CN110032184A - A kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling - Google Patents
A kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling Download PDFInfo
- Publication number
- CN110032184A CN110032184A CN201910185329.2A CN201910185329A CN110032184A CN 110032184 A CN110032184 A CN 110032184A CN 201910185329 A CN201910185329 A CN 201910185329A CN 110032184 A CN110032184 A CN 110032184A
- Authority
- CN
- China
- Prior art keywords
- unmanned boat
- waterborne
- path
- dynamic barrier
- dynamic
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000002265 prevention Effects 0.000 title claims abstract description 20
- 238000005070 sampling Methods 0.000 title claims abstract description 9
- 230000004888 barrier function Effects 0.000 claims abstract description 103
- 230000002068 genetic effect Effects 0.000 claims abstract description 13
- 230000033001 locomotion Effects 0.000 claims description 21
- 210000000349 chromosome Anatomy 0.000 claims description 12
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 9
- 239000003643 water by type Substances 0.000 claims description 6
- 230000007613 environmental effect Effects 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 230000002093 peripheral effect Effects 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 238000010353 genetic engineering Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000035699 permeability Effects 0.000 claims description 3
- 108090000623 proteins and genes Proteins 0.000 claims description 3
- 230000009471 action Effects 0.000 claims description 2
- 230000008439 repair process Effects 0.000 claims description 2
- 230000000977 initiatory effect Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- LFULEKSKNZEWOE-UHFFFAOYSA-N propanil Chemical compound CCC(=O)NC1=CC=C(Cl)C(Cl)=C1 LFULEKSKNZEWOE-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/0206—Control of position or course in two dimensions specially adapted to water vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Traffic Control Systems (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 waterborne based on heredity with slip window sampling.Step 1: unmanned boat waterborne carries out global active path planning, and step 2: unmanned boat waterborne analyzes dynamic barrier, step 3: unmanned boat waterborne carries out sliding window algorithm avoidance, step 4: unmanned boat waterborne continues traveling and repeats above step；Relative to traditional path planning, genetic algorithm and sliding window algorithm are combined the path planning to form dynamic collision prevention by the present invention, and path planning is more effective, quick, stable.
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 heredity and sliding window
The unmanned boat collision prevention method waterborne of mouth method.
Background technique
The path planning collision prevention method of unmanned boat (USV) waterborne is that unmanned boat autonomous control waterborne seeks in optimal path to move
The key point of state avoidance research.
Since marine environment is complicated and changeable, while the moment also being wanted to consider that marine stormy waves stream does unmanned boat waterborne
The effect of disturbing.So unmanned boat waterborne before entering waters, first can cook up one according to known related Ocean environment information
The path of the item overall situation, into after waters, unmanned boat waterborne is navigated by water according to the route planned.Due in advance can not
Predict during navigation it is possible that various situations, such as ship, the drifting objects of burst accident, therefore, nothing waterborne
The situation that people's ship will also detect surrounding in real time during navigation obtains the information of ambient enviroment, for these feelings that happen suddenly
Condition, unmanned boat waterborne should be able to quickly perceive, and make for different situations and quickly and accurately coping with, and adjustment is certainly
The operational configuration of body, finally evades out obstructing objects, and specified target point is reached according to mission requirements.
Summary of the invention
The unmanned boat collision prevention method waterborne based on heredity with slip window sampling that the purpose of the present invention is to provide a kind of, with reality
Existing unmanned boat waterborne dynamic collision prevention on optimum path planning, solves unmanned boat waterborne dynamic barrier on path planning
Avoidance problem.
The embodiment of the present invention provides a kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling, comprising:
Step 1: unmanned boat waterborne carries out global active path planning: after into waters according to the path of advance planning before
Row does not in advance consider there is no burst dynamic barrier or obstruction, then unmanned boat waterborne is pressed if ambient enviroment is all gone well
Continue navigation according to original path and constantly detect peripheral information, until reaching target point；If unmanned boat waterborne is in water-area navigation
During detect dynamic barrier, then it is moved by for information about dynamic predict, then plan new route
Avoiding obstacles eventually arrive at target point；
Step 2: unmanned boat waterborne analyzes dynamic barrier: if unmanned boat waterborne is during water-area navigation
Dynamic barrier is detected, by waterborne in stroke after dynamic window module and the judgement of detection device marine radar and AIS
The motion conditions of unmanned boat and dynamic barrier obtain the solution for different situations:
If (a) dynamic barrier traffic direction is opposite with unmanned boat waterborne navigation or deviates unmanned boat course line waterborne,
It can not crash, unmanned boat waterborne can continue to move ahead according to original motion state, enter step four；
If (b) unmanned boat waterborne is in next T0Period unmanned boat waterborne can enter the risk of collision area of dynamic barrier,
Then unmanned boat waterborne should carry out immediately evading for dynamic barrier, enter step three；
If (c) between unmanned boat and dynamic barrier waterborne the case where is quasi- dangerous situation, unmanned boat waterborne is at this
Between initiate in section to evade action to dynamic barrier, enter step three；
If (d) between unmanned boat and dynamic barrier waterborne the case where is to enter step four there are potential danger situation；
Step 3: unmanned boat waterborne carries out sliding window algorithm avoidance: according to the motion information of the dynamic barrier of acquisition
It is analyzed, is calculated by genetic algorithm and sliding window algorithm, obtain dynamic barrier evasion tactics, and then adjust waterborne
Unmanned boat displacement posture, evades dynamic barrier；
Step 4: unmanned boat waterborne continues traveling and repeats above step: unmanned boat waterborne is needed according to task to target point
It moves ahead, while reentering step 1；
The step 1, comprising:
Unmanned boat waterborne carries out global active path planning: moving ahead after into waters according to the path of advance planning, if all
Collarette border is all gone well, and does not in advance consider there is no burst dynamic barrier or obstruction, then unmanned boat waterborne is according to original road
Diameter continues to navigate by water and constantly detect peripheral information, until reaching target point；If unmanned boat waterborne is during water-area navigation
It detects dynamic barrier, then moves dynamic to it by for information about and predict, then plan that new route avoids obstacle
Object eventually arrives at target point；
It is wherein, described to carry out global active path planning method particularly includes:
Different path solutions is evaluated in the space encoder of itself by fitness function according to GA algorithm, is copied
Cross and variation process during living nature population genetic is iterated update by multiple and different path solutions, obtains optimal road
Diameter；Each chromosome i (i=1,2,3...n) will represent the one of unmanned boat waterborne in the initialization population of unmanned boat on the water
The path of item, as follows:
(xi1,yi1)→(xi2,yi2)→···(xili,yili)
Wherein, (xi1,yi1) and (xili,yili) indicate starting point on unmanned boat i-th (i=1,2,3...n) paths waterborne
And terminal, (xij,yij) (j=2 ..., li) it is then the coordinate of some point other than two-end-point on path, length
For variable li, the space entirely solved is traversed using the coding mode of genetic algorithm, each path is divided into multistage, it is right respectively
The path difference section is calculated, and obtains path-related information, and the division length value of any paths, k are indicated with deltaδ
It indicates positive scale parameter value, by following formula, obtains the total d for the section that the paths are divided intonum:
In above formula, (xmax,ymax) and (xmin,ymin) be artificial window coordinate range, s (xs,ys) and e (xe,ye) it is water
The initial position and target point of upper unmanned boat；
Assuming that the i-th paths are expressed as P by a total of n item of initial original pathiIf path Pi3 sections are divided into, i.e.,
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 two points
It is arbitrarily generated a little in rangeThis abscissa is limited inIn range, and ordinate is confined to
Path initial position s (xs,ys) and target point e (xe,ye) between, i.e.,Then sequentially by pi(2j-1),pi(2j) is connected, and ultimately generates path Pi:
PiIndicate i-th (i=1,2,3...n) a feasible motion path in population, then the node of passage path itself
Coordinate encodes the paths accordingly；
The feasible path collection of unmanned boat waterborne isWherein the i-th paths are expressed as Pi=[pi1,
pi2,···pili] (i=1,2,3, n), pij(I=1,2, li) it is the path paths traversed point, pi1
For the initial point in path, piliFor the terminal in path:
Value(P*)=min [f1(P),f2(P),f3(P),f4(P)]
In above formula, f1It (P) is the safety of planning path, f2It (P) is the feasibility of planning path, f3It (P) is planning path
Flatness, f4It (P) is the economy of planning path；
The length in path:
The path length of i-th chromosome of unmanned boat waterborne:
In above formula, miIt is PiInfeasible path number, C in path1For a suitable positive number；
The slickness in path:
Work as liWhen > 2, i.e., the number that the path of chromosome divides gene loci is greater than 2, the P of unmanned boat waterborneiPath is average
Turning value:
In above formula, aij(j=2, liIt -1) is pi(j-1)pijWith pijpi(j+1)Between angle (0≤aij≤π),mi
And kiIt is aijIn be not less than pi/2 number, i.e., if some turning be not less than pi/2 when, punishment meter is carried out to target value
It calculates, C2It is a suitable positive number；
Work as liWhen=2, path PiFor the line of initial point to target point:
Turning(Pi)=mi×C2
Safety:
If path PiIt is feasible, danger (di)=1/di, wherein di> 0 indicates the navigation route of unmanned boat waterborne
Minimum value apart from static-obstacle thing；If path PiIt is infeasible, danger (Pi)=mi × C3, miFor path individual
Route segment and the distance between barrier be less than the quantity of safe distance, C3For a positive number appropriate.
According to the value of fitness function, follow-on chromosome is selected by ＂ roulette ＂ method and intersected, make a variation, repair
Multiple genetic manipulation optimizes population, obtains the feasible path that can evade static-obstacle thing；
The step 2, comprising:
Unmanned boat waterborne analyzes dynamic barrier: if unmanned boat waterborne detected during water-area navigation it is dynamic
State barrier, by after dynamic window module and the judgement of detection device marine radar and AIS in stroke unmanned boat waterborne and
The motion conditions of dynamic barrier obtain the solution for different situations；
It is wherein, described that dynamic barrier analyzed method particularly includes:
The speed of a ship or plane of unmanned boat waterborne is VUSV, the speed of dynamic barrier is VObz, polar axis and VUSVBetween angle α=∠
(VUSV,ex), exFor polar axis, polar axis and VObzBetween angle β=∠ (VObs,ex), polar axis and unmanned boat waterborne and dynamic barrier
Angle theta=∠ (L between circle center line connectingRO,ex), Δ V and VUVSBetween angleμ=∠ (LRO, tanL),
TanL is dynamic barrier circle tangent line, γ=∠ (Δ V, LRO) and in (LRO-μ,LRO+ μ) corresponding value is chosen outside region；
Δ v is resolved into the component Δ v along dynamic barrier center of circle direction0With vertically-oriented component Δ vr, wherein Δ v0
Unmanned boat waterborne is set to advance towards the direction of barrier, Δ vrMake unmanned boat waterborne far from barrier；
In above formula:
Assuming that the speed v of dynamic barrier within a short period of timeObsAnd direction β variation is less, can not consider, i.e. dvObs
β=0=0, d, then have:
By vUSV、vObs, Δ v composition speed triangle can be obtained there are following formula relationship:
Wherein Φ is vUSV、vObsBetween angle；
In order to allow unmanned boat waterborne can be far from (LRO-μ,LRO+ μ) danger zone, Δ γ must satisfy following condition limit
System:
Above formula shows the variation delta v of Δ γ Yu unmanned boat speed waterborneUSVAnd the variation delta α of the direction of motion exists
Certain connection, adjusts the course Δ α and speed of a ship or plane Δ v of unmanned boat waterborneUSV, then respectively correspond two kinds of unmanned boat waterborne and effectively keep away
Barrier behavior, it is possible to use only one such collision prevention behavior completion dynamic barrier evades task；
Evading for dynamic barrier is carried out about navigation direction is changed, the value of evacuation amplitude, ao C is solved by following formula,
By being obtained to the meet research of situation of unmanned boat waterborne and dynamic barrier:
In above formula, RTFor unmanned boat waterborne between dynamic barrier at a distance from, θTFor the relative bearing of dynamic barrier, β is
The corner of relative movement line,For the speed of a ship or plane ratio of dynamic barrier and unmanned boat waterborne, Δ C is the steering angle of evacuation；
Steering angle Δ C is acquired by iterative method:
The step 3, comprising:
Unmanned boat waterborne carries out sliding window algorithm avoidance: being divided according to the motion information of the dynamic barrier of acquisition
Analysis is calculated by genetic algorithm and sliding window algorithm, obtains dynamic barrier evasion tactics, and then adjust unmanned boat waterborne
Displacement posture, evades dynamic barrier；
It is wherein, described to carry out sliding window algorithm avoidance method particularly includes:
Ambient ocean environmental information generates sub-objective P after the completion of updating in each sliding windowwin(t), pass through inspiration
Formula function minf (P)=g (P)+h (P) determines that sub-goal, g (P) are the generation that unmanned boat waterborne is run to P position by current point
Valence, and the position of P point and current environmental information determine g (P) value, h (P) is cost of the P position to terminating point, it
Value is obtained by evaluating the spacing of P position to final position, sub-goal Pwin(t) window of minimum function value is used
Boundary point P:
Through P whether in the range of feasible zone determining g (P), it may be assumed that
In above formula, FD indicates that the free space of clear, NFD indicate the space of barrier
The select permeability of sub-goal is simplified to optimization problem:
It is P that letter, which is regarded as and seeks the P for the shortest path length for reaching target point,win(t), if Pwin(t) not unique, then arbitrarily select one
It is a；
The radius R of sliding window is determining by following formula, window monitoring range of the unmanned boat waterborne at the position P (t):
Win (P (t))=P | P ∈ W, d (P, P (t))≤R }
If the route speed of unmanned boat waterborne is V, period T, specific item punctuate is Pwin(t), then it is transported in a period
Dynamic step-length is ε=T × Δ V, and wherein Δ V is unmanned boat waterborne average speed within the time period and 0 < ε < R, nothing waterborne
Area of feasible solutions within the scope of people's ship sliding window can indicate are as follows:
In above formula, SOnumRepresent static-obstacle thing number, DOnumRepresent known dynamic barrier number, DOUnumIt represents
Unknown dynamic barrier number；
The beneficial effects of the present invention are:
1. invention introduces genetic algorithm and sliding window algorithm, for global path planning and local path avoidance
Unmanned boat waterborne can satisfy optimal path planning requirement, to avoid unmanned boat waterborne on Global motion planning path
Local obstacle avoidance；
2. the present invention can make fast reaction to the real-time dynamic environment information of the unmanned boat waterborne of acquisition；
3. the present invention introduced genetic algorithm and sliding window algorithm, two kinds of algorithms are combined to form dynamic collision prevention
Path planning, compared to traditional path planning, path planning is more effective, quick, stable；
4. the present invention is while focusing on control effect in view of the real navigation situation and efficiency of navigation of unmanned boat.
Detailed description of the invention
Fig. 1 is that the present invention calculates Obstacle avoidance model figure；
Fig. 2 is that the present invention calculates relative velocity exploded view；
Fig. 3 is that present invention unmanned boat waterborne evades dynamic barrier flow chart；
Fig. 4 is sliding window path planning algorithm flow chart of the present invention；
Fig. 5 (a) is that dynamic barrier of the present invention evades analogous diagram；
Fig. 5 (b) is that dynamic barrier of the present invention evades analogous diagram；
Fig. 5 (c) is that dynamic barrier of the present invention evades analogous diagram；
Fig. 5 (d) is that dynamic barrier of the present invention evades analogous diagram；
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, the global active path planning of water surface unmanned boat
GA mainly passes through fitness function and evaluates in the space encoder of itself different path solutions, by imitative
According to the cross and variation process during living nature population genetic, how different path solutions is iterated update, finally obtains one
Meet the optimal path of mission requirements.On the water each chromosome i in the initialization population of unmanned boat (USV) (i=1,2,
3...n one path of unmanned boat waterborne (USV) will) be represented.It is as follows:
(xi1,yi1)→(xi2,yi2)→···(xili,yili)
Wherein, (xi1,yi1) and (xili,yili) indicate on unmanned boat (USV) i-th (i=1,2..., n) paths waterborne
Beginning and end, (xij,yij) (j=2 ..., li) it is then the coordinate of some point other than two-end-point on path,
Length is variable li, the space entirely solved is traversed using the coding mode of this genetic algorithm.Each path can be by it
It is divided into multistage part, calculating analysis is carried out to the paths difference section part respectively, obtains the relevant information of the paths.With
Delta indicates the length value of the division section of any paths, kδThe scale parameter value positive as one can be found out by following formula
The total d for the section that the paths are divided intonum。
(xmax,ymax) and (xmin,ymin) be artificial window coordinate range, s (xs,ys) and e (xe,ye) be it is waterborne nobody
The initial position and target point of ship (USV).
Assuming that the i-th paths are expressed as pi by a total of n item of initial original path, if path pi is divided into 3 sections,
That is 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 two points
In the range of arbitrarily generate a littleThis abscissa is limited inIn range, and ordinate limits to
In 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) is connected, and ultimately generates path Pi:
Pi indicates a feasible motion path of i-th (i=1,2 .., n) in population.Then the node of passage path itself is sat
Mark to encode the paths accordingly.
The feasible path collection of unmanned boat (USV) waterborne isWherein, the i-th paths are expressed as Pi=
[pi1,pi2,···pili] (i=1,2,3, n), pij(I=1,2, li) it is the path paths traversed
Point, pi1For the initial point in path, piliFor the terminal in path.
Value(P*)=min [f1(P),f2(P),f3(P),f4(P)]
f1It (P) is the safety of planning path, f2It (P) is the feasibility of planning path, f3It (P) is the smooth of planning path
Property, f4It (P) is the economy of planning path.
The length in path:
The path length of i-th chromosome of unmanned boat (USV) waterborne:
miIt is PiInfeasible path number, C in path1For a suitable positive number.
The slickness in path:
As li > 2, i.e., the number that the path of chromosome divides gene loci is greater than 2, the road Pi of unmanned boat (USV) waterborne
Diameter average corner value:
Wherein, aij(j=2, liIt -1) is pi(j-1)pijWith pijpi(j+1)Between angle (0≤aij≤π),miAnd ki
It is aijIn be not less than pi/2 number, i.e., if some turning be not less than pi/2 when, punishment calculating is carried out to target value,
C2It is a suitable positive number, works as liWhen=2, path P i is line of the initial point to target point, Turning (Pi)=mi×C2
Safety:
If path PiIt is feasible, danger (di)=1/di, wherein di> 0 indicates the navigation of unmanned boat waterborne (USV)
Minimum value of the route apart from static-obstacle thing；If path PiIt is infeasible, danger (Pi)=mi×C3, miFor the path
The distance between route segment and barrier of individual are less than the quantity of safe distance, C3It is then a positive number appropriate.
According to the value of fitness function, selected by ＂ roulette ＂ method follow-on chromosome intersected, made a variation,
The genetic manipulations such as reparation, to optimize population, output can evade the feasible path of static-obstacle thing.
2, analysis of the unmanned boat waterborne to dynamic barrier
Follow unmanned boat sail collision regulation waterborne carry out barrier evade can by change navigation direction and
Change the mode of movement velocity to realize.
The speed of a ship or plane that attached drawing 1 illustrates unmanned boat waterborne (USV) is VUSV, the speed of dynamic mobile object is VObz.Polar axis with
VUSVBetween angle α=∠ (VUSV,ex), exFor polar axis；Polar axis and VObzBetween angle β=∠ (VObs,ex)；Polar axis and unmanned boat
Angle theta=∠ (L between dynamic mobile object circle center line connectingRO,ex), Δ V and VUVSBetween angleμ=
∠(LRO, tanL), tanL is dynamic mobile object circle tangent line.γ=∠ (Δ V, LRO) and will be in (LRO-μ,LRO+ μ) outside region
To choose corresponding value.
Attached drawing 2 illustrates the component Δ v resolved into Δ v along dynamic mobile object center of circle direction0And point of vertical direction
Measure Δ vr.Wherein Δ v0So that unmanned boat (USV) waterborne is advanced towards the direction of barrier, promote Δ vrMake unmanned boat waterborne
(USV) far from barrier.
In formula:
Assuming that within the shorter time dynamic mobile object speed vObsAnd direction β variation is less, can not consider, i.e.,
dvObsβ=0=0, d, then have:
By vUSV、vObs, Δ v composition speed triangle, there are the relationships of following formula, carry it into above formula, can be obtained:
Wherein Φ is vUSV、vObsBetween angle.
In order to allow unmanned boat waterborne can be far from (LRO-μ,LRO+ μ) danger zone, Δ γ must satisfy following condition limit
System:
Show the variation delta v of Δ γ Yu unmanned boat waterborne (USV) speedUSVAnd the variation delta α of the direction of motion is deposited
Just become the Solve problems of the two values to unmanned boat waterborne (USV) in the problem of certain connection, such collision prevention.It adjusts
The course Δ α and speed of a ship or plane Δ v of whole unmanned boat waterborne (USV)USV, then two kinds of unmanned boat waterborne (USV) has been respectively corresponded effectively
Avoid-obstacle behavior, it is possible to use only one such collision prevention behavior can complete dynamic and evade task in barrier.
Evading for barrier is carried out about navigation direction is changed, can solve evacuation amplitude, ao C's by following formula
Value.By being obtained to the meet research of situation of unmanned boat waterborne (USV) and dynamic mobile object:
R in formulaTFor unmanned boat waterborne between dynamic barrier at a distance from；θTFor the relative bearing of dynamic barrier；β is phase
To the corner of the line of motion；For the speed of a ship or plane ratio of dynamic barrier and unmanned boat waterborne；Δ C is the angle of evacuation.
During unmanned boat waterborne evades barrier by changing navigation direction, main side in need of consideration
There are DCPA, R in faceT、θT,k.Steering angle Δ C is acquired by iterative method:
3, unmanned boat waterborne uses sliding window algorithm
During being evaded using sliding window module progress dynamic barrier, the ambient ocean in each sliding window
Environmental information generates sub-objective P after the completion of updatingwin(t), it is determined by heuristic function minf (P)=g (P)+h (P)
Sub-goal out, g (P) is the cost that unmanned boat waterborne is run to P position by current point, and the position of P point and current environment
Information determines the certain situation of g (P) value, and h (P) is cost of the P position to terminating point, its value is then by evaluating P position
To final position spacing and obtain.Sub-goal Pwin(t) the window edge point P of minimum function value is used:
In order to make that simplification is calculated, we decrease in the requirement to optimization, and carry out weight to the algorithm of feasibility
Point ground research, it is only necessary to which whether g (P) in the range of feasible zone is determined by P, it may be assumed that
FD indicates that the free space of clear, NFD indicate the space of barrier in formula.
The select permeability of sub-goal has just been simplified to following optimization problem:
It is P that letter, which is regarded as and seeks the P for the shortest path length for reaching target point,win(t), if Pwin(t) not unique, then it can at will select
One out.
The radius R problem of sliding window can be determined by following research.Unmanned boat waterborne is at the position P (t)
Window monitoring range:
Win (P (t))=P | P ∈ W, d (P, P (t))≤R }
If the route speed of unmanned boat waterborne is V, period T, specific item punctuate is PwinIt (t), then can be a time
The step-length moved in section is write as ε=T × Δ V (Δ V is unmanned boat waterborne average speed within the time period and 0 < ε < R).
Area of feasible solutions within the scope of unmanned boat sliding window waterborne can indicate are as follows:
Wherein, SOnumRepresent static-obstacle thing number, DOnumRepresent known dynamic barrier number, DOUnumIt represents not
The dynamic barrier number known.
4, unmanned boat normally travel waterborne
Unmanned boat waterborne continues a period of time unmanned boat waterborne that moves ahead to target point according to the needs of task, reenters the
1 step.
Claims (4)
1. a kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling characterized by comprising
Step 1: unmanned boat waterborne carries out global active path planning: it moves ahead after into waters according to the path of advance planning, if
Ambient enviroment is all gone well, and does not in advance consider there is no burst dynamic barrier or obstruction, then unmanned boat waterborne is according to original
Path continues to navigate by water and constantly detect peripheral information, until reaching target point；If unmanned boat waterborne is in the process of water-area navigation
In detect dynamic barrier, then it is moved by for information about dynamic predict, then plan new route avoid hindering
Hinder object, eventually arrives at target point；
Step 2: unmanned boat waterborne analyzes dynamic barrier: if unmanned boat waterborne detects during water-area navigation
To dynamic barrier, by stroke after dynamic window module and the judgement of detection device marine radar and AIS it is waterborne nobody
The motion conditions of ship and dynamic barrier obtain the solution for different situations:
If (a) dynamic barrier traffic direction is opposite with unmanned boat waterborne navigation or deviates unmanned boat course line waterborne, can not
It can crash, unmanned boat waterborne can continue to move ahead according to original motion state, enter step four；
If (b) unmanned boat waterborne is in next T0Period unmanned boat waterborne can enter the risk of collision area of dynamic barrier, then waterborne
Unmanned boat should carry out immediately evading for dynamic barrier, enter step three；
If (c) between unmanned boat and dynamic barrier waterborne the case where is quasi- dangerous situation, unmanned boat waterborne is in the period
Interior initiation evades action to dynamic barrier, enters step three；
If (d) between unmanned boat and dynamic barrier waterborne the case where is to enter step four there are potential danger situation；
Step 3: unmanned boat waterborne carries out sliding window algorithm avoidance: being carried out according to the motion information of the dynamic barrier of acquisition
Analysis is calculated by genetic algorithm and sliding window algorithm, obtains dynamic barrier evasion tactics, and then adjust unmanned boat waterborne
Displacement posture, evades dynamic barrier；
Step 4: unmanned boat waterborne continues traveling and repeats above step: unmanned boat waterborne needs to move ahead to target point according to task,
Step 1 is reentered simultaneously.
2. a kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling according to claim 1, feature
It is, the step 1, comprising:
Unmanned boat waterborne carries out global active path planning: moving ahead after into waters according to the path of advance planning, if surrounding ring
Border is all gone well, and does not in advance consider there is no burst dynamic barrier or obstruction, then unmanned boat waterborne according to original path after
Continuation of the journey row simultaneously constantly detects peripheral information, until reaching target point；If unmanned boat waterborne detects during water-area navigation
To dynamic barrier, then dynamic is moved to it by for information about and is predicted, then plans new route avoiding obstacles, most
Zhongdao reaches target point；
It is wherein, described to carry out global active path planning method particularly includes:
Different path solutions is evaluated in the space encoder of itself by fitness function according to GA algorithm, copies biology
Cross and variation process during boundary's population genetic is iterated update by multiple and different path solutions, obtains optimal path；In water
Each chromosome i (i=1,2,3...n) will represent one road of unmanned boat waterborne in the initialization population of upper unmanned boat
Diameter, as follows:
(xi1,yi1)→(xi2,yi2)→…(xili,yili)
Wherein, (xi1,yi1) and (xili,yili) indicate starting point and end on unmanned boat i-th (i=1,2,3...n) paths waterborne
Point, (xij,yij) (j=2 ..., li) it is then the coordinate of some point other than two-end-point on path, the length is can
The l of changei, the space entirely solved is traversed using the coding mode of genetic algorithm, each path is divided into multistage, respectively to the path
Different sections are calculated, and obtain path-related information, and the division length value of any paths, k are indicated with deltaδIt indicates just
Scale parameter value the total d for the section that the paths are divided into is obtained by following formulanum:
In above formula, (xmax,ymax) and (xmin,ymin) be artificial window coordinate range, s (xs,ys) and e (xe,ye) it is nothing waterborne
The initial position and target point of people's ship；
Assuming that the i-th paths are expressed as P by a total of n item of initial original pathiIf path PiIt is divided into 3 sections, i.e. 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 range of the two points
Interior any generation pointThis abscissa is limited inIn range, and ordinate is confined to path
Initial position s (xs,ys) and target point e (xe,ye) between, i.e.,Then sequentially by pi(2j-1),pi
(2j) is connected, and ultimately generates path Pi:
PiIndicate i-th (i=1,2,3...n) a feasible motion path in population, then the node coordinate of passage path itself comes
The paths are encoded accordingly；
The feasible path collection of unmanned boat waterborne isWherein the i-th paths are expressed as Pi=[pi1,pi2,…
pili] (i=1,2,3 ..., n), pij(i=1,2 ..., li) it is the path paths traversed point, pi1For the initial point in path,
piliFor the terminal in path:
Value(P*)=min [f1(P),f2(P),f3(P),f4(P)]
In above formula, f1It (P) is the safety of planning path, f2It (P) is the feasibility of planning path, f3It (P) is the flat of planning path
Slip, f4It (P) is the economy of planning path；
The length in path:
The path length of i-th chromosome of unmanned boat waterborne:
In above formula, miIt is PiInfeasible path number, C in path1For a suitable positive number；
The slickness in path:
Work as liWhen > 2, i.e., the number that the path of chromosome divides gene loci is greater than 2, the P of unmanned boat waterborneiPath average corner
Value:
In above formula, aij(j=2 ..., liIt -1) is pi(j-1)pijWith pijpi(j+1)Between angle (0≤aij≤π),miAnd kiIt is aijIn
Not less than the number of pi/2, i.e., if to carry out punishment calculating, C to target value when some turning is not less than pi/22It is a conjunction
Suitable positive number；
Work as liWhen=2, path PiFor the line of initial point to target point:
Turning(Pi)=mi×C2
Safety:
If path PiIt is feasible, danger (di)=1/di, wherein di> 0 indicates the navigation route of unmanned boat waterborne apart from quiet
The minimum value of state barrier；If path PiIt is infeasible, danger (Pi)=mi×C3, miFor the route segment of path individual
The distance between barrier is less than the quantity of safe distance, C3For a positive number appropriate；
According to the value of fitness function, follow-on chromosome is selected by ＂ roulette ＂ method and intersected, make a variation, repair
Genetic manipulation optimizes population, obtains the feasible path that can evade static-obstacle thing.
3. a kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling according to claim 1, feature
It is: the step 2, comprising:
Unmanned boat waterborne analyzes dynamic barrier: if unmanned boat waterborne detects that dynamic hinders during water-area navigation
Hinder object, passes through unmanned boat waterborne in stroke after dynamic window module and the judgement of detection device marine radar and AIS and dynamic
The motion conditions of barrier obtain the solution for different situations；
It is wherein, described that dynamic barrier analyzed method particularly includes:
The speed of a ship or plane of unmanned boat waterborne is VUSV, the speed of dynamic barrier is VObz, polar axis and VUSVBetween angle α=∠ (VUSV,
ex), exFor polar axis, polar axis and VObzBetween angle β=∠ (VObs,ex), polar axis and unmanned boat waterborne and the dynamic barrier center of circle connect
Angle theta=∠ (L between lineRO,ex), Δ V and VUVSBetween angleμ=∠ (LRO, tanL), tanL is
State barrier justifies tangent line, γ=∠ (Δ V, LRO) and in (LRO-μ,LRO+ μ) corresponding value is chosen outside region；
Δ v is resolved into the component Δ v along dynamic barrier center of circle direction0With vertically-oriented component Δ vr, wherein Δ v0Make water
Upper unmanned boat is advanced towards the direction of barrier, Δ vrMake unmanned boat waterborne far from barrier；
In above formula:
Assuming that the speed v of dynamic barrier within a short period of timeObsAnd direction β variation is less, can not consider, i.e. dvObs=0, d
β=0 then has:
By vUSV、vObs, Δ v composition speed triangle can be obtained there are following formula relationship:
Wherein Φ is vUSV、vObsBetween angle；
In order to allow unmanned boat waterborne can be far from (LRO-μ,LRO+ μ) danger zone, Δ γ must satisfy the limitation of following condition:
Above formula shows the variation delta v of Δ γ Yu unmanned boat speed waterborneUSVAnd there are certain by the variation delta α of the direction of motion
Connection, adjusts the course Δ α and speed of a ship or plane Δ v of unmanned boat waterborneUSV, then two kinds of unmanned boat effective avoidance rows waterborne are respectively corresponded
For, it is possible to use only one such collision prevention behavior completion dynamic barrier evades task；
Evading for dynamic barrier is carried out about navigation direction is changed, the value of evacuation amplitude, ao C is solved by following formula, is passed through
The meet research of situation of unmanned boat waterborne and dynamic barrier is obtained:
In above formula, RTFor unmanned boat waterborne 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 ratio of dynamic barrier and unmanned boat waterborne, Δ C is the steering angle of evacuation；
Steering angle Δ C is acquired by iterative method:
4. a kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling according to claim 1, feature
It is: the step 3, comprising:
Unmanned boat waterborne carries out sliding window algorithm avoidance: being analyzed, is led to according to the motion information of the dynamic barrier of acquisition
It crosses genetic algorithm and sliding window algorithm calculates, obtain dynamic barrier evasion tactics, and then adjust unmanned boat itself fortune waterborne
Dynamic posture, evades dynamic barrier；
It is wherein, described to carry out sliding window algorithm avoidance method particularly includes:
Ambient ocean environmental information generates sub-objective P after the completion of updating in each sliding windowwin(t), pass through heuristic letter
Number minf (P)=g (P)+h (P) determines sub-goal, and g (P) is the cost that unmanned boat waterborne is run to P position by current point, and P
The position and current environmental information of point determine g (P) value, and h (P) is cost of the P position to terminating point, its value is then logical
Appraisal P position is crossed to obtain, sub-goal P to the spacing of final positionwin(t) the window edge point P of minimum function value is used:
Through P whether in the range of feasible zone determining g (P), it may be assumed that
In above formula, FD indicates that the free space of clear, NFD indicate the space of barrier
The select permeability of sub-goal is simplified to optimization problem:
It is P that letter, which is regarded as and seeks the P for the shortest path length for reaching target point,win(t), if Pwin(t) not unique, then arbitrarily select one i.e.
It can；
The radius R of sliding window is determining by following formula, window monitoring range of the unmanned boat waterborne at the position P (t):
Win (P (t))=P | P ∈ W, d (P, P (t))≤R }
If the route speed of unmanned boat waterborne is V, period T, specific item punctuate is Pwin(t), then it is moved in a period
Step-length is ε=T × Δ V, and wherein Δ V is unmanned boat waterborne average speed within the time period and 0 < ε < R, and unmanned boat waterborne is sliding
Area of feasible solutions in dynamic window ranges can indicate are as follows:
In above formula, SOnumRepresent static-obstacle thing number, DOnumRepresent known dynamic barrier number, DOUnumIt represents unknown
Dynamic barrier number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910185329.2A CN110032184A (en) | 2019-03-12 | 2019-03-12 | A kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910185329.2A CN110032184A (en) | 2019-03-12 | 2019-03-12 | A kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110032184A true CN110032184A (en) | 2019-07-19 |
Family
ID=67235936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910185329.2A Pending CN110032184A (en) | 2019-03-12 | 2019-03-12 | A kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110032184A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111399506A (en) * | 2020-03-13 | 2020-07-10 | 大连海事大学 | Global-local hybrid unmanned ship path planning method based on dynamic constraints |
| CN111580518A (en) * | 2020-05-12 | 2020-08-25 | 哈尔滨工程大学 | Unmanned ship layered obstacle avoidance method based on improved drosophila optimization and dynamic window method |
| CN112797987A (en) * | 2021-03-23 | 2021-05-14 | 陕西欧卡电子智能科技有限公司 | Navigation method and device for obstacle avoidance of unmanned ship, computer equipment and storage medium |
| CN113110460A (en) * | 2021-04-20 | 2021-07-13 | 哈尔滨工程大学 | Method for acquiring heading feasible interval of unmanned surface vehicle in dynamic environment |
| CN115328122A (en) * | 2022-08-02 | 2022-11-11 | 柳州城市职业学院 | Unmanned ship obstacle avoidance method and system based on multi-radar sensing |
-
2019
- 2019-03-12 CN CN201910185329.2A patent/CN110032184A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 付杨: "水面无人艇的路径规划及避碰方法研究", 《万方数据库在线》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111399506A (en) * | 2020-03-13 | 2020-07-10 | 大连海事大学 | Global-local hybrid unmanned ship path planning method based on dynamic constraints |
| CN111399506B (en) * | 2020-03-13 | 2023-04-25 | 大连海事大学 | Global-local hybrid unmanned ship path planning method based on dynamic constraint |
| CN111580518A (en) * | 2020-05-12 | 2020-08-25 | 哈尔滨工程大学 | Unmanned ship layered obstacle avoidance method based on improved drosophila optimization and dynamic window method |
| CN111580518B (en) * | 2020-05-12 | 2022-08-02 | 哈尔滨工程大学 | Unmanned ship layered obstacle avoidance method based on improved drosophila optimization and dynamic window method |
| CN112797987A (en) * | 2021-03-23 | 2021-05-14 | 陕西欧卡电子智能科技有限公司 | Navigation method and device for obstacle avoidance of unmanned ship, computer equipment and storage medium |
| CN113110460A (en) * | 2021-04-20 | 2021-07-13 | 哈尔滨工程大学 | Method for acquiring heading feasible interval of unmanned surface vehicle in dynamic environment |
| CN113110460B (en) * | 2021-04-20 | 2022-09-02 | 哈尔滨工程大学 | Method for acquiring heading feasible interval of unmanned surface vehicle in dynamic environment |
| CN115328122A (en) * | 2022-08-02 | 2022-11-11 | 柳州城市职业学院 | Unmanned ship obstacle avoidance method and system based on multi-radar sensing |
| CN115328122B (en) * | 2022-08-02 | 2023-03-14 | 柳州城市职业学院 | Unmanned ship obstacle avoidance method and system based on multi-radar sensing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110032184A (en) | A kind of unmanned boat collision prevention method waterborne based on heredity with slip window sampling | |
| Polvara et al. | Obstacle avoidance approaches for autonomous navigation of unmanned surface vehicles | |
| Wang et al. | Roboat: An autonomous surface vehicle for urban waterways | |
| CN107329477B (en) | Unmanned ship navigation and automatic driving equipment and method thereof | |
| Candeloro et al. | A Voronoi-diagram-based dynamic path-planning system for underactuated marine vessels | |
| CN105955268B (en) | A kind of UUV moving-target sliding mode tracking control methods considering Local obstacle avoidance | |
| CN112327885B (en) | Unmanned ship self-adaptive global-local mixed path planning method | |
| Liu et al. | The fast marching method based intelligent navigation of an unmanned surface vehicle | |
| Lazarowska | Ship's trajectory planning for collision avoidance at sea based on ant colony optimisation | |
| CN109933067B (en) | Unmanned ship collision avoidance method based on genetic algorithm and particle swarm algorithm | |
| CN109298708B (en) | Unmanned ship autonomous obstacle avoidance method integrating radar and photoelectric information | |
| CN108803313A (en) | A kind of paths planning method based on ocean current prediction model | |
| Bartolomei et al. | Perception-aware path planning for uavs using semantic segmentation | |
| Perera et al. | Autonomous guidance and navigation based on the COLREGs rules and regulations of collision avoidance | |
| CN113853533A (en) | Yaw rate from radar data | |
| KR102099699B1 (en) | Device and method avoiding collision of autonomous surface vehicle considering uncertainty of trajectory prediction | |
| Lazarowska | Ant colony optimization based navigational decision support system | |
| CN113341980A (en) | Automatic ship collision avoidance and path planning method based on improved artificial potential field method | |
| CN108536140A (en) | A kind of unmanned boat autonomous navigation system and method | |
| Enevoldsen et al. | COLREGs-informed RRT* for collision avoidance of marine crafts | |
| Chen et al. | Tracking with UAV using tangent-plus-Lyapunov vector field guidance | |
| CN104571128A (en) | Obstacle avoidance method used for underwater robot and based on distance and parallax information | |
| Sedova et al. | Automated stationary obstacle avoidance when navigating a marine craft | |
| Fu et al. | Path planning and decision making for autonomous vehicle in urban environment | |
| CN108562289A (en) | Quadrotor laser radar air navigation aid in continuous polygon geometry environment |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190719 |