CN111338356A - Multi-target unmanned ship collision avoidance path planning method for improving distributed genetic algorithm - Google Patents
Multi-target unmanned ship collision avoidance path planning method for improving distributed genetic algorithm Download PDFInfo
- Publication number
- CN111338356A CN111338356A CN202010262916.XA CN202010262916A CN111338356A CN 111338356 A CN111338356 A CN 111338356A CN 202010262916 A CN202010262916 A CN 202010262916A CN 111338356 A CN111338356 A CN 111338356A
- Authority
- CN
- China
- Prior art keywords
- ship
- subgroup
- collision avoidance
- population
- 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.)
- Pending
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 53
- 230000002068 genetic effect Effects 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000008569 process Effects 0.000 claims abstract description 21
- 230000006870 function Effects 0.000 claims description 30
- 230000005012 migration Effects 0.000 claims description 19
- 238000013508 migration Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 6
- QZXCCPZJCKEPSA-UHFFFAOYSA-N chlorfenac Chemical compound OC(=O)CC1=C(Cl)C=CC(Cl)=C1Cl QZXCCPZJCKEPSA-UHFFFAOYSA-N 0.000 claims description 6
- LFULEKSKNZEWOE-UHFFFAOYSA-N propanil Chemical compound CCC(=O)NC1=CC=C(Cl)C(Cl)=C1 LFULEKSKNZEWOE-UHFFFAOYSA-N 0.000 claims description 6
- 238000011160 research Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000010521 absorption reaction Methods 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims 2
- 238000012544 monitoring process Methods 0.000 claims 1
- 206010063385 Intellectualisation Diseases 0.000 abstract description 2
- 230000002265 prevention Effects 0.000 abstract description 2
- 230000036544 posture Effects 0.000 description 11
- 238000011156 evaluation Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 2
- 210000000349 chromosome Anatomy 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010353 genetic engineering Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
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
- G05D1/0088—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
-
- 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)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention relates to a multi-target unmanned ship collision avoidance path planning method for improving a distributed genetic algorithm. The invention comprises the following steps: (1) acquiring the position of the obstacle and the position and the posture of the unmanned ship; (2) calculating the collision risk of the unmanned ship; (3) and planning the collision avoidance path of the multi-target unmanned ship based on the improved distributed genetic algorithm. The invention completes a series of processes of avoiding the obstacles by collecting and analyzing the obstacle information of the ship and making the collision prevention countermeasures, effectively improves the intellectualization of the obstacle avoidance of the ship and reduces the workload of operators.
Description
Technical Field
The invention relates to a multi-target unmanned ship collision avoidance path planning method for improving a distributed genetic algorithm.
Background
On the basis of research on the ship collision avoidance risk and the ship collision avoidance rule, the idea of planning a ship collision avoidance path by using an improved distributed genetic algorithm is proposed by using surface navigation equipment such as a GSP (generalized general purpose processor) and an automatic plotting instrument. The distributed genetic algorithm is a parallel algorithm, has a quick random search capability, is generally accepted by scholars at home and abroad after being proposed, and is firstly converted into a mathematical form when the distributed genetic algorithm is used for planning a collision avoidance line of a ship. Considering that each route is composed of a series of nodes and line segments connecting the nodes, the nodes are actually determined in route optimization, the nodes are converted into genetic factors of a genetic algorithm through coding, constraint conditions of the route are converted into an objective function, and a fitness function for judging the advantages and the disadvantages of the route is determined through the objective function. The invention plans the ship collision avoidance path by using an improved distributed genetic algorithm, and realizes the planning of the collision avoidance path of the multi-target unmanned ship by designing a target function of reference safety and conformity to traffic rules.
Disclosure of Invention
The invention aims to provide a navigation path of an unmanned ship, so that the unmanned ship can effectively realize multi-target unmanned ship collision avoidance path planning, and the invention aims to realize the following steps:
the multi-target unmanned ship collision avoidance path planning method for improving the distributed genetic algorithm comprises the following steps:
(1) acquiring the position of the obstacle and the position and the posture of the unmanned ship;
(2) calculating the collision risk of the unmanned ship;
(3) and planning the collision avoidance path of the multi-target unmanned ship based on the improved distributed genetic algorithm.
The acquisition of barrier position and the acquisition of unmanned ship position gesture, specific process is:
in an unmanned ship intelligent collision avoidance module based on an improved distributed genetic algorithm, an optimal path is planned for the known surrounding marine environment, when the collision avoidance requirement is not met, a ship can navigate in the sea area according to the planned path, and the surrounding marine environment is continuously scanned and monitored to judge whether a collision avoidance target exists. During navigation, if other targets exist on the sea surface, the position and the attitude of the unmanned ship and the position of the obstacle can be respectively measured by using a measuring system on the unmanned ship.
Calculating the collision risk of the unmanned ship, and the specific process is as follows:
in the research of the ship collision risk, the DCPA and the TCPA are two important reasons influencing the ship risk. In order to make the calculation speed faster, the values of the DCPA and the TCPA are used as the input of a BP neural network, and the ship collision risk is used as the network output. And (4) obtaining the connection weight and the threshold of each neuron through learning of expert data, and outputting the ship collision risk.
The multi-target unmanned ship collision avoidance path planning method based on the improved distributed genetic algorithm,
the specific process for improving the distributed genetic algorithm comprises the following steps:
1. subgroup redistribution: in the migration process, fitness calculation and queuing are carried out on the optimal individual of each subgroup, and the space size obtained by the next evolution of the subgroups is in a linear relation with the current fitness thereof:
pithe number of individuals assigned to the next ith subgroup, f (p)i) Is the fitness value of the best individual of the ith subgroup,is the sum of fitness values of all subgroups and p is the individual number of the ith subgroup.
2. Individual migration: and selecting individuals in the sub-population according to the size of the fitness value as a migration object, namely selecting the individuals with high fitness function values. The sub-population with the lower fitness function migrates the local optimal solution of the sub-population to the sub-population with the higher fitness function, but cannot migrate to the sub-population with the lower fitness function. The following formula:
Mpi={pj|f(pi)≤f(pj)}
wherein M ispiA sub-species cluster representing an ith sub-population individual whose fitness value is smaller than itself is acceptable. When the size of the sub-population is reduced, the part of individuals with the minimum fitness is discarded; when the size of the sub-population is increased, in addition to the absorption of the optimum individual from the outside,but also randomly copies itself to fill the space. The sub-population with the highest fitness will not migrate any individual to other sub-populations, whereas the sub-population with the lowest fitness will not have an individual migrated and will also lose a portion of the individuals.
The improved distributed genetic algorithm flow is as follows:
step 1: and (5) encoding.
Step 2: and initializing the population.
And step 3: generating subgroups.
And 4, step 4: and (5) executing a traditional genetic algorithm in the subgroup when the migration period T is not reached, and executing the step 5 after the integral multiple of the migration period T is reached.
And 5: and obtaining the optimal individuals of each subgroup, comparing to obtain the optimal individuals of the subgroup, judging whether the maximum iteration times is reached, outputting the result if the maximum iteration times is reached, otherwise, performing the migration operation, and executing the step 6.
Step 6: and calculating the distribution space of each subgroup according to a subgroup distribution formula, if the distribution space is larger than the original space of the subgroup, comparing and judging whether the space of the subgroup i is larger than that of the subgroup j, if the space of the subgroup i is larger than that of the subgroup j, copying the optimal individual of the subgroup j into the subgroup i, otherwise, copying the optimal individual of the subgroup i into the subgroup j. If the allocated space < the original space of the subgroup, part of the individuals are discarded at will.
And 7: and (4) recombining the subgroups and returning to the step 3.
The multi-target unmanned ship collision avoidance path planning method based on the improved distributed genetic algorithm comprises the following specific steps:
step 1: when the ship navigates according to the optimal path planned by the distributed genetic algorithm before starting, other targets on the sea surface are monitored, and the risk degree of the ship is judged according to the navigation parameters of the target ship.
Step 2: if the risk degree of the target ship is higher than 0.3, entering collision avoidance maneuver; otherwise, the operation is continued along the set route.
And step 3: and judging the meeting posture of the ship and the target ship.
And 4, step 4: and planning a reasonable collision avoidance path according to the information acquired by the system.
And 5: and if all the target risk degrees are less than 0.3, returning to the original path to continue navigation.
Step 6: and after the iteration is finished, outputting an optimal feasible path capable of avoiding the obstacle.
The invention has the following beneficial effects:
1. the invention completes a series of processes of avoiding the obstacles by collecting and analyzing the obstacle information of the ship and making the collision prevention countermeasures, effectively improves the intellectualization of the obstacle avoidance of the ship and reduces the workload of operators.
2. The invention uses the improved distributed genetic algorithm to plan the ship navigation path, overcomes the problem that the traditional genetic algorithm is easy to fall into the local optimal solution, and improves the quality of the planned path.
3. The invention is beneficial to planning a safe and economic route by setting different fitness functions under different meeting postures.
4. The invention plans the path by improving the distributed genetic algorithm, realizes that the unmanned ship can find the obstacles in time when executing tasks such as search and rescue, exploration and the like, implements the collision avoidance strategy according to the ship danger degree and the conditions of the sea collision avoidance rules, and ensures the safety of the ship during navigation.
Drawings
FIG. 1 is a decision making process for collision avoidance for a ship;
FIG. 2 is a flow chart of intelligent collision avoidance for a ship;
FIG. 3 is a flow chart of a conventional genetic algorithm process;
FIG. 4 is a flowchart of an improved distributed algorithm process of the present invention;
FIG. 5 shows the initial positions of the target and home vessels according to the present invention;
fig. 6 shows the movement locus of the target ship and the optimal collision-preventing path of the ship in the invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
In order to make the aforementioned objects, features and advantages of the present invention more comprehensible, the present invention is described in further detail with reference to fig. 1 to 5 and a detailed description thereof, wherein fig. 1 is a ship collision avoidance decision process; FIG. 2 is a flow chart of intelligent collision avoidance for a ship; FIG. 3 is a flow chart of a conventional genetic algorithm process; FIG. 4 is a flowchart of an improved distributed algorithm process of the present invention; FIG. 4 shows the initial positions of the target and home vessels in accordance with the present invention; fig. 5 shows the movement locus of the target ship and the optimal collision-preventing path of the ship in the invention.
The invention relates to a multi-target unmanned ship collision avoidance path planning method based on an improved distributed genetic algorithm. The invention aims to provide an unmanned ship autonomous collision avoidance control method, which improves a distributed genetic algorithm, realizes that an unmanned ship can autonomously search obstacles when executing tasks such as search and rescue, exploration and the like, and implements a collision avoidance strategy according to the distribution condition of the obstacles. 1. The method comprises the steps of obtaining the position of an obstacle and the position and the attitude of an unmanned ship, measuring the position information of the unmanned ship and the obstacle by position reference systems such as a satellite, a tension lock, underwater sound, laser and radar, and measuring the attitude information of a ship such as the heading by an electric compass and a motion reference unit. 2. And calculating the collision risk of the ship by evaluating the navigational speed and the attitude of the ship and the position of the obstacle in the navigation process of the ship. 3. And planning the collision avoidance path of the multi-target unmanned ship based on the improved distributed genetic algorithm. The invention can plan the air route for the unmanned ship to avoid the obstacle when the unmanned ship encounters the obstacle in the sailing process.
The purpose of the invention is realized by the following steps:
1. acquisition of obstacle position and acquisition of ship position attitude
The position information of the unmanned ship and the obstacles is measured by position reference systems such as a satellite, a tension lock, underwater sound, laser, radar and the like, and the attitude information of the unmanned ship such as the heading and the like is measured by an electric compass, a motion reference unit and the like. And filtering and time-space alignment are carried out on the acquired attitude and position information to obtain the accurate position attitude of the unmanned ship and the position of the obstacle.
2. And establishing a collision risk degree model of the unmanned ship, and calculating the collision risk degree of the ship.
3. An objective function is determined.
4. And planning the collision avoidance path of the multi-target unmanned ship based on the improved distributed genetic algorithm.
The specific details include:
1. acquisition of position of obstacle and acquisition of position and posture of unmanned boat
In an unmanned ship intelligent collision avoidance module based on an improved distributed genetic algorithm, an optimal path is planned for the known surrounding marine environment, when the collision avoidance requirement is not met, a ship can navigate in the sea area according to the planned path, and the surrounding marine environment is continuously scanned and monitored to judge whether a collision avoidance target exists. During navigation, if other targets exist on the sea surface, the position and the attitude of the unmanned ship and the position of the obstacle can be respectively measured by using a measuring system on the unmanned ship.
in the formula vOx,vOyThe speed of the unmanned ship on the x axis and the y axis is respectively.
The relative distance between the unmanned boat and the obstacle is calculated according to the geographic coordinates of the unmanned boat and the obstacle as follows:
in the formula, xT,yTIs the horizontal and vertical coordinates of the obstacle; x is the number ofO,yOIs the horizontal and vertical coordinates of the unmanned boat.
The true orientation of the obstacle relative to the unmanned surface is θT:
The true orientation of the unmanned vehicle relative to the obstacle is theta0:
the phase orientation of the obstacle is αT:
The relative velocity components of the obstacle relative to the unmanned vehicle in the x, y axes are:
2. and establishing a collision avoidance risk degree model of the unmanned ship, and calculating a collision risk degree value of the target and the ship.
During navigation, if other targets exist on the sea surface, calculating the ship risk by using target navigation data measured by a measuring system on the unmanned ship.
Nearest meeting distance:
recent times encountered:
in the research of the ship collision risk, the DCPA and the TCPA are two important reasons influencing the ship risk. In order to make the calculation speed faster, the values of the DCPA and the TCPA are used as the input of a BP neural network, and the ship collision risk is used as the network output. And obtaining the connection weight and the threshold of each neuron by learning the expert data.
As the ship is a large inertia model, the ship and the target ship are considered to do uniform linear motion within 1 second, so that the data acquisition period of the system is set to be 1 second. When a plurality of targets appear on the water surface, collision is avoided for the target with the highest risk degree within 1 second according to the collision risk degree between each target and the ship, so that intelligent collision avoidance under multiple targets is realized. The method is characterized in that other targets on the sea surface are detected in real time during navigation of the ship, and when no dangerous target is found, the ship navigates according to a set air route. If there are targets with a risk greater than the risk threshold, i.e., a risk >0.3, the system will enter a collision avoidance state until the risk of all targets is below 0.3. The intelligent ship collision avoidance flow chart is shown in fig. 2, and the whole obstacle avoidance process is as follows:
1. when the unmanned ship navigates according to the planned optimal path before starting, other targets on the sea surface are monitored, and the risk degree of the unmanned ship is judged according to navigation parameters of the target ship.
2. If the risk degree of the target ship is higher than 0.3, entering collision avoidance maneuver; otherwise, the operation is continued along the set route.
3. And judging the meeting posture of the ship and the target ship.
4. And planning a reasonable collision avoidance path according to the information acquired by the system.
5. And if all the target risk degrees are less than 0.3, returning to the original path to continue navigation.
3. Determination of an objective function
According to the marine traffic rule, the system can make different collision avoidance instructions according to the specific meeting posture of the ship and the target ship. That is, different fitness functions are set under different meeting postures. In order to obtain a safe and economical route, the fitness function should consist of two parts.
The first part is a safety evaluation function of the air route, firstly, a target with the maximum risk is found, the shortest distance between the target and the ship is obtained, and the safety of the air route is evaluated according to the preset safety distance. The security evaluation function is as follows:
wherein, x is chromosome to represent a flight path, i is any node on the flight path, the value of i is from 1 to N-1, N is the total number of nodes in a chromosome (flight path), that is to say, each flight path has N-1 segments
When the ship navigates along the collision-prevention path, different meeting situations can be met on different small sections, and when the ship runs to a node, all surrounding targets can be detected, the collision risk degree of each target can be calculated, and the target with the highest collision risk degree with the ship can be found. giThe minimum distance between the target with the maximum collision risk degree with the ship and the ship is shown between the ith node and the (i + 1) th node when the ship runs. d represents the set minimum safe distance. Since the minimum value is found for the target, g is the minimum value in a safe situationiWhen > d, clear (x)i) The value of (A) will be small, in case of insecurity, the opposite is clear (x)i) The value of (a) is large. Where k and h are proportionality coefficients, and k is 0.01 and h is 10.
The second part is that the evaluation function given about the marine traffic rule needs to make corresponding avoidance action according to the meeting posture of the ship and the opposite ship under the general avoidance condition. When the ship navigates in an avoidance path, meeting postures of the ship and the opposite ship are changed at different nodes, so that the algorithm is difficult to realize, and corresponding simplification is performed. Firstly, finding a target with the maximum collision risk degree between a certain node and the ship, judging the meeting posture of the ship and the ship, and judging the steering condition of the ship by analyzing the position condition of the ship from the current node to the next node. The traffic rule evaluation function is as follows:
in the formula:
setting T _ Cost (x)i)∈(0,1]The smaller the value of the function is, the more the sea traffic rule is obeyed by the section of the route. Firstly, the collision risk degrees of each target ship and the ship on each section of track are sequenced, so that the target ship j with the largest collision risk degree with the ship is found, and the conformity of each route to ocean traffic rules is judged according to a meeting pattern formed by the target ship j and the ship.
And (3) adding a proper weight according to the two evaluation functions to form an objective function:
f(x)=0.5S(x)+0.5T(x)
because the minimum problem is to be solved, according to the basic knowledge of the genetic algorithm, the fitness function is determined as follows:
according to the fitness function, when the evolution degree of the population meets the termination condition, the individual with the maximum fitness value in the last generation of individuals is the optimal solution required by the problem.
4. Path planning is carried out by multi-target unmanned ship collision avoidance path planning method based on improved distributed genetic algorithm
The whole population is firstly decomposed into several sub-populations, each sub-population individually executes genetic algorithm, and specific individual migration among the sub-populations is periodically carried out. The improved algorithm can make the evolution of the population jump out of the local optimum and make the search more flexible.
1) Subgroup redistribution: in the migration process, fitness calculation and queuing are carried out on the optimal individual of each subgroup, and the space size obtained by the next evolution of the subgroups is in a linear relation with the current fitness thereof:
pithe number of individuals assigned to the next ith subgroup, f (p)i) Is the fitness value of the best individual of the ith subgroup,is the sum of fitness values of all subgroups and p is the individual number of the ith subgroup.
However, as the individual population subjected to genetic manipulation becomes smaller, the diversity of individuals decreases, and the evolutionary capacity is also weakened. To overcome this problem, distributed genetic algorithm migration strategies were developed.
2) And selecting individuals in the sub-population according to the size of the fitness value as a migration object, namely selecting the individuals with high fitness function values. The sub-population with the lower fitness function migrates the local optimal solution of the sub-population to the sub-population with the higher fitness function, but cannot migrate to the sub-population with the lower fitness function. The following formula:
Mpi={pj|f(pi)≤f(pj)}
wherein M ispiA sub-species cluster representing an ith sub-population individual whose fitness value is smaller than itself is acceptable. When the size of the sub-population is reduced, the part of individuals with the minimum fitness is discarded; when the size of the sub-population is increased, the optimum individual is absorbed from outside
Besides, randomly copying itself to fill the space. The sub-population with the highest fitness will not migrate any individual to other sub-populations, whereas the sub-population with the lowest fitness will not have an individual migrated and will also lose a portion of the individuals.
The improved distributed genetic algorithm flow is as follows:
step 1: and (5) encoding.
Step 2: and initializing the population.
And step 3: generating subgroups.
And 4, step 4: and (5) executing a traditional genetic algorithm in the subgroup when the migration period T is not reached, and executing the step 5 after the integral multiple of the migration period T is reached.
And 5: and obtaining the optimal individuals of each subgroup, comparing to obtain the optimal individuals of the subgroup, judging whether the maximum iteration times is reached, outputting the result if the maximum iteration times is reached, otherwise, performing the migration operation, and executing the step 6.
Step 6: and calculating the distribution space of each subgroup according to a subgroup distribution formula, if the distribution space is larger than the original space of the subgroup, comparing and judging whether the space of the subgroup i is larger than that of the subgroup j, if the space of the subgroup i is larger than that of the subgroup j, copying the optimal individual of the subgroup j into the subgroup i, otherwise, copying the optimal individual of the subgroup i into the subgroup j. If the allocated space < the original space of the subgroup, part of the individuals are discarded at will.
And 7: and (4) recombining the subgroups and returning to the step 3.
On the basis of path planning of a traditional genetic algorithm, the multi-target unmanned ship collision avoidance path planning method based on the improved distributed genetic algorithm comprises the following steps:
step 1: when the ship navigates according to the optimal path planned by the distributed genetic algorithm before starting, other targets on the sea surface are monitored, and the risk degree of the ship is judged according to the navigation parameters of the target ship.
Step 2: if the risk degree of the target ship is higher than 0.3, entering collision avoidance maneuver; otherwise, the operation is continued along the set route.
And step 3: and judging the meeting posture of the ship and the target ship.
And 4, step 4: and planning a reasonable collision avoidance path according to the information acquired by the system.
And 5: and if all the target risk degrees are less than 0.3, returning to the original path to continue navigation.
And 6, outputting an optimal feasible path capable of avoiding the obstacle after iteration is completed.
In a simulation test, three target ships are designed, and three meeting patterns of port crossing, starboard crossing and encounter are simulated respectively. The simulation results are shown in FIG. 6.
Claims (5)
1. The multi-target unmanned ship collision avoidance path planning method for improving the distributed genetic algorithm is characterized by comprising the following steps of:
(1) acquiring the position of the obstacle and the position and the posture of the unmanned ship;
(2) calculating the collision risk of the unmanned ship;
(3) and planning the collision avoidance path of the multi-target unmanned ship based on the improved distributed genetic algorithm.
2. The method for planning the collision avoidance path of the multi-target unmanned ship based on the improved distributed genetic algorithm as claimed in claim 1, wherein the acquiring of the position of the obstacle and the acquiring of the position and the posture of the unmanned ship comprise the following specific processes:
in an unmanned ship intelligent collision avoidance module based on an improved distributed genetic algorithm, firstly, an optimal path is planned for the known surrounding marine environment, when the collision avoidance requirement is not met, a ship navigates in the sea area according to the planned path, and the surrounding marine environment is continuously scanned and monitored to judge whether a collision avoidance target exists; during navigation, if other targets exist on the sea surface, the position and the attitude of the unmanned ship and the position of the obstacle can be respectively measured by using a measuring system on the unmanned ship.
3. The multi-target unmanned ship collision avoidance path planning method based on the improved distributed genetic algorithm as claimed in claim 1, wherein the unmanned ship collision risk is calculated by the following specific process:
in the research of the ship collision risk, the DCPA and the TCPA are two important reasons influencing the ship risk; in order to make the calculation speed faster, the values of the DCPA and the TCPA are used as the input of a BP neural network, and the ship collision risk is used as the network output; and (4) obtaining the connection weight and the threshold of each neuron through learning of expert data, and outputting the ship collision risk.
4. The multi-target unmanned ship collision avoidance path planning method based on the improved distributed genetic algorithm as claimed in claim 1, wherein the specific process of the improved distributed genetic algorithm is as follows:
(1) subgroup redistribution: in the migration process, fitness calculation and queuing are carried out on the optimal individual of each subgroup, and the space size obtained by the next evolution of the subgroups is in a linear relation with the current fitness thereof:
pithe number of individuals assigned to the next ith subgroup, f (p)i) Is the fitness value of the best individual of the ith subgroup,is the sum of fitness values of all subgroups, p is the individual number of the ith subgroup;
(2) individual migration: selecting individuals in the sub-population according to the size of the fitness value as a migration object, namely selecting the individuals with high fitness function values; the sub-population with the lower fitness function migrates the local optimal solution of the sub-population to the sub-population with the higher fitness function, but cannot migrate to the sub-population with the lower fitness function; the following formula:
Mpi={pj|f(pi)≤f(pj)}
wherein M ispiA sub-species cluster representing an ith sub-population individual having a smaller acceptable fitness value than the ith sub-population individual; when the size of the sub-population is reduced, the part of individuals with the minimum fitness is discarded; when the size of the sub-population is increased, in addition to the absorption of the optimum individual from the outside,randomly copying itself to fill the space; the sub-population with the highest fitness does not transfer any individual to other sub-populations, while the sub-population with the lowest fitness does not transfer any individual and loses a part of individuals;
the process is as follows:
step 3.1: coding;
step 3.2: initializing a population;
step 3.3: generating a subgroup;
step 3.4: executing a traditional genetic algorithm in the subgroup when the migration period T is not reached, and executing the step 5 after the integral multiple of the migration period T is reached;
step 3.5: obtaining the optimal individuals of each subgroup, comparing the optimal individuals of the subgroups to obtain the optimal individuals of the population, judging whether the maximum iteration times is reached, if the maximum iteration times is reached, outputting a result, otherwise, performing a migration operation, and executing the step 6;
step 3.6: calculating the distribution space of each subgroup according to a subgroup distribution formula, if the distribution space is larger than the original space of the subgroup, comparing and judging whether the space of the subgroup i is larger than that of the subgroup j, if the space of the subgroup i is larger than that of the subgroup j, copying the optimal individual of the subgroup j into the subgroup i, otherwise, copying the optimal individual of the subgroup i into the subgroup j; if the allocation space is less than the original space of the subgroup, part of individuals are discarded at will;
step 3.7: and (4) recombining the subgroups and returning to the step 3.
5. The multi-target unmanned ship collision avoidance path planning method for the improved distributed genetic algorithm according to claim 1, is characterized in that the multi-target unmanned ship collision avoidance path planning method comprises the following specific steps:
step 4.1: when a ship navigates according to an optimal path planned by a distributed genetic algorithm before starting, monitoring that other targets exist on the sea surface, and judging the risk degree of the ship according to navigation parameters of a target ship;
step 4.2: if the risk degree of the target ship is higher than 0.3, entering collision avoidance maneuver; otherwise, continuing to run along the established route;
step 4.3: judging the meeting posture of the ship and the target ship;
step 4.4: planning a reasonable collision avoidance path according to the information acquired by the system;
step 4.5: if all the target risk degrees are less than 0.3, returning to the original path to continue navigation;
step 4.6: and after the iteration is finished, outputting an optimal feasible path capable of avoiding the obstacle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010262916.XA CN111338356A (en) | 2020-04-07 | 2020-04-07 | Multi-target unmanned ship collision avoidance path planning method for improving distributed genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010262916.XA CN111338356A (en) | 2020-04-07 | 2020-04-07 | Multi-target unmanned ship collision avoidance path planning method for improving distributed genetic algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111338356A true CN111338356A (en) | 2020-06-26 |
Family
ID=71186395
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010262916.XA Pending CN111338356A (en) | 2020-04-07 | 2020-04-07 | Multi-target unmanned ship collision avoidance path planning method for improving distributed genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111338356A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111829527A (en) * | 2020-07-23 | 2020-10-27 | 中国石油大学(华东) | Unmanned ship path planning method based on deep reinforcement learning and considering marine environment elements |
CN113703463A (en) * | 2021-09-23 | 2021-11-26 | 大连海事大学 | Elite multi-population evolution algorithm-based ship collision avoidance path planning method |
CN114387824A (en) * | 2022-01-13 | 2022-04-22 | 上海大学 | Collision-prevention steering judgment method conforming to international maritime collision-prevention rule |
CN115407785A (en) * | 2022-11-01 | 2022-11-29 | 中国船舶集团有限公司第七〇七研究所 | Ship collision avoidance control method, device, equipment and storage medium |
CN115829179A (en) * | 2022-11-30 | 2023-03-21 | 中国人民解放军91977部队 | Ship path planning method and device |
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 |
-
2020
- 2020-04-07 CN CN202010262916.XA patent/CN111338356A/en active Pending
Non-Patent Citations (2)
Title |
---|
袁宇祺: "多目标下的船舶智能避碰方法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
郭云皓: "船舶自动识别***的研究和船舶自动避碰***的实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111829527A (en) * | 2020-07-23 | 2020-10-27 | 中国石油大学(华东) | Unmanned ship path planning method based on deep reinforcement learning and considering marine environment elements |
CN111829527B (en) * | 2020-07-23 | 2021-07-20 | 中国石油大学(华东) | Unmanned ship path planning method based on deep reinforcement learning and considering marine environment elements |
CN113703463A (en) * | 2021-09-23 | 2021-11-26 | 大连海事大学 | Elite multi-population evolution algorithm-based ship collision avoidance path planning method |
CN113703463B (en) * | 2021-09-23 | 2023-12-05 | 大连海事大学 | Ship collision avoidance path planning method based on elite multi-population evolution algorithm |
CN114387824A (en) * | 2022-01-13 | 2022-04-22 | 上海大学 | Collision-prevention steering judgment method conforming to international maritime collision-prevention rule |
CN114387824B (en) * | 2022-01-13 | 2023-06-23 | 上海大学 | Anti-collision steering judging method conforming to international offshore anti-collision rule |
CN115407785A (en) * | 2022-11-01 | 2022-11-29 | 中国船舶集团有限公司第七〇七研究所 | Ship collision avoidance control method, device, equipment and storage medium |
CN115829179A (en) * | 2022-11-30 | 2023-03-21 | 中国人民解放军91977部队 | Ship path planning method and device |
CN115829179B (en) * | 2022-11-30 | 2023-08-01 | 中国人民解放军91977部队 | Ship path planning method and device |
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 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111338356A (en) | Multi-target unmanned ship collision avoidance path planning method for improving distributed genetic algorithm | |
Zhou et al. | The review unmanned surface vehicle path planning: Based on multi-modality constraint | |
CN109933067B (en) | Unmanned ship collision avoidance method based on genetic algorithm and particle swarm algorithm | |
CN109765929B (en) | UUV real-time obstacle avoidance planning method based on improved RNN | |
CN109597417B (en) | Multi-USV group collaborative collision avoidance planning method based on collision avoidance criterion | |
CN109784201B (en) | AUV dynamic obstacle avoidance method based on four-dimensional risk assessment | |
CN107037809A (en) | A kind of unmanned boat collision prevention method based on improvement ant group algorithm | |
CN113156947B (en) | Method for planning path of ship in dynamic environment | |
Wang et al. | Cooperative collision avoidance for unmanned surface vehicles based on improved genetic algorithm | |
CN113032896B (en) | Anti-collision auxiliary decision-making method based on preference of ship driver | |
Li et al. | Distributed multi-objective algorithm for preventing multi-ship collisions at sea | |
Song et al. | Method of emergency collision avoidance for unmanned surface vehicle (USV) based on motion ability database | |
JP2021018484A (en) | Peripheral situation rendering method, navigation avoidance operation learning program, navigation avoidance operation learning system, and marine vessel | |
CN109765890B (en) | Multi-USV group collaborative collision avoidance planning method based on genetic algorithm | |
Sun et al. | A novel fuzzy control algorithm for three-dimensional AUV path planning based on sonar model | |
Sedova et al. | Automated stationary obstacle avoidance when navigating a marine craft | |
Guan et al. | Autonomous collision avoidance of unmanned surface vehicles based on improved A-star and dynamic window approach algorithms | |
Wu et al. | Multi-vessels collision avoidance strategy for autonomous surface vehicles based on genetic algorithm in congested port environment | |
CN116360434A (en) | Ship path planning method based on improved CSAC-APF algorithm | |
CN114625150A (en) | Rapid ant colony unmanned ship dynamic obstacle avoidance method based on danger index and distance function | |
Vagale et al. | Evaluation of path planning algorithms of autonomous surface vehicles based on safety and collision risk assessment | |
CN116700295A (en) | Ship autonomous collision prevention method for complex water area | |
CN112947438B (en) | AUV (autonomous Underwater vehicle) online path planning method based on full-oscillation type invasive weed optimization algorithm | |
Wen et al. | Online heuristically planning for relative optimal paths using a stochastic algorithm for USVs | |
CN113538973A (en) | Automatic ship collision avoidance method based on improved particle swarm optimization |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200626 |