes negative sign, otherwise takes just
Number；
It analyzes and researches to the calculated result of DCPA and TCPA, obtains the possibility that crashes between unmanned boat and mobile object
The comprehensive evaluation index of property:
3. a kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm according to claim 1, feature
It is: the step 2, comprising:
Unmanned surface vehicle based on genetic algorithm evades the path planning of static-obstacle thing: navigating by water waters by above-mentioned unmanned boat
Environmental correclation parameter information initializes the relevant parameter of genetic algorithm, generates the initial population in path, and enter heredity
The iterative cycles of algorithm calculate fitness function；According to obtained fitness function value, selected by " roulette " method next
The chromosome in generation is intersected, is made a variation, repairing operation optimization population, after the completion of iteration obtains that static-obstacle thing can be evaded most
Shortest path；
Wherein, the initial population in the path are as follows:
Original motion path is generated i.e. according to the relevant information of the demand of task and the initial position of unmanned boat and target point
Every chromosome:
(xi1, yi1)→(xi2, yi2)→…→(xili, yili)
In above formula, i (1=0,1,2...n) represents a paths of unmanned boat navigation, and n indicates initial population number, and will dyeing
Body is encoded into type real, wherein if between unmanned boat initial position and target point and be not present obstructing objects, it is concluded that
One smooth shortest path of original path, if there are obstructing objects between unmanned boat initial position and target point, this
It can be iterated update on the basis of original path and generate a plurality of new path, multistage can be classified as each path
Part carries out calculating analysis to the paths difference section part respectively, obtains the relevant information of the paths, indicated with delta
The length value of the division section of any paths, kδThe scale parameter value positive as one, dnumIndicate what the paths were divided into
The sum of section:
In above formula, (xmax, ymax) and (xmin, ymin) be artificial window coordinate range, s (xs, ys) and e (xe, ye) it is unmanned boat
Initial position and target point；
After the computation partition of path number of segment above, the method initialized by heuristic genetic group is subjected to population
Initialization, generates corresponding dye gene, if the i-th paths are expressed as P by a total of n item of initial original pathiIf dnum
=3, then appoint at random in jth (j=1,2,3) section and takes two point pi(2j-1)And pi(2j), then appoint in the range of the two points
Meaning generates pointThis abscissa is limited inIn range, it is initial that ordinate is confined to path
Position s (xs, ys) and target point e (xe, ye) between, i.e.,Then according to priority by pi(2j-1),
pi(2j)It is connected, ultimately generates path Pi:
Obtain feasible path collection
Wherein, the fitness function are as follows:
It is Value (P by individual adaptation degree function setup*)=min [f1(P), f2(P), f3(P)]:
(1)The length of delegated path, the path length of i-th chromosome are as follows:
In above formula, miIt is PiInfeasible path number, C in path1For a suitable positive number；
(2)The slickness for indicating path, when the path of chromosome divides the number of gene loci
Greater than 2, the P of unmanned surface vehicleiPath average corner value:
Wherein, aij(j=2 ..., liIt -1) is pi(j-1)pijWith pijpi(j+1)Between angle (0≤aij≤π),miAnd kiIt is aijIn not
It is less thanNumber, i.e., if some turning is not less thanWhen, then punishment calculating, C are carried out to target value2Be one it is suitable just
Number, works as liWhen=2, path PiFor the line of initial point to target point, Turning (Pi)=mi×C2；
(3)The safety for indicating path, if path PiIt is feasible, danger (di)=1/
Di, wherein di > 0 indicates minimum value of the navigation route of unmanned boat apart from static-obstacle thing；If path PiBe it is infeasible,
danger(Pi)=mi×C3, miIt is less than the quantity of safe distance for the distance between the route segment of path individual and barrier,
C3It is then a positive number appropriate.
4. a kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm according to claim 1, feature
It is: the step 3, comprising:
The dynamic collision prevention path planning combined based on genetic algorithm with particle swarm algorithm: around real-time monitoring marine environment whether
There are dynamic barriers, if there is dynamic barrier, judge unmanned boat between dynamic barrier at a distance from, calculate collision danger
Dangerous degree if possible generates collision, judges whether the motion state of dynamic barrier can be surveyed；If can survey, using genetic algorithm
In conventional collision prevention；If can not survey, evading for dynamic barrier is carried out using particle swarm algorithm；It is excellent to finally complete path
Change, exports feasible collision prevention path and restore navigation；
Wherein, the dynamic collision prevention method combined based on genetic algorithm with particle swarm algorithm are as follows:
Δ v is resolved into Δ voWith Δ vr,
Assuming that the speed v of dynamic mobile object in a relatively short period of timeobsAnd direction change is little, can not consider, i.e.,
dvobsβ=0=0, d
In above formula,For vUSV、vobsBetween angle, Δ γ must satisfy the limitation of following condition:
By the multidate information of relevant equipment monitoring ambient ocean environment, if it find that dynamic barrier, judges that dynamic hinders
Whether hinder the motion state of object can survey；If can survey, since genetic algorithm is adapted to solve complicated optimization problem, can find out
The globally optimal solution of optimization problem then carries out the dynamic collision prevention based on genetic algorithm；If can not survey, since particle swarm algorithm has
There is the quickish speed for approaching optimal solution, effectively the parameter of system can be optimized；It then carries out calculating based on population
The dynamic collision prevention of method；
Collision prevention based on genetic algorithm can regard the objective optimisation problems under a many condition as:
f(ΔvUSV, Δ α) and it is fitness function in genetic algorithm, the optimal solution found out will not both be touched with known quiescent state barrier
Hit, it is also necessary to the constraint for meeting unmanned boat Rules of Navigation, according to the information of unmanned boat and dynamic barrier, calculate them it
Between distance, DCPA and TCPA value carries out collision prevention processing if the danger that can be collided, and collision prevention processing can be by changing
Become navigation direction to realize:
R in above formulaTFor unmanned boat between dynamic barrier at a distance from；θTFor the relative bearing of dynamic barrier；β is relative motion
The corner of line；For the speed of a ship or plane of dynamic barrier and unmanned surface vehicle ratio；Δ C is the angle of evacuation；
Steering angle Δ C is acquired by iterative method:
The motion state that surrounding dynamic barrier is monitored by equipment moment such as relevant AIS radars predicts its movement trend, so
The location of current unmanned boat is set as new initial position afterwards, obtains one by the distance of route segment of the length of safeD
Specific item punctuate, by the iteration optimization of genetic algorithm, the path that neighbouring barrier reaches specific item punctuate can be evaded by cooking up one,
Then new initial position is designated as with the specific item again, is repeated the above process, specified target point is successfully arrived at and completes task；
During the collision prevention of unpredictable dynamic barrier motion state, the advance each time of unmanned boat will have anticipation, benefit
With polar coordinates, the position of unmanned boat and dynamic barrier is described, the length of unmanned boat is r, and mobile radius is ρ, is currently located position
It is set to (A1, B1), next step target point is (A2, B2), the position of dynamic barrier at this time is (C1, D1)；WhenBarrier is indicated in the range of unmanned boat safety moving, need at this time into
Row is handled based on particle swarm algorithm collision prevention；
Inertia weight maximum value is set as ωmax, minimum value ωmin, Studying factors are respectively C1, C2, group D, greatest iteration
Number is Dmax, using current point as origin, ρ is that radius makees polar coordinates, carries out n equal portions, and m particle generates the particle of m × n at random
Group and position X and speed V, are arranged current optimal locationForm initial population t0；Make Pbesti
Represent the optimal value that i-th of particle search arrives, GbestiThe optimal value that entire group search arrives is represented, the suitable of initial population is calculated
Angle value is answered, to global optimum GbestiIt updates, wherein fitness function includes path length and degree of safety, shortest path length
Objective function:
In above formula, (xji, yji) be path j on i point coordinate, (xji-1, yji-1) it is (x on the j of pathji, yji) coordinate,
At this point, next coordinate (x of unmanned boat path planning predictionjn, yjn), it needs this point and target point (g1, g2) connect
To calculate path, entire path length function are as follows:
Y=Yfit1+Yfit2
Degree of safety function:
In above formula, xkFor the position coordinates of barrier；
In summary, fitness function: F=A × Y+B × OB (i), wherein A, the weighted factor of two function of B, for more than or equal to zero
Any real number, for complete collision prevention, B < A, according to speed, position iterative formula updates, and generates new population t1, particle
Group velocity iterative formula are as follows:
VI, j(t+1)=ω VI, j(t)+c1r1(PI, j(t)-xI, j(t))+c2r2(PG, j(t)-xI, j(t))
ωk+1=ωmin+(ωk-ωmin)×((Kmax-k)/Kmax)n
In above formula, r1, r2Random real number greater than 0 less than 1, PI, jIt is the optimal location that particle i is searched out so far, PG, iIt is complete
Office's optimal location, ωkFor the resulting value of current iteration, initial value ωmax, k expression current iteration number；KmaxRepresent maximum
The number of iterations；
Calculate new population t1Fitness value, if better than generating new population t if previous generation2, it is otherwise converted into rectangular co-ordinate, is recycled
The new population of grey iterative generation, until finding optimal collision prevention path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910185335.8A CN109933067B (en) | 2019-03-12 | 2019-03-12 | Unmanned ship collision avoidance method based on genetic algorithm and particle swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910185335.8A CN109933067B (en) | 2019-03-12 | 2019-03-12 | Unmanned ship collision avoidance method based on genetic algorithm and particle swarm algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109933067A true CN109933067A (en) | 2019-06-25 |
| CN109933067B CN109933067B (en) | 2022-07-15 |
Family
ID=66987032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910185335.8A Active CN109933067B (en) | 2019-03-12 | 2019-03-12 | Unmanned ship collision avoidance method based on genetic algorithm and particle swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109933067B (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110703752A (en) * | 2019-10-15 | 2020-01-17 | 哈尔滨工程大学 | Unmanned ship double-layer path planning method based on immune heredity-artificial potential field method |
| CN111813128A (en) * | 2020-07-29 | 2020-10-23 | 浙江北鲲智能科技有限公司 | Unmanned ship autonomous navigation performance evaluation method |
| CN112542064A (en) * | 2019-09-23 | 2021-03-23 | 廖秉军 | Dynamic collision avoidance method for fast moving object and slow moving object |
| CN113419522A (en) * | 2021-05-21 | 2021-09-21 | 北京航天控制仪器研究所 | Simulation method and system for unmanned ship path planning algorithm |
| CN113721620A (en) * | 2021-08-30 | 2021-11-30 | 山东交通学院 | Particle swarm-genetic hybrid algorithm-based vehicle transverse PID control method |
| CN113917929A (en) * | 2021-11-11 | 2022-01-11 | 中国船舶重工集团公司第七一九研究所 | Unmanned ship path optimization method and system based on hybrid particle swarm algorithm |
| CN114169628A (en) * | 2021-12-14 | 2022-03-11 | 西南交通大学 | Shipboard aircraft scheduling optimization method and system based on A-star algorithm and genetic algorithm |
| CN114911240A (en) * | 2022-05-27 | 2022-08-16 | 哈尔滨工程大学 | Unmanned aerial vehicle-assisted unmanned ship dynamic obstacle avoidance path planning method |
| CN115167445A (en) * | 2022-07-28 | 2022-10-11 | 中国矿业大学 | Particle swarm optimization-based maritime intelligent search and rescue method |
| US20220371709A1 (en) * | 2021-05-21 | 2022-11-24 | Wuhan University Of Technology | Path planning system and method for sea-aerial cooperative underwater target tracking |
| CN115420289A (en) * | 2022-08-17 | 2022-12-02 | 中国电子科技集团公司第二十八研究所 | Unmanned ship route planning method based on particle swarm improved artificial potential field method |
| CN116107328A (en) * | 2023-02-09 | 2023-05-12 | 陕西科技大学 | Optimal automatic obstacle avoidance method for ornithopter based on improved genetic algorithm |
| CN116126032A (en) * | 2023-04-17 | 2023-05-16 | 华南农业大学 | Unmanned aerial vehicle group path planning method based on improved multi-objective evolutionary algorithm |
| CN116386389A (en) * | 2023-03-21 | 2023-07-04 | 中国南方航空股份有限公司 | Civil aviation route planning method with limit |
| CN117742323A (en) * | 2023-12-06 | 2024-03-22 | 江苏大学 | Target distribution and route planning method for multi-agent unmanned ship |
| CN117968689A (en) * | 2023-12-25 | 2024-05-03 | 荆楚理工学院 | Unmanned aerial vehicle path planning method and system based on genetic particle mixing algorithm |
| CN115167445B (en) * | 2022-07-28 | 2024-10-29 | 中国矿业大学 | Particle swarm optimization-based offshore intelligent search and rescue method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102768536A (en) * | 2012-07-20 | 2012-11-07 | 哈尔滨工程大学 | Route planning method based on multi-target glowworm swarm algorithm |
| CN106444755A (en) * | 2016-09-22 | 2017-02-22 | 江苏理工学院 | Mobile robot path planning method and system based on improved genetic algorithm |
| CN108564202A (en) * | 2018-03-18 | 2018-09-21 | 哈尔滨工程大学 | A kind of unmanned boat Route optimization method based on environmental forecasting information |
| CN108983789A (en) * | 2018-08-20 | 2018-12-11 | 广东华中科技大学工业技术研究院 | A kind of path planning of unmanned boat and lay dispatching method |
| CN109375625A (en) * | 2018-11-12 | 2019-02-22 | 智慧航海（青岛）科技有限公司 | A kind of intelligent ship paths planning method based on fast search genetic algorithm |
-
2019
- 2019-03-12 CN CN201910185335.8A patent/CN109933067B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102768536A (en) * | 2012-07-20 | 2012-11-07 | 哈尔滨工程大学 | Route planning method based on multi-target glowworm swarm algorithm |
| CN106444755A (en) * | 2016-09-22 | 2017-02-22 | 江苏理工学院 | Mobile robot path planning method and system based on improved genetic algorithm |
| CN108564202A (en) * | 2018-03-18 | 2018-09-21 | 哈尔滨工程大学 | A kind of unmanned boat Route optimization method based on environmental forecasting information |
| CN108983789A (en) * | 2018-08-20 | 2018-12-11 | 广东华中科技大学工业技术研究院 | A kind of path planning of unmanned boat and lay dispatching method |
| CN109375625A (en) * | 2018-11-12 | 2019-02-22 | 智慧航海（青岛）科技有限公司 | A kind of intelligent ship paths planning method based on fast search genetic algorithm |
Non-Patent Citations (4)
| Title |
|---|
| GUANGYU LI等: ""Fractional-Order PID Controller of USV Course-Keeping Using Hybrid GA-PSOAlgorithm"", 《2015 8TH INTERNATIONAL SYMPOSIUM ON COMPUTATIONAL INTELLIGENCE AND DESIGN (ISCID)》 * |
| XIAOGONG LIN等: ""Research of USV obstacle avoidance strategy based on dynamic window"", 《2017 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION (ICMA)》 * |
| 付杨: ""水面无人艇的路径规划及避碰方法研究"", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》 * |
| 杜开君等: ""基于海事规则的水面无人艇动态障碍规避方法"", 《船海工程》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN112542064A (en) * | 2019-09-23 | 2021-03-23 | 廖秉军 | Dynamic collision avoidance method for fast moving object and slow moving object |
| CN112542064B (en) * | 2019-09-23 | 2024-03-26 | 廖秉军 | Dynamic collision prevention method for fast moving object and slow moving object |
| CN110703752A (en) * | 2019-10-15 | 2020-01-17 | 哈尔滨工程大学 | Unmanned ship double-layer path planning method based on immune heredity-artificial potential field method |
| CN110703752B (en) * | 2019-10-15 | 2023-01-03 | 哈尔滨工程大学 | Unmanned ship double-layer path planning method based on immune heredity-artificial potential field method |
| CN111813128A (en) * | 2020-07-29 | 2020-10-23 | 浙江北鲲智能科技有限公司 | Unmanned ship autonomous navigation performance evaluation method |
| CN111813128B (en) * | 2020-07-29 | 2023-01-13 | 浙江北鲲智能科技有限公司 | Unmanned ship autonomous navigation performance evaluation method |
| CN113419522A (en) * | 2021-05-21 | 2021-09-21 | 北京航天控制仪器研究所 | Simulation method and system for unmanned ship path planning algorithm |
| CN113419522B (en) * | 2021-05-21 | 2022-08-12 | 航天时代(青岛)海洋装备科技发展有限公司 | Simulation method and system for unmanned ship path planning algorithm |
| US20220371709A1 (en) * | 2021-05-21 | 2022-11-24 | Wuhan University Of Technology | Path planning system and method for sea-aerial cooperative underwater target tracking |
| CN113721620A (en) * | 2021-08-30 | 2021-11-30 | 山东交通学院 | Particle swarm-genetic hybrid algorithm-based vehicle transverse PID control method |
| CN113917929A (en) * | 2021-11-11 | 2022-01-11 | 中国船舶重工集团公司第七一九研究所 | Unmanned ship path optimization method and system based on hybrid particle swarm algorithm |
| CN114169628A (en) * | 2021-12-14 | 2022-03-11 | 西南交通大学 | Shipboard aircraft scheduling optimization method and system based on A-star algorithm and genetic algorithm |
| CN114169628B (en) * | 2021-12-14 | 2023-04-07 | 西南交通大学 | Shipboard aircraft scheduling optimization method and system based on A-star algorithm and genetic algorithm |
| CN114911240A (en) * | 2022-05-27 | 2022-08-16 | 哈尔滨工程大学 | Unmanned aerial vehicle-assisted unmanned ship dynamic obstacle avoidance path planning method |
| CN115167445A (en) * | 2022-07-28 | 2022-10-11 | 中国矿业大学 | Particle swarm optimization-based maritime intelligent search and rescue method |
| CN115167445B (en) * | 2022-07-28 | 2024-10-29 | 中国矿业大学 | Particle swarm optimization-based offshore intelligent search and rescue method |
| CN115420289A (en) * | 2022-08-17 | 2022-12-02 | 中国电子科技集团公司第二十八研究所 | Unmanned ship route planning method based on particle swarm improved artificial potential field method |
| CN115420289B (en) * | 2022-08-17 | 2024-05-21 | 中国电子科技集团公司第二十八研究所 | Unmanned ship route planning method based on particle swarm improvement artificial potential field method |
| CN116107328A (en) * | 2023-02-09 | 2023-05-12 | 陕西科技大学 | Optimal automatic obstacle avoidance method for ornithopter based on improved genetic algorithm |
| CN116386389A (en) * | 2023-03-21 | 2023-07-04 | 中国南方航空股份有限公司 | Civil aviation route planning method with limit |
| CN116386389B (en) * | 2023-03-21 | 2023-12-29 | 中国南方航空股份有限公司 | Civil aviation route planning method with limit |
| CN116126032A (en) * | 2023-04-17 | 2023-05-16 | 华南农业大学 | Unmanned aerial vehicle group path planning method based on improved multi-objective evolutionary algorithm |
| CN117742323A (en) * | 2023-12-06 | 2024-03-22 | 江苏大学 | Target distribution and route planning method for multi-agent unmanned ship |
| CN117742323B (en) * | 2023-12-06 | 2024-07-12 | 江苏大学 | Target distribution and route planning method for multi-agent unmanned ship |
| CN117968689A (en) * | 2023-12-25 | 2024-05-03 | 荆楚理工学院 | Unmanned aerial vehicle path planning method and system based on genetic particle mixing algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109933067B (en) | 2022-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109933067A (en) | A kind of unmanned boat collision prevention method based on genetic algorithm and particle swarm algorithm | |
| CN112015174B (en) | Multi-AGV motion planning method, device and system | |
| Babinec et al. | VFH* TDT (VFH* with Time Dependent Tree): A new laser rangefinder based obstacle avoidance method designed for environment with non-static obstacles | |
| CN109657863B (en) | Firefly algorithm-based unmanned ship global path dynamic optimization method | |
| US11941553B1 (en) | Methods, electronic devices and storage media for ship route optimization | |
| CN108279692A (en) | A kind of UUV dynamic programming methods based on LSTM-RNN | |
| Wang et al. | Cooperative collision avoidance for unmanned surface vehicles based on improved genetic algorithm | |
| Xu et al. | Heuristic and random search algorithm in optimization of route planning for Robot’s geomagnetic navigation | |
| CN111338350A (en) | Unmanned ship path planning method and system based on greedy mechanism particle swarm algorithm | |
| CN113093733B (en) | Sea-to-sea striking method for unmanned boat cluster | |
| Zheng et al. | A Decision‐Making Method for Ship Collision Avoidance Based on Improved Cultural Particle Swarm | |
| CN110906935A (en) | Unmanned ship path planning method | |
| CN113538973B (en) | Automatic ship collision avoidance method based on improved particle swarm optimization | |
| CN113848974A (en) | Aircraft trajectory planning method and system based on deep reinforcement learning | |
| Tian et al. | A two-level optimization algorithm for path planning of bionic robotic fish in the three-dimensional environment with ocean currents and moving obstacles | |
| Liu et al. | Intelligent path planning for AUVs in dynamic environments: An EDA-based learning fixed height histogram approach | |
| CN113391633A (en) | Urban environment-oriented mobile robot fusion path planning method | |
| CN113985899A (en) | Underwater robot global path planning method based on interval multi-objective optimization | |
| Xiao et al. | Artificial force fields for multi-agent simulations of maritime traffic: a case study of Chinese waterway | |
| Vagale et al. | Evaluation of path planning algorithms of autonomous surface vehicles based on safety and collision risk assessment | |
| Smierzchalski et al. | Path planning in dynamic environments | |
| KR20240080189A (en) | Distance measurement method and distance measurement device using the same | |
| CN114879660A (en) | Robot environment sensing method based on target driving | |
| CN116448134B (en) | Vehicle path planning method and device based on risk field and uncertain analysis | |
| CN112799414B (en) | AUV relaxation trajectory planning method |
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 |