CN115143970B - Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation - Google Patents
Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation Download PDFInfo
- Publication number
- CN115143970B CN115143970B CN202211065443.XA CN202211065443A CN115143970B CN 115143970 B CN115143970 B CN 115143970B CN 202211065443 A CN202211065443 A CN 202211065443A CN 115143970 B CN115143970 B CN 115143970B
- Authority
- CN
- China
- Prior art keywords
- underwater vehicle
- underwater
- motion
- threat degree
- network model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
- G01C21/203—Specially adapted for sailing ships
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
Abstract
The invention provides an obstacle avoidance method and system of an underwater vehicle based on threat degree evaluation, wherein the method comprises the steps of acquiring the speed and the position of the underwater vehicle moving towards all possible directions; determining the optimal position of each moving direction of the underwater vehicle passing through and the optimal position in all the moving directions at the target moment; the optimal position in all the motion directions is the next motion position of the underwater vehicle; updating the speed and the position of each motion direction of the underwater vehicle at each moment; constructing a space path planning model of the underwater vehicle; constructing a threat degree evaluation network model; and (4) carrying out obstacle avoidance on the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle. When uncertain things occur in the process of executing tasks, the underwater vehicle can be prevented from colliding with obstacles or other underwater vehicles, threat assessment on the uncertain things can be avoided, and safety of the underwater vehicle with multiple underwater vehicles and task completion efficiency can be improved.
Description
Technical Field
The invention belongs to the technical field of underwater vehicle control, and particularly relates to an underwater vehicle obstacle avoidance method and system based on threat degree evaluation.
Background
An Autonomous Underwater Vehicle (AUV) is an unmanned carrying platform which has Autonomous navigation and planning capability and can replace human beings to execute Underwater operation tasks, has the unique advantages of good concealment, high maneuverability, flexible use, wide range of motion and the like compared with a manned submersible Vehicle or a cabled unmanned Vehicle, represents the future development direction of the Underwater Vehicle, and is widely applied to various military and civil fields. However, as the application field and application requirements of the AUV are continuously expanded, the task scene and task requirements of the AUV are increasingly complicated.
The AUV path planning refers to planning a path from a starting point to a target point, meeting AUV performance constraints and being optimal under a certain task cost index (such as minimum energy consumption, shortest navigation time and the like) on the premise of safely avoiding various obstacle threats according to known environment information or environment information (such as obstacle information, ocean current information and the like) detected in real time from a sensor (such as a forward looking sonar, a high-frequency radar and the like). The core thought of the problem mainly comprises two parts of environment modeling and optimized searching: firstly, dividing a planning space according to certain rules, thereby modeling a path planning problem into an optimization search problem in a specific space; then, a suitable optimization search algorithm is adopted to find a feasible optimal path under a certain index. Scholars at home and abroad carry out a great deal of research on AUV path planning problems from different angles, and from the angle of planning space modeling, the method mainly comprises a graph-based method, a space decomposition method, a stochastic programming method, a mathematical programming method and an artificial potential field method.
However, most of the AUV path planning techniques are only applicable to the environment without obstacles or with sparse obstacles, and the AUV path planning problem in the complex marine environment has not been solved effectively. The difficulty of the problem is mainly reflected in that the marine environment has complexity, specifically includes a three-dimensional search space, an environment unstructured (for example, including various types of dense underwater obstacles, existence of non-convex regions, and the like), an environment dynamic property (for example, existence of dynamic ocean currents, motion threats, and the like), an environment uncertainty (partial or complete unknown planning space information), and the like, while the existing method has great limitations and is difficult to meet requirements of path optimality, feasibility, real-time property, complex environment constraint, AUV performance constraint, and the like, so that the path planning capability of the AUV in the complex marine environment still has great space improvement.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides an obstacle avoidance method and system of an underwater vehicle based on threat degree evaluation.
In a first aspect, the invention provides an obstacle avoidance method for an underwater vehicle based on threat degree assessment, which comprises the following steps:
acquiring the speed and position of the underwater vehicle moving towards all possible directions;
determining the optimal position of each moving direction of the underwater vehicle at the target moment and the optimal position in all the moving directions; the optimal position in all the motion directions is the next motion position of the underwater vehicle;
updating the speed and the position of each motion direction of the underwater vehicle at each moment;
constructing a space path planning model of the underwater vehicle;
constructing a threat degree evaluation network model;
and (4) carrying out obstacle avoidance on the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle.
Further, the updating the speed and the position of each moving direction of the underwater vehicle at each moment comprises:
and updating the speed and the position of each motion direction of the underwater vehicle at each moment according to the following formula:
wherein the content of the first and second substances,v id (t+ 1) istThe speed of each motion direction of the underwater vehicle at +1 moment;w s is an inertia weight factor used for controlling the speed of the underwater vehicle;v id (t) Is composed oftThe speed of each motion direction of the underwater vehicle at the moment;c 1 an acceleration constant, namely an acceleration weight factor for balancing the extreme value of the motion direction;c 2 an acceleration constant, namely an acceleration weight factor of a global extreme value of the balanced motion direction;r 1 andr 2 are random numbers uniformly distributed between 0 and 1;p id (t) Is composed oftThe optimal position of the underwater vehicle in one motion direction at the moment;x id (t) Is composed oftThe position of the underwater vehicle at the moment;g gd (t) Is composed oftThe optimal position of the underwater vehicle in all motion directions at the moment;x id (t+ 1) ist+1 position of the underwater vehicle;dis a spatial dimension;iis as followsiThe direction of movement possible.
Further, the expression of the inertia weight factor is:
wherein the content of the first and second substances,w max the maximum value of the inertia weight factor is obtained;w min the minimum value of the inertia weight factor is taken;iterNumthe current iteration number is used as the current iteration number;iterNum max is the set maximum number of iterations.
Further, the expressions of the acceleration weight factors of the self extreme value and the global extreme value of the balanced motion direction are respectively as follows:
wherein the content of the first and second substances,c 1min the minimum value of the acceleration weight factor which is the extreme value of the balance motion direction;c 1max the maximum value of the acceleration weight factor which is the extreme value of the balance motion direction;c 2min the minimum value of the acceleration weight factor which is the global extreme value of the balanced motion direction;c 2max the maximum value of the acceleration weight factor is the global extreme value of the balanced motion direction.
Further, the constructing a spatial path planning model of the underwater vehicle comprises:
the method comprises the following steps of constructing an underwater space path planning model of the underwater vehicle, wherein the expression is as follows:
wherein the content of the first and second substances,Mthe final output variable of the underwater vehicle controller, namely the acceleration of the underwater vehicle propeller;gan output data variable for an underwater vehicle controller;R(V' cm ) Selecting a regurator for processing the dynamic barrier, wherein the size of the regurator depends on the angle difference between the motion direction of the underwater vehicle and the movement direction of the dynamic barrier;V' cm is the maximum value of the velocity of the underwater vehicle;V o the moving speed of the dynamic barrier is obtained;d o is a dynamic obstacle motion speed threshold;αis an activating factor;Disthe distance between the underwater vehicle and the dynamic barrier;d 1 is the safe distance between the underwater vehicle and the dynamic barrier.
Further, the constructing a spatial path planning model of the underwater vehicle further comprises:
the expression of the space path planning model of the underwater vehicle on the horizontal plane is established as follows:
wherein the content of the first and second substances,m h outputting data variables for a horizontal plane controller of an underwater vehicle;a l the acceleration of the left propeller of the underwater vehicle;a r the acceleration of the right propeller of the underwater vehicle;M h the final output quantity of the underwater vehicle horizontal plane controller is obtained;g l the output quantity of the horizontal plane left fuzzy controller is obtained;g r the output quantity of the right fuzzy controller of the horizontal plane is obtained;R(u' cm ) A regurator for selecting and processing the horizontal plane dynamic barrier;u' cm is the linear velocity of the underwater vehicleu'The maximum value of (a);R l a left-side regulator for horizontal dynamic obstacles;R r a right calibrator for horizontal plane dynamic barrier; the angle of the motion direction of the underwater vehicle is set asφ R The angle of the direction of motion of the dynamic barrier isφ o ,φ m =φ R -φ o (ii) a When 0 <φ m When the temperature is less than or equal to 30 ℃,R l <R r the underwater vehicle selects left turning to avoid dynamic obstacles; when the temperature is less than or equal to 30 ℃ below zeroφ m When the ratio is less than 0, the reaction solution is,R l >R r the underwater vehicle chooses to turn right to avoid dynamic obstacles.
Further, the constructing a threat degree evaluation network model includes:
parameters defining a threat level assessment network modelθThe prior probability of (2) obeys Dirichlet distribution, and the posterior probability of the threat degree evaluation network model is as follows:
wherein, the first and the second end of the pipe are connected with each other,P(D|G) Evaluating a posterior probability of the network model for the threat level;Iis a firstIA variable;nis the total number of variables;q I is the total number of nodes of the Bayesian network;Jare nodes in a bayesian network;;kis a nodeJA parent node of (a);;r k is a nodeJThe number of parent nodes of;α IJk individual elements that satisfy a condition in the dataset;N IJk number of instances in the dataset that satisfy the condition;
the Bayesian statistical score is as follows:
wherein logP(D|G) Evaluating Bayes statistical scoring of the network model for the threat degree;
suppose thatD=(D 1 ,D 2 ,…,D m ) Is a group of independent and same-distribution underwater environment observation data, and any observation data is subjected toD l Definition ofX l Is composed ofD l The set of all of the missing variables in (a),is prepared from radix GinsengNumber ofθIs estimated at the current time of the current estimation,is based onWill be provided withD=(D 1 ,D 2 ,…,D m ) Repairing the obtained complete underwater observation data; definition ofθBased onThe log-likelihood function of (a) is:
wherein the content of the first and second substances,is a weight; when the temperature is higher than the set temperatureX l When = phi, definePhi is a null set;
to proceed witht 1 Sub-iteration, calculating the expected log-likelihood functionAndto the maximumθIs a value of (a), wherein;
Defining underwater environment observation data samplesX l =x l ,D l The characteristic function of (A) is:
definition ofX I =kAnd isπ(X I )=J,π(X I ) Taking values for the father node;
to obtain
To obtain
Definition of,For data after patchingAll satisfyX I =kAnd isπ(X I )=JThe sum of the weights of the environmental samples; obtaining:
therefore, whenθWhen the following values are taken, the following values are obtained,the maximum is reached, namely:
wherein, the first and the second end of the pipe are connected with each other,is the maximum bayesian network parameter;θ IJk are bayesian network parameters.
In a second aspect, the present invention provides an obstacle avoidance system for an underwater vehicle based on threat level assessment, including:
the acquisition module is used for acquiring the speed and the position of the underwater vehicle moving towards all possible directions;
the position determining module is used for determining the passing optimal position of each motion direction of the underwater vehicle at the target moment and the optimal position in all the motion directions; the optimal position in all the motion directions is the next motion position of the underwater vehicle;
the updating module is used for updating the speed and the position of each motion direction of the underwater vehicle at each moment;
the path planning model construction module is used for constructing a space path planning model of the underwater vehicle;
the threat degree evaluation network model building module is used for building a threat degree evaluation network model;
and the underwater vehicle obstacle avoidance module is used for avoiding obstacles of the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle.
The invention provides an obstacle avoidance method and system of an underwater vehicle based on threat degree evaluation, wherein the method comprises the steps of acquiring the speed and the position of the underwater vehicle moving towards all possible directions; determining the optimal position of each moving direction of the underwater vehicle at the target moment and the optimal position in all the moving directions; the optimal position in all the motion directions is the next motion position of the underwater vehicle; updating the speed and the position of each motion direction of the underwater vehicle at each moment; constructing a space path planning model of the underwater vehicle; constructing a threat degree evaluation network model; and (4) carrying out obstacle avoidance on the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle. By adopting the scheme, when the underwater vehicle has uncertain things in the task execution process, the collision between the underwater vehicle and the barrier or other underwater vehicles can be avoided, the threat assessment on the uncertain things can be avoided, and the safety of the multi-underwater vehicle and the task completion efficiency can be improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained based on these drawings without creative efforts.
Fig. 1 is a flowchart of an obstacle avoidance method for an underwater vehicle based on threat degree assessment according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an underwater vehicle obstacle avoidance system based on threat degree evaluation provided by an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The task allocation of the underwater vehicle is to allocate the task into a plurality of subtasks, then allocate the subtasks to the underwater vehicle, and plan the sequence of executing the subtasks. The work efficiency can be improved through task allocation, and unnecessary time and energy waste are reduced. The mission planning comprises time series planning and space path planning. As shown in fig. 1, an embodiment of the present invention provides an obstacle avoidance method for an underwater vehicle based on threat level assessment, including:
the speed and position of the underwater vehicle moving in all possible directions are acquired 101.
The time series planning is actually a typical Traveling Salesman Problem (TSP), a task target area needing underwater operation is regarded as a city needing traversal in the TSP Problem, and an optimization algorithm is used for finding out a city needing traversalThe bar can complete all tasks and the shortest path. The method realizes time sequence planning of underwater vehicle tasks based on Particle Swarm Optimization (PSO). Suppose that the H directions in which an underwater vehicle may move at a certain position are defined as H particles in the PSO algorithm. The PSO algorithm first randomly initializes H particles in the feasible solution space and velocity space, i.e., determines the velocity and position of the vehicle in all possible directions, where the position is used to characterize the problem solution, under the conditions that the motion of the underwater vehicle allows. By evaluating the objective function for each direction of movement, determiningtThe best position each particle (i.e. each direction of motion) passes at the moment and the best position found in the population of particles.
And 103, updating the speed and the position of each motion direction of the underwater vehicle at each moment.
In the step, the speed and the position of each motion direction of the underwater vehicle at each moment are updated according to the following formulas:
wherein the content of the first and second substances,v id (t+ 1) istThe speed of each motion direction of the underwater vehicle at +1 moment;w s is an inertia weight factor used for controlling the speed of the underwater vehicle;v id (t) Is composed oftThe speed of each motion direction of the underwater vehicle at the moment;c 1 an acceleration constant, namely an acceleration weight factor for balancing the extreme value of the motion direction;c 2 an acceleration constant, namely an acceleration weight factor of a global extreme value of the balanced motion direction;r 1 andr 2 are random numbers uniformly distributed between 0 and 1;p id (t) Is composed oftUnderwater navigation at all timesOptimal position of the device in one direction of motion;x id (t) Is composed oftThe position of the underwater vehicle at the moment;g gd (t) Is composed oftThe optimal position of the underwater vehicle in all motion directions at the moment;x id (t+ 1) istThe position of the underwater vehicle at +1 moment;dis a spatial dimension;iis a firstiThe possible direction of movement.
Optionally, the expression of the inertia weight factor is:
wherein, the first and the second end of the pipe are connected with each other,w max the maximum value of the inertia weight factor is obtained;w min the minimum value of the inertia weight factor is obtained;iterNumthe current iteration number is;iterNum max the method has the advantages that the maximum iteration times are set, the larger inertia weight traversal range can be ensured when the method starts, and the gradually reduced weight also has better local searching capability along with the increase of the iteration times and just accords with the PSO evolution rule. The trend of keeping consistent with the setting of the inertia weight value,r 1 andr 2 are random numbers evenly distributed between 0 and 1.
The expressions of the acceleration weight factors of the self extreme value and the global extreme value of the balanced motion direction are respectively as follows:
wherein the content of the first and second substances,c 1min the minimum value of the acceleration weight factor which is the extreme value of the balance motion direction;c 1max the maximum value of the acceleration weight factor which is the extreme value of the balance motion direction;c 2min the minimum value of the acceleration weight factor which is the global extreme value of the balanced motion direction;c 2max the maximum value of the acceleration weight factor for balancing the global extreme value of the motion direction.c 1 Is marked by the size of the underwater vehicle facing a certain directionThe cognitive ability of the particles is greatly influenced by the memory of the particles;c 2 the size of (2) indicates the social information exchange sharing capability of the particle, and is greatly influenced by group information.
And 104, constructing a space path planning model of the underwater vehicle.
In the step, an expression for constructing an underwater space path planning model of the underwater vehicle is as follows:
wherein the content of the first and second substances,Mthe final output variable of the underwater vehicle controller, namely the acceleration of the underwater vehicle propeller;gan output data variable for an underwater vehicle controller;R(V' cm ) Selecting a regurator for processing the dynamic barrier, wherein the size of the regurator depends on the angle difference between the motion direction of the underwater vehicle and the movement direction of the dynamic barrier;V' cm is the maximum value of the velocity of the underwater vehicle;V o the moving speed of the dynamic barrier is obtained;d o is a dynamic barrier motion speed threshold;αis an activating factor;Disthe distance between the underwater vehicle and the dynamic barrier;d 1 the safe distance between the underwater vehicle and the dynamic barrier. When the dynamic barrier approaches the underwater vehicle and enters the safe distance, namelyDis≤d 1 When the temperature of the water is higher than the set temperature,α=1, otherwiseα=0 is not activated.
The expression for constructing the space path planning model of the underwater vehicle on the horizontal plane is as follows:
wherein the content of the first and second substances,m h is the horizontal plane of an underwater vehicleThe controller outputs a data variable;a l the acceleration of the left propeller of the underwater vehicle;a r the acceleration of the right propeller of the underwater vehicle;M h the final output quantity of the underwater vehicle horizontal plane controller is obtained;g l the output quantity of the left fuzzy controller of the horizontal plane is obtained;g r the output quantity of the right fuzzy controller of the horizontal plane is obtained;R(u' cm ) A regurator for selecting and processing the horizontal plane dynamic barrier;u' cm is the linear velocity of the underwater vehicleu'Maximum value of (d);R l a left-side regulator for horizontal dynamic obstacles;R r a right calibrator for horizontal plane dynamic barrier; the angle of the motion direction of the underwater vehicle is set asφ R The angle of the direction of motion of the dynamic barrier isφ o ,φ m =φ R -φ o (ii) a When 0 <φ m When the temperature is less than or equal to 30 ℃,R l <R r the underwater vehicle selects left turning to avoid dynamic obstacles; when the temperature is less than or equal to 30 ℃ below zeroφ m When the ratio is less than 0, the reaction mixture is,R l >R r the underwater vehicle chooses to turn right to avoid dynamic obstacles.
And 105, constructing a threat degree evaluation network model.
In the step, the threat degree evaluation network model is constructed, the prior probability of the network model is combined, and the maximum posterior probability is selected. Network model for estimating assumed threat degreeGHas a prior probability ofP(G) For a known underwater environment data setDNetwork modelGThe posterior probability of (a) is:
the network model for obtaining the maximum posterior probability is that in the above formulaP(G)P(D|G) Take the maximum value, so logP(G,D)=log(P(G)P(D|G) A bayesian statistical score of the network model is evaluated for threat. The higher the score is, the better the network model is proved to be, and the threat degree of the underwater uncertain event can be more accurately evaluated.
Parameters defining a threat level assessment network modelθThe prior probability of (2) obeys Dirichlet distribution, and the posterior probability of the threat degree evaluation network model is as follows:
wherein the content of the first and second substances,P(D|G) Evaluating a posterior probability of the network model for the threat level;Iis as followsIA variable;nis the total number of variables;q I is the total number of nodes of the Bayesian network;Jare nodes in a bayesian network;;kis a nodeJA parent node of (a);;r k is a nodeJThe number of parent nodes of (2);α IJk individual elements that satisfy a condition in the dataset;N IJk is the number of instances in the dataset that satisfy the condition.
The Bayesian statistical score is as follows:
wherein logP(D|G) And evaluating the Bayesian statistical score of the network model for the threat degree.
The accuracy of the evaluation of different threat degrees of the parameters is different for the same network structure. In order to more accurately evaluate the threat level of the uncertain events, the method optimizes the evaluation network parameters. The idea of the optimization algorithm is that under the condition that the threat degree evaluation network structure and the underwater environment observation data are determined, bayesian reasoning is used for calculating the probability of missing data, an expected sufficient statistical factor is used for completing a missing data set, and the optimal parameters of the current model are re-estimated.
Suppose thatD=(D 1 ,D 2 ,…,D m ) Is a group of independent and same-distribution underwater environment observation data, and any observation data is subjected toD l Definition ofX l Is composed ofD l The set of all of the missing variables in (a),as a parameterθIs estimated at the current time of the current estimation,is based onWill be provided withD=(D 1 ,D 2 ,…,D m ) Repairing the obtained complete underwater observation data; definition ofθBased onThe log-likelihood function of (a) is:
wherein the content of the first and second substances,is a weight; when in useX l When = phi, definePhi is an empty set;is formed byAndDif it is determined thatCan be written asIs called asθBased onDThe expected log-likelihood function of (a).
The network parameter optimization method comprises the following steps: in the first placet 1 In the process of sub-iteration, calculating expected log-likelihood functionAndto the maximumθIs a value of wherein。
Defining underwater environment observation data samplesX l =x l ,D l The characteristic function of (A) is:
definition ofX I =kAnd isπ(X I )=J,π(X I ) And taking values for the father node.
To obtain
To obtain
Definition of,For patched dataAll of them satisfyingX I =kAnd isπ(X I )=JThe sum of the weights of the environmental samples; obtaining:
wherein the content of the first and second substances,is the maximum bayesian network parameter;θ IJk are bayesian network parameters.
And 106, carrying out obstacle avoidance on the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle.
And downloading the tasks to an underwater vehicle system, and executing the tasks by the underwater vehicle according to distribution. At the same time, the underwater vehicle transmits the collected data (including the uncertain events) back to the control system through the sensor system. And the control system carries out threat degree evaluation on the acquired information for uncertain events. And if the uncertain events threaten the safety of the underwater vehicle, the control system makes a decision to control the underwater vehicle to avoid collision. Otherwise, the task is autonomously completed according to the task plan and collision is autonomously avoided.
Based on the same inventive concept, the embodiment of the invention also provides an underwater vehicle obstacle avoidance system based on threat degree evaluation, and as the problem solving principle of the system is similar to that of the underwater vehicle obstacle avoidance method based on threat degree evaluation, the implementation of the system can refer to the implementation of the underwater vehicle obstacle avoidance method based on threat degree evaluation, and repeated parts are not repeated.
An underwater vehicle obstacle avoidance system based on threat level assessment provided by an embodiment of the invention, as shown in fig. 2, includes:
the acquiring module 10 is used for acquiring the speed and the position of the underwater vehicle moving towards all possible directions;
the position determining module 20 is used for determining the optimal position passed by each motion direction of the underwater vehicle at the target moment and the optimal position in all the motion directions; the optimal position in all the motion directions is the next motion position of the underwater vehicle;
the updating module 30 is used for updating the speed and the position of each motion direction of the underwater vehicle at each moment;
a path planning model construction module 40, configured to construct a spatial path planning model of the underwater vehicle;
a threat degree evaluation network model building module 50, configured to build a threat degree evaluation network model;
and the underwater vehicle obstacle avoidance module 60 is used for carrying out obstacle avoidance on the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle.
For more specific working processes of the above modules, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not described herein again.
The invention has been described in detail with reference to specific embodiments and illustrative examples, but the description is not intended to be construed in a limiting sense. Those skilled in the art will appreciate that various equivalent substitutions, modifications or improvements may be made to the technical solution of the present invention and its embodiments without departing from the spirit and scope of the present invention, which fall within the scope of the present invention. The scope of the invention is defined by the appended claims.
Claims (7)
1. An obstacle avoidance method of an underwater vehicle based on threat degree assessment is characterized by comprising the following steps:
acquiring the speed and position of the underwater vehicle moving towards all possible directions;
determining the optimal position of each moving direction of the underwater vehicle passing through and the optimal position in all the moving directions at the target moment; the optimal position in all the motion directions is the next motion position of the underwater vehicle;
updating the speed and the position of each motion direction of the underwater vehicle at each moment;
constructing a space path planning model of the underwater vehicle;
constructing a threat degree evaluation network model, which comprises the following steps:
the prior probability of a parameter theta of the threat degree evaluation network model is defined to obey Dirichlet distribution, and the posterior probability of the threat degree evaluation network model is as follows:
wherein, P (D | G) is the posterior probability of the threat degree evaluation network model; i is the variable of the I; n is the total number of variables; q. q.s I Is the total number of nodes of the Bayesian network; j is a node in the Bayesian network; alpha (alpha) ("alpha") IJ =∑ k α IJk (ii) a k is a parent node of the node J;r k the number of parents of node J; alpha is alpha IJk Individual elements that satisfy a condition in the dataset; n is a radical of hydrogen IJk Number of instances in the dataset that satisfy the condition;
the Bayesian statistical score is as follows:
wherein logP (D | G) is a Bayesian statistical score of the threat degree evaluation network model;
let D = (D) 1 ,D 2 ,…,D m ) A group of independent and identically distributed underwater environment observation data is obtained by carrying out comparison on any one observation data D l Definition of X l Is D l The set of all of the missing variables in (a),for the current estimation of the parameter theta,is based onMixing D = (D) 1 ,D 2 ,…,D m ) Repairing the obtained complete underwater observation data; defining theta based onThe log-likelihood function of (a) is:
wherein, the first and the second end of the pipe are connected with each other,is a weight; when X is l When = phi, definePhi is a null set;
go on to the t 1 Sub-iteration of calculating the expected log-likelihood functionAndto a maximum value of theta, wherein
Defining an underwater environment observation data sample X l =x l ,D l The characteristic function of (A) is:
definition of X I K and pi (X) I )=J,π(X I ) Taking values for the father node;
to obtain
To obtain
Definition of For data after patchingAll of them satisfying X I K and pi (X) I ) Weight sum of environmental samples of = J; obtaining:
wherein, the first and the second end of the pipe are connected with each other,is the maximum bayesian network parameter; theta IJk Is a Bayesian network parameter;
and (4) carrying out obstacle avoidance on the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle.
2. The method for avoiding obstacles for an underwater vehicle based on threat level assessment as claimed in claim 1, wherein the updating of the speed and the position of each motion direction of the underwater vehicle at each moment comprises:
and updating the speed and the position of each motion direction of the underwater vehicle at each moment according to the following formula:
wherein v is id (t + 1) is the speed of each motion direction of the underwater vehicle at the moment of t + 1; w is a s Is an inertia weight factor used for controlling the speed of the underwater vehicle; v. of id (t) the speed of each moving direction of the underwater vehicle at the moment t; c. C 1 An acceleration constant, namely an acceleration weight factor for balancing the extreme value of the motion direction; c. C 2 An acceleration constant, namely an acceleration weight factor of a global extreme value of the balanced motion direction; r is 1 And r 2 Are random numbers uniformly distributed between 0 and 1; p is a radical of formula id (t) the optimal position of the underwater vehicle in one direction of motion at time t; x is the number of id (t) the position of the underwater vehicle at time t; g gd (t) the optimal position of the underwater vehicle in all directions of motion at time t; x is the number of id (t + 1) is the position of the underwater vehicle at the time of t + 1; d is the spatial dimension; i is the ith possible direction of motion.
3. The threat level assessment based underwater vehicle obstacle avoidance method according to claim 2, wherein the expression of the inertia weight factor is as follows:
wherein, w max The maximum value of the inertia weight factor is obtained; w is a min The minimum value of the inertia weight factor is obtained; iterNum is the current number of iterations; iterNum max Is the set maximum number of iterations.
4. The obstacle avoidance method for the underwater vehicle based on the threat degree assessment as claimed in claim 3, wherein the expressions of the acceleration weight factors of the self extreme value and the global extreme value of the equilibrium motion direction are respectively:
wherein, c 1min The minimum value of the acceleration weight factor which is the extreme value of the balance motion direction; c. C 1max The maximum value of the acceleration weight factor which is the extreme value of the balance motion direction; c. C 2min The minimum value of the acceleration weight factor which is the global extreme value of the balanced motion direction; c. C 2max The maximum value of the acceleration weight factor for balancing the global extreme value of the motion direction.
5. The method for obstacle avoidance of an underwater vehicle based on threat degree assessment as claimed in claim 1, wherein the constructing of the space path planning model of the underwater vehicle comprises:
the method comprises the following steps of constructing an underwater space path planning model of the underwater vehicle, wherein the expression is as follows:
wherein M is the final output variable of the underwater vehicle controller, namely the acceleration of the underwater vehicle propeller; g is an output data variable of the underwater vehicle controller; r (V' cm ) Selecting a regurator for processing the dynamic barrier, wherein the size of the regurator depends on the angle difference between the motion direction of the underwater vehicle and the movement direction of the dynamic barrier; v' cm Is the maximum value of the velocity of the underwater vehicle; v o The moving speed of the dynamic barrier is obtained; d is a radical of o Is a dynamic obstacle motion speed threshold; alpha is an activating factor; dis is the distance between the underwater vehicle and the dynamic barrier; d is a radical of 1 The safe distance between the underwater vehicle and the dynamic barrier.
6. The method for obstacle avoidance of an underwater vehicle based on threat level assessment as claimed in claim 5, wherein the constructing a spatial path planning model of the underwater vehicle further comprises:
the expression for constructing the space path planning model of the underwater vehicle on the horizontal plane is as follows:
wherein m is h Outputting data variables for a horizontal plane controller of an underwater vehicle; a is a l The acceleration of the left propeller of the underwater vehicle; a is a r The acceleration of the right propeller of the underwater vehicle is obtained; m h The final output quantity of the underwater vehicle horizontal plane controller is obtained; g l The output quantity of the horizontal plane left fuzzy controller is obtained; g is a radical of formula r The output quantity of the right fuzzy controller of the horizontal plane is obtained; r (u' cm ) A regurator for selecting and processing the horizontal plane dynamic barrier; u' cm Is the maximum value of the linear velocity u' of the underwater vehicle; r is l A left-side regulator for horizontal dynamic obstacles; r r A right regulator for horizontal plane dynamic barrier; the angle of the motion direction of the underwater vehicle is set asThe angle of the direction of motion of the dynamic barrier isWhen in useIn degree, R l <R r The underwater vehicle selects left turning to avoid dynamic obstacles; when the temperature is higher than the set temperatureWhen R is l >R r The underwater vehicle chooses to turn right to avoid dynamic obstacles.
7. An underwater vehicle obstacle avoidance system based on threat level assessment is characterized by comprising:
the acquisition module is used for acquiring the speed and the position of the underwater vehicle moving towards all possible directions;
the position determining module is used for determining the passing optimal position of each motion direction of the underwater vehicle at the target moment and the optimal position in all the motion directions; the optimal position in all the motion directions is the next motion position of the underwater vehicle;
the updating module is used for updating the speed and the position of each motion direction of the underwater vehicle at each moment;
the path planning model construction module is used for constructing a space path planning model of the underwater vehicle;
the threat degree assessment network model building module is used for building a threat degree assessment network model and comprises the following steps:
the prior probability of a parameter theta of the threat degree evaluation network model is defined to obey Dirichlet distribution, and the posterior probability of the threat degree evaluation network model is as follows:
wherein, P (D | G) is the posterior probability of the threat degree evaluation network model; i is the variable I; n is the total number of variables; q. q.s I Is the total number of nodes of the Bayesian network; j is a node in the Bayesian network; alpha is alpha IJ =∑ k α IJk (ii) a k is a parent node of the node J;r k the number of parents of node J; alpha is alpha IJk Individual elements that satisfy a condition in the dataset; n is a radical of hydrogen IJk Number of instances in the dataset that satisfy the condition;
the Bayesian statistical score is as follows:
wherein logP (D | G) is a Bayesian statistical score of the threat degree evaluation network model;
let D = (D) 1 ,D 2 ,…,D m ) A group of independent and identically distributed underwater environment observation data is obtained by carrying out comparison on any one observation data D l Definition of X l Is D l The set of all of the missing variables in (a),for the current estimation of the parameter theta,is based onD = (D) 1 ,D 2 ,…,D m ) Repairing the obtained complete underwater observation data; defining θ based onThe log likelihood function of (d) is:
wherein the content of the first and second substances,is a weight; when X is l = Φ, definePhi is a null set;
to carry out the t 1 Sub-iteration of calculating the expected log-likelihood functionAndto a maximum value of theta, wherein
Defining underwater environment observation data sample X l =x l ,D l The characteristic function of (A) is:
definition of X I K and pi (X) I )=J,π(X I ) Taking values for the father node;
to obtain
To obtain
Definition of For patched dataAll of them satisfying X I K and pi (X) I ) = the weighted sum of the environmental samples of J; obtaining:
wherein the content of the first and second substances,is a maximum Bayesian network parameterCounting; theta.theta. IJk Is a Bayesian network parameter;
and the underwater vehicle obstacle avoidance module is used for avoiding obstacles of the underwater vehicle according to the space path planning model and the threat degree evaluation network model of the underwater vehicle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211065443.XA CN115143970B (en) | 2022-09-01 | 2022-09-01 | Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211065443.XA CN115143970B (en) | 2022-09-01 | 2022-09-01 | Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115143970A CN115143970A (en) | 2022-10-04 |
CN115143970B true CN115143970B (en) | 2022-11-29 |
Family
ID=83416207
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211065443.XA Active CN115143970B (en) | 2022-09-01 | 2022-09-01 | Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115143970B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117606490A (en) * | 2024-01-23 | 2024-02-27 | 吉林大学 | Collaborative search path planning method for autonomous underwater vehicle |
CN117826826A (en) * | 2024-03-04 | 2024-04-05 | 广东电网有限责任公司阳江供电局 | Autonomous obstacle avoidance method and system for underwater robot |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20160048530A (en) * | 2014-10-24 | 2016-05-04 | 국방과학연구소 | Method and apparatus for generating pathe of autonomous vehicle |
CN109471446A (en) * | 2018-11-08 | 2019-03-15 | 淮阴师范学院 | A kind of complete coverage path planning method neural network based |
CN109711087A (en) * | 2019-01-14 | 2019-05-03 | 哈尔滨工程大学 | A kind of UUV dynamic threats method for situation assessment |
CN113010963A (en) * | 2021-03-04 | 2021-06-22 | 山东大学 | Variable-quality underwater vehicle obstacle avoidance method and system based on deep reinforcement learning |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107966153B (en) * | 2017-11-24 | 2020-02-18 | 中国海洋大学 | Underwater vehicle path planning algorithm |
CN110782481B (en) * | 2019-10-18 | 2023-09-05 | 华中光电技术研究所(中国船舶重工集团有限公司第七一七研究所) | Unmanned ship intelligent decision-making method and system |
CN112733273A (en) * | 2021-01-14 | 2021-04-30 | 齐齐哈尔大学 | Method for determining Bayesian network parameters based on genetic algorithm and maximum likelihood estimation |
CN114266977B (en) * | 2021-12-27 | 2023-04-07 | 青岛澎湃海洋探索技术有限公司 | Multi-AUV underwater target identification method based on super-resolution selectable network |
-
2022
- 2022-09-01 CN CN202211065443.XA patent/CN115143970B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20160048530A (en) * | 2014-10-24 | 2016-05-04 | 국방과학연구소 | Method and apparatus for generating pathe of autonomous vehicle |
CN109471446A (en) * | 2018-11-08 | 2019-03-15 | 淮阴师范学院 | A kind of complete coverage path planning method neural network based |
CN109711087A (en) * | 2019-01-14 | 2019-05-03 | 哈尔滨工程大学 | A kind of UUV dynamic threats method for situation assessment |
CN113010963A (en) * | 2021-03-04 | 2021-06-22 | 山东大学 | Variable-quality underwater vehicle obstacle avoidance method and system based on deep reinforcement learning |
Non-Patent Citations (1)
Title |
---|
Deep Reinforcement Learning Based Optimal Trajectory Tracking Control of Autonomous Underwater Vehicle;Runsheng Yu 等;《IEEE》;20170726;第4958-4965页 * |
Also Published As
Publication number | Publication date |
---|---|
CN115143970A (en) | 2022-10-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Hu et al. | Voronoi-based multi-robot autonomous exploration in unknown environments via deep reinforcement learning | |
CN115143970B (en) | Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation | |
Zhu et al. | Complete coverage path planning of autonomous underwater vehicle based on GBNN algorithm | |
Wang et al. | A multilayer path planner for a USV under complex marine environments | |
CN106845716B (en) | Navigation error constraint-based water surface unmanned ship local hierarchical path planning method | |
CN108319293B (en) | UUV real-time collision avoidance planning method based on LSTM network | |
MahmoudZadeh et al. | Uninterrupted path planning system for Multi-USV sampling mission in a cluttered ocean environment | |
CN110926477A (en) | Unmanned aerial vehicle route planning and obstacle avoidance method | |
MahmoudZadeh et al. | Hybrid motion planning task allocation model for AUV’s safe maneuvering in a realistic ocean environment | |
CN110764518B (en) | Underwater dredging robot path planning method and device, robot and storage medium | |
Yao et al. | A hierarchical architecture using biased min-consensus for USV path planning | |
Ni et al. | An improved DSA-based approach for multi-AUV cooperative search | |
Guo et al. | A fusion method of local path planning for mobile robots based on LSTM neural network and reinforcement learning | |
Yao et al. | Lstm-pso: Long short-term memory ship motion prediction based on particle swarm optimization | |
Huang | A novel three-dimensional path planning method for fixed-wing UAV using improved particle swarm optimization algorithm | |
Li et al. | Three-dimensional path planning for AUVs in ocean currents environment based on an improved compression factor particle swarm optimization algorithm | |
Yao et al. | Multi-USV cooperative path planning by window update based self-organizing map and spectral clustering | |
Ramezani et al. | UAV path planning employing MPC-reinforcement learning method considering collision avoidance | |
Zhang et al. | Intelligent vector field histogram based collision avoidance method for auv | |
Wang et al. | Research on sbmpc algorithm for path planning of rescue and detection robot | |
CN115933637A (en) | Path planning method and device for substation equipment inspection robot and storage medium | |
CN113110530B (en) | Underwater robot path planning method for three-dimensional environment | |
Zhi-Wen et al. | Path planning for UUV in dynamic environment | |
Wang et al. | Multi-AUG three-dimensional path planning and secure cooperative path following under DoS attacks | |
Ma et al. | Distributed multiagent control approach for multitarget tracking |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |