CN111538349A - Long-range AUV autonomous decision-making method oriented to multiple tasks - Google Patents
Long-range AUV autonomous decision-making method oriented to multiple tasks Download PDFInfo
- Publication number
- CN111538349A CN111538349A CN202010304780.4A CN202010304780A CN111538349A CN 111538349 A CN111538349 A CN 111538349A CN 202010304780 A CN202010304780 A CN 202010304780A CN 111538349 A CN111538349 A CN 111538349A
- Authority
- CN
- China
- Prior art keywords
- task
- auv
- task sequence
- time
- long
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000000694 effects Effects 0.000 claims abstract description 12
- 230000007246 mechanism Effects 0.000 claims abstract description 10
- 230000001960 triggered effect Effects 0.000 claims abstract description 6
- 238000011156 evaluation Methods 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 8
- 238000013461 design Methods 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 2
- 230000035484 reaction time Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 12
- 230000008569 process Effects 0.000 abstract description 6
- 238000007726 management method Methods 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 abstract description 3
- 239000002699 waste material Substances 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241001137309 Jordanella floridae Species 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000011160 research Methods 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
- G05D1/10—Simultaneous control of position or course in three dimensions
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06316—Sequencing of tasks or work
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Automation & Control Theory (AREA)
- Tourism & Hospitality (AREA)
- Radar, Positioning & Navigation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Remote Sensing (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Educational Administration (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a long-range AUV autonomous decision-making method for multiple tasks, which mainly solves the problems of task planning and scheduling: (1) the method for representing the index data of the event is introduced through an event triggering mechanism, and the decision calculation is carried out only when the event is triggered, so that the decision calculation is changed from periodicity to aperiodicity, unnecessary optimization calculation is reduced, and the function of responding to the emergency in real time is realized; (2) meanwhile, the effect function is constructed, safety and efficiency are quantified, two targets of safety and efficiency are synchronously optimized, the long-range AUV self-adaptive dynamic planning is realized, the operation efficiency is effectively improved on the premise of ensuring the sailing safety of the long-range AUV, the energy waste in the sailing process is reduced, the optimal task execution sequence is found, and the marine survey task with long duration and complex sailing environment is efficiently completed. The scheme reasonably balances the importance and priority of the tasks in the face of the multi-task autonomous decision problem, improves the autonomy of the aircraft, realizes a strong task management strategy, and has the advantages of strong real-time performance, small calculation amount and the like.
Description
Technical Field
The invention belongs to the field of underwater robot control, is applied to a long-range autonomous underwater vehicle for multiple tasks, and particularly relates to a long-range AUV autonomous decision-making method for multiple tasks.
Background
Due to the dynamic and variable marine environment and unknown complexity, the long-range AUV faces the influence of various factors in the driving process. When highly complex tasks are executed, resources must be managed effectively and rationally and orchestrated so that the AUV can adapt to complex marine environments. Due to the high uncertainty of the marine environment, the current long-range AUV autonomous decision making technology cannot adapt to complex and variable marine environments, and the multi-task-oriented scheduling capability needs to be improved.
The autonomy is a characteristic that an AUV must have for completing various tasks, the existing autonomy decision technology only aims at the AUV of a general voyage, a hierarchical decision method is generally adopted, task coordination and allocation are carried out by exhausting tasks and events, for example, aiming at the task Planning problem of the autonomous decision of the AUV, Zadeh and the like design an AUV autonomous decision framework (Zadeh S M, Powers DM W, Sammut K, et al. persistent AUV Operations Using a Robust Read Missino and Path Planning (RRMPP) Architecture [ J ].2016 ]), the AUV task Planning and time management are completed by combining with dynamic constraints of ocean current data, uncertain obstacles and aircrafts through lower-level motion Planning, the framework is composed of a plurality of modules in a hierarchical form, and each module adopts a series of algorithms to research the efficiency and robustness of the autonomy decision method under different task scenes.
However, in the prior art, when multi-task scheduling is implemented, a periodic sampling mode is generally adopted, and a dynamic programming algorithm is continuously used for decision calculation. In order to make a decision in real time, the sampling period is short, the data acquisition amount in unit time is huge, the decision calculation frequency is high, and the calculation and storage resources of the AUV are limited, so that a large amount of system resources are easily occupied, and the overall data calculation and communication performance of the system is influenced; and the marine environment self-adaptive control system is closely connected with a mother ship, has short working time and good sea conditions of a working sea area, so that the marine environment self-adaptive control system is weaker in self-contained capability, is only suitable for a simple marine environment, cannot respond to unpredictable emergencies in real time, and is difficult to apply to a dynamic complex marine environment.
The long-range AUV has more environmental conditions and difficult problems than the common AUV, so a higher-level autonomous decision system is required to be provided to process the execution problem of multiple tasks under complex conditions.
Disclosure of Invention
Aiming at the defects of the existing autonomous decision making technology, the invention provides a long-range AUV autonomous decision making method facing multiple tasks, mainly solving the problems of task planning and scheduling, not only effectively managing resources of the long-range AUV and reasonably and comprehensively planning multiple tasks, but also being capable of adapting to more complex environment without manual participation and completing the execution of complex tasks.
The invention is realized by adopting the following technical scheme: a long-range AUV autonomous decision-making method for multitasking comprises the following steps:
a, executing a basic task sequence, wherein the basic task sequence comprises a planning task, an obstacle avoidance task, a tracking task, a control task and an emergency task so as to ensure the task execution under the long-range AUV (autonomous underwater vehicle) general state;
step B, expressing the triggering event of the long-range AUV by adopting a data index, and designing a task triggering mechanism to trigger a corresponding task sequence;
step C, constructing an autonomous decision system and performing dynamic self-adaptive planning, wherein the autonomous decision system comprises a control variable mapping module, an AUV state prediction module and a task sequence effect evaluation module, and specifically comprises the following steps;
c1, generating control variables according to AVU states based on the control variable mapping module;
c2, based on AUV state prediction module, calculating the next time AUV state according to the control variable and the current time AUV state;
c3, based on the task sequence effect evaluation module, constructing a utility function to evaluate the control variable and judging whether the current task sequence is the optimal task sequence;
and D, finally, executing the optimal task sequence according to the task sequence triggering result and the evaluation result.
Further, the step B specifically adopts the following manner:
(2) using the index data to represent the task sequence, and periodically sampling the index data, including:
acquiring data of the distance between the aircraft and the obstacle in real time, and taking the distance data as index data to represent an obstacle avoidance task sequence;
collecting the flow speed and direction of the surrounding ocean current, and representing the flow speed and flow direction data as a current avoidance event;
acquiring current and voltage data and other sensor related data of the AUV aircraft, and taking the current and voltage data and other sensor related data as index data to represent a fault event;
(2) design task triggering mechanism
Setting a task sequence trigger threshold to eTThe sampling reference value is dbThe sampled value is dkDefining the sampling reference value as a safety distance;
for the obstacle avoidance task sequence, the sampling value is the real-time distance between the AUV (autonomous Underwater vehicle) and the obstacle, and the trigger error of the task sequence is ek=|dk-dbIf ek<eTIt means that a task sequence event is triggered, which is called the trigger time.
6. The multitask-oriented long-range AUV autonomous decision method according to claim 1, characterized in that: in the step C1, the relationship between the AUV state and the control variable is mapped based on the control variable mapping module:
u(k)=f(x(k))
where x (k) represents the system state at time k, and u (k) represents the control variables generated from the state of the long-range AUV system, including but not limited to system voltage, system current, communication frequency, mission status, speed, heading, altitude, and depth.
Further, in the step C2, calculating the AUV state at the next time according to the control variable and the AUV state at the current time is specifically implemented by:
x(k+1)=F(x(k),u(k),k)
where x (k) is the system state at time k, and x (k) ═ Wk,Ik,Fk,Sk,Vk,Ak,Hk,DkW represents system voltage, I represents system current, F represents communication frequency, S represents mission status, V represents navigational speed, a represents heading, H represents height of AUV from sea floor, D represents depth of AUV from sea surface, u (k) is a control variable, and u (k) is { t }1,j,k,t2,q,k,…tn,l,k0, 1, …, N-1, N is positive integer, i, q, l ∈ [1, m]And m is a positive integer.
Further, in step C3, generating a utility function based on the task sequence effect evaluation module, and evaluating the control variable specifically includes:
1) and outputting an approximate performance index function according to the current AUV state:
wherein U (x (i), U (i)) is a utility function, γ is a discount factor, and 0 < γ ≦ 1;
wherein V (i) is the navigation speed of the aircraft at the i-th moment; t is tsRequired for the aircraft to receive feedback to take actionReaction time set to a constant; DB (i) is the distance between the aircraft and the obstacle at the moment i; MDT is the distance from the task starting point to the task ending point; DT (i) is the distance between the aircraft and the task end point at the moment i;
3) judging the current task sequence
When judging whether the current task sequence is the optimal task sequence, setting an optimal target, namely searching a decision sequence u (i), wherein i is k, k +1, … and N-1, so that a performance index function is minimum;
assume that the optimal performance indicator function value for all AUV states x (k +1) from time k +1 is J*(x (k +1)), and the optimal decision sequence starting from the time k +1 is u*(k+1),u*(k+2),…,u*(N-1), the performance indicator function at time k is expressed as:
U(x(k),u(k))+γJ*(x(k+1));
wherein x (k) is the state at time k, x (k +1) is determined by a system equation, and the optimal performance index function value at time k is expressed as:
the obtained optimal solution u*(k) Comprises the following steps:
i.e. the optimal decision at time k is u*(k),u*(k) The task sequence is the optimal task sequence, so that the optimal task sequence which needs to be executed by the long-range AUV at present is obtained.
Compared with the prior art, the invention has the advantages and positive effects that:
the autonomous decision method provided by the scheme reasonably balances task importance and priority in the face of a multi-task autonomous decision problem, improves the autonomous weight of the aircraft, realizes a strong task management strategy, and has the advantages of strong real-time performance, small calculated amount and the like;
the method for expressing the index data of the event is introduced through an event trigger mechanism, decision calculation is performed only when the event is triggered, so that the decision calculation is changed from periodicity to aperiodicity, unnecessary optimization calculation is reduced, the function of responding to the emergency in real time is realized, a large amount of calculation resources are saved, various tasks can be managed integrally, the autonomous decision level of the long-range AUV is improved practically, and the safety of the AUV is guaranteed while the tasks are completed;
meanwhile, in the long-range AUV task execution process, factors such as ocean currents have great influence on energy consumption of the aircraft, the long-range AUV carries limited resources, the scheme synchronously optimizes two targets of safety and high efficiency by constructing an effect function, so that the long-range AUV self-adaptive dynamic planning is realized, the operation efficiency is effectively improved on the premise of ensuring the navigation safety of the long-range AUV, the energy waste in the navigation process is reduced, an optimal task execution sequence is found, and the marine survey task with long duration and complex navigation environment is efficiently completed.
Drawings
Fig. 1 is a schematic diagram illustrating an autonomous decision method according to an embodiment of the present invention.
Detailed Description
In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be further described with reference to the accompanying drawings and examples. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and thus, the present invention is not limited to the specific embodiments disclosed below.
The implementation provides a long-range Autonomous Underwater Vehicle (AUV) decision-making method for multiple tasks, which mainly comprises the following steps as shown in figure 1:
a, executing a basic task sequence, wherein the basic task sequence comprises a planning task, an obstacle avoidance task, a tracking task, a control task, an emergency task and the like, so as to ensure the task execution under the long-range AUV (autonomous underwater vehicle) general state;
step B, expressing the triggering event of the long-range AUV by adopting a data index, and designing a task triggering mechanism to trigger a corresponding task sequence;
step C, designing an autonomous decision system and carrying out dynamic self-adaptive planning, wherein the autonomous decision system comprises a control variable mapping module, an AUV state prediction module and a task sequence effect evaluation module;
(1) generating control variables according to AVU states based on a control variable mapping module;
(2) based on the AUV state prediction module, calculating the AUV state at the next moment according to the control variable and the AUV state at the current moment;
(3) based on a task sequence effect evaluation module, constructing a utility function to evaluate the control variable and judging whether the current task sequence is the optimal task sequence;
and D, finally, executing the optimal task sequence according to the task sequence triggering result and the evaluation result.
Specifically, in this embodiment, taking a flag fish type 210-AUV of the chinese ocean university as an example of an experimental platform, a detailed description is given to an implementation process of the long-range AUV autonomous decision method:
one, executing basic task sequence
The basic task sequence is a basic guarantee for executing tasks, and tasks such as planning tasks, obstacle avoidance tasks, tracking tasks, control tasks, emergency tasks and the like are started when the tasks start.
In the embodiment, a planning task is set to be in three modes of no planning, traversal planning and end-to-end planning, an obstacle avoidance task is set to be in three modes of no obstacle avoidance, static obstacle avoidance and dynamic obstacle avoidance, a tracking task is set to be in two modes of single-point tracking and multipoint continuous tracking, a control task is set to be in three modes of rudder piece control, propeller control and rudder piece propeller synchronous control, and an emergency task is set to be in three modes of mild emergency, moderate emergency and high risk emergency. Each task pattern represents a specific task execution mode, and each task pattern is numbered as shown in table 1.
TABLE 1 task numbering
Numbering | Task name |
1 | No planning |
2 | Traversal planning |
3 | End-to-end planning |
4 | Without obstacle avoidance |
5 | Static obstacle avoidance |
6 | Dynamic obstacle avoidance |
7 | Without tracking |
8 | Single point tracking |
9 | Multipoint continuous tracking |
10 | Without control |
11 | Rudder blade control |
12 | Propeller control |
13 | Synchronous control of rudder propeller |
14 | Without emergency |
15 | Light emergency |
16 | Moderate emergency |
17 | High-risk emergency |
The experiment is submarine target searching and detecting, basic task sequences are determined to be 2, 5, 8, 13 and 14 according to experiment tasks, and after the AUV is laid, the basic task sequences are executed to ensure that the AUV starts tasks normally.
Expressing the trigger event by adopting a data index, and designing a task trigger mechanism to trigger a corresponding task sequence;
(1) using the index data to represent the task sequence, and periodically sampling the index data:
for example, data of the distance between a vehicle and an obstacle is collected in real time, and the data is used as index data to represent an obstacle avoidance task sequence (also called an event); collecting the flow speed and direction of the surrounding ocean current, and representing the flow speed and flow direction data as a current avoidance event; collecting current and voltage data of the aircraft and relevant data of other sensors, and using the data as index data to represent fault events and the like;
(2) design task triggering mechanism
Setting task sequence triggersA threshold value of eTThe sampling reference value is dbThe sampled value is dkAnd defining the sampling reference value as a safe distance, for example, for an obstacle avoidance task sequence, if the sampling value is the real-time distance between an aircraft and an obstacle, the triggering error of the task sequence is ek=|dk-dbL. If ek<eTIt means that a task sequence event is triggered, which is called the trigger time.
For example, in this embodiment, an altimeter mounted on the AUV is used to collect distance data between the AUV and the obstacle in real time, and an obstacle avoidance decision triggering threshold e is setT2, safety distance db4, distance d between current AUV and obstaclekIf 5, then the event triggers error ek=|dk-db1 due to ek<eTTherefore, an obstacle avoidance event is triggered, and the AUV starts to make autonomous decisions.
A general decision scheme adopts a fixed period decision method, so that a large amount of system computing resources are easily occupied;
three, adaptive dynamic programming
Designing an autonomous decision system and carrying out dynamic self-adaptive planning, wherein the autonomous decision system comprises a control variable mapping module, an AUV (autonomous Underwater vehicle) state prediction module and a task sequence effect evaluation module; in this embodiment, the long-range AUV autonomous decision system is regarded as a discrete nonlinear system, the state variable represents a variable capable of describing a complete system, the system here is a discrete task system and is represented by a set T, where T ═ T1,T2,…,TnAnd each task in the system is a necessary task for enabling the AUV to normally operate, and needs to be executed synchronously, and the relationship among the tasks is a parallel relationship, such as a path planning task, a tracking task, an underlying control task and the like. Each task can be divided into a plurality of subtasks, i.e. the ith task Ti∈{ti1,ti2,…timThat the subtasks cannot be performed synchronously, only at a timeThe method comprises the steps of selecting and executing a subtask, wherein path planning tasks comprise traversal planning, obstacle avoidance planning, flow avoidance planning and the like, and the planning tasks are mutually exclusive relations, namely when one task is executed, other tasks need to be closed.
The dynamic adaptive planning specifically includes:
(1) mapping the relationship between the AUV state and the control variable based on a control variable mapping module:
u(k)=f(x(k))
wherein x (k) represents the system state at time k, u (k) represents control variables, the control variables including but not limited to system voltage, system current, communication frequency, mission status, speed, heading, altitude, and depth are generated based on the state of the long-range AUV system;
the state variable and the control variable are key elements capable of reasonably expressing a decision-making system, and the embodiment extracts key index data from a long-range AUV complex system, constructs the state variable and the control variable and lays a foundation for solving a multi-task-oriented autonomous decision-making problem;
(2) based on the AUV state prediction module, calculating the AUV state at the next moment according to the control variable and the AUV state at the current moment;
x(k+1)=F(x(k),u(k),k)
where x (k) is the system state at time k, and x (k) ═ Wk,Ik,Fk,Sk,Vk,Ak,Hk,DkW represents system voltage, I represents system current, F represents communication frequency, S represents mission status, V represents navigational speed, a represents heading, H represents height of AUV from sea floor, D represents depth of AUV from sea surface, u (k) is a control variable, and u (k) is { t }1,j,k,t2,q,k,…tn,l,k0, 1, …, N-1, N is positive integer, i, q, l ∈ [1, m]And m is a positive integer.
(3) The method comprises the following steps of generating a utility function based on a task sequence effect evaluation module, evaluating a control variable, judging whether a current task sequence is an optimal task sequence, evaluating whether the AUV state is safe or not and whether the task sequence is optimal or not through the utility function, realizing the consideration of two targets of safety and efficiency through the constructed utility function, and performing multi-target optimization on the task sequence, specifically:
1) and outputting an approximate performance index function according to the current AUV state based on the evaluation network module:
wherein U (x (i), U (i)) is a utility function, γ is a discount factor, and 0 < γ ≦ 1;
wherein V (i) is the navigation speed of the aircraft at the i-th moment; t is tsSetting the reaction time required for the aircraft to receive feedback and make an action as a constant; DB (i) is the distance between the aircraft and the obstacle at the moment i; MDT is the distance from the task starting point to the task ending point; DT (i) is the distance between the aircraft and the task end point at the moment i;
4) judging the current task sequence
When judging whether the current task sequence is the optimal task sequence, setting an optimal target, namely searching a decision sequence u (i), wherein i is k, k +1, … and N-1, so that a performance index function is minimum;
assume that the optimal performance indicator function value for all possible AUV states x (k +1) starting from time k +1 is J*(x (k +1)), and the optimal decision sequence starting from the time k +1 is u*(k+1),u*(k+2),…,u*(N-1), the performance indicator function at time k is expressed as:
U(x(k),u(k))+γJ*(x(k+1));
wherein x (k) is the state at time k, x (k +1) is determined by a system equation, and the optimal performance index function value at time k is expressed as:
the obtained optimal solution u*(k) Comprises the following steps:
i.e. the optimal decision at time k is u*(k)。u*(k) The task sequence is the optimal task sequence, so that the optimal task sequence which needs to be executed by the long-range AUV at present is obtained.
D, finally, the autonomous decision system transmits the information of the optimal task sequence to an execution mechanism through a communication element, and starts to execute the optimal task sequence; and at each moment, generating an optimal task sequence according to the flow, and executing each stage of subtasks according to the selected task sequence until the task is completed.
The above description is only a preferred embodiment of the present invention, and not intended to limit the present invention in other forms, and any person skilled in the art may apply the above modifications or changes to the equivalent embodiments with equivalent changes, without departing from the technical spirit of the present invention, and any simple modification, equivalent change and change made to the above embodiments according to the technical spirit of the present invention still belong to the protection scope of the technical spirit of the present invention.
Claims (5)
1. A long-range AUV autonomous decision-making method for multitasking is characterized by comprising the following steps:
a, executing a basic task sequence, wherein the basic task sequence comprises a planning task, an obstacle avoidance task, a tracking task, a control task and an emergency task so as to ensure the task execution under the long-range AUV (autonomous underwater vehicle) general state;
step B, expressing the triggering event of the long-range AUV by adopting a data index, and designing a task triggering mechanism to trigger a corresponding task sequence;
step C, constructing an autonomous decision system and performing dynamic self-adaptive planning, wherein the autonomous decision system comprises a control variable mapping module, an AUV state prediction module and a task sequence effect evaluation module, and specifically comprises the following steps;
c1, generating control variables according to AVU states based on the control variable mapping module;
c2, based on AUV state prediction module, calculating the next time AUV state according to the control variable and the current time AUV state;
c3, based on the task sequence effect evaluation module, constructing a utility function to evaluate the control variable and judging whether the current task sequence is the optimal task sequence;
and D, finally, executing the optimal task sequence according to the task sequence triggering result and the evaluation result.
2. The multitask-oriented long-range AUV autonomous decision method according to claim 1, characterized in that: the step B specifically adopts the following mode:
(1) using the index data to represent the task sequence, and periodically sampling the index data, including:
acquiring data of the distance between the aircraft and the obstacle in real time, and taking the distance data as index data to represent an obstacle avoidance task sequence;
collecting the flow speed and direction of the surrounding ocean current, and representing the flow speed and flow direction data as a current avoidance event;
acquiring current and voltage data and other sensor related data of the AUV aircraft, and taking the current and voltage data and other sensor related data as index data to represent a fault event;
(2) design task triggering mechanism
Setting a task sequence trigger threshold to eTThe sampling reference value is dbThe sampled value is dkDefining the sampling reference value as a safety distance;
for the obstacle avoidance task sequence, the sampling value is the real-time distance between the AUV (autonomous Underwater vehicle) and the obstacle, and the trigger error of the task sequence is ek=|dk-dbIf ek<eTIt means that a task sequence event is triggered, which is called the trigger time.
3. The multitask-oriented long-range AUV autonomous decision method according to claim 1, characterized in that: in the step C1, the relationship between the AUV state and the control variable is mapped based on the control variable mapping module:
u(k)=f(x(k))
where x (k) represents the system state at time k, and u (k) represents the control variables generated from the state of the long-range AUV system, including but not limited to system voltage, system current, communication frequency, mission status, speed, heading, altitude, and depth.
4. The multitask-oriented long-range AUV autonomous decision method according to claim 1, characterized in that: in the step C2, calculating the AUV state at the next time according to the control variable and the AUV state at the current time is specifically implemented by the following method:
x(k+1)=F(x(k),u(k),k)
where x (k) is the system state at time k, and x (k) ═ Wk,Ik,Fk,Sk,Vk,Ak,Hk,DkW represents system voltage, I represents system current, F represents communication frequency, S represents mission status, V represents navigational speed, a represents heading, H represents height of AUV from sea floor, D represents depth of AUV from sea surface, u (k) is a control variable, and u (k) is { t }1,j,k,t2,q,k,…tn,l,k0, 1, …, N-1, N is positive integer, i, q, l ∈ [1, m]And m is a positive integer.
5. The multitask-oriented long-range AUV autonomous decision method according to claim 1, characterized in that: in step C3, a utility function is generated based on the task sequence effect evaluation module, and the evaluation of the control variable specifically includes:
1) and outputting an approximate performance index function according to the current AUV state:
wherein U (x (i), U (i)) is a utility function, γ is a discount factor, and 0 < γ ≦ 1;
wherein V (i) is the navigation speed of the aircraft at the i-th moment; t is tsSetting the reaction time required for the aircraft to receive feedback and make an action as a constant; DB (i) is the distance between the aircraft and the obstacle at the moment i; MDT is the distance from the task starting point to the task ending point; DT (i) is the distance between the aircraft and the task end point at the moment i;
2) judging the current task sequence
When judging whether the current task sequence is the optimal task sequence, setting an optimal target, namely searching a decision sequence u (i), wherein i is k, k +1, … and N-1, so that a performance index function is minimum;
assume that the optimal performance indicator function value for all AUV states x (k +1) from time k +1 is J*(x (k +1)), and the optimal decision sequence starting from the time k +1 is u*(k+1),u*(k+2),…,u*(N-1), the performance indicator function at time k is expressed as:
U(x(k),u(k))+γJ*(x(k+1));
wherein x (k) is the state at time k, x (k +1) is determined by a system equation, and the optimal performance index function value at time k is expressed as:
the obtained optimal solution u*(k) Comprises the following steps:
i.e. the optimal decision at time k is u*(k),u*(k) The task sequence is the optimal task sequence, so that the optimal task sequence which needs to be executed by the long-range AUV at present is obtained.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010304780.4A CN111538349B (en) | 2020-04-17 | 2020-04-17 | Long-range AUV autonomous decision-making method oriented to multiple tasks |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010304780.4A CN111538349B (en) | 2020-04-17 | 2020-04-17 | Long-range AUV autonomous decision-making method oriented to multiple tasks |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111538349A true CN111538349A (en) | 2020-08-14 |
CN111538349B CN111538349B (en) | 2021-03-23 |
Family
ID=71976900
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010304780.4A Active CN111538349B (en) | 2020-04-17 | 2020-04-17 | Long-range AUV autonomous decision-making method oriented to multiple tasks |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111538349B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112862251A (en) * | 2021-01-12 | 2021-05-28 | 西北工业大学 | Distributed multi-AUV task allocation method based on improved PI algorithm |
CN113836781A (en) * | 2021-05-31 | 2021-12-24 | 北京科技大学 | Large-scale robot crowd intelligent cooperative decision-making method oriented to personalized customization mode |
CN117111620A (en) * | 2023-10-23 | 2023-11-24 | 山东省科学院海洋仪器仪表研究所 | Autonomous decision-making method for task allocation of heterogeneous unmanned system |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101551250A (en) * | 2008-04-02 | 2009-10-07 | 南开大学 | Mobile robot landmark dynamic configuration method and device searching facing to unknown environments |
CN101825903A (en) * | 2010-04-29 | 2010-09-08 | 哈尔滨工程大学 | Water surface control method for remotely controlling underwater robot |
WO2013154231A1 (en) * | 2012-04-09 | 2013-10-17 | 주식회사 싸이트로닉 | Method and system for static and dynamic positioning of marine structure by using real-time monitoring of mooring line |
CN103439935A (en) * | 2013-08-15 | 2013-12-11 | 青岛远创机器人自动化有限公司 | Underwater robot control system based on state machine model |
CN105828287A (en) * | 2016-03-11 | 2016-08-03 | 南京航空航天大学 | Reinforcement learning collaborative tracking algorithm (RLTCA) of wireless sensor network |
EP3107806A2 (en) * | 2014-02-19 | 2016-12-28 | CGG Services SA | Method and autonomous underwater vehicle able to maintain a planned arrangement |
CN107844460A (en) * | 2017-07-24 | 2018-03-27 | 哈尔滨工程大学 | A kind of underwater multi-robot based on P MAXQ surrounds and seize method |
CN108489491A (en) * | 2018-02-09 | 2018-09-04 | 上海交通大学 | A kind of Three-dimensional Track Intelligent planning method of autonomous underwater vehicle |
CN108594846A (en) * | 2018-03-23 | 2018-09-28 | 哈尔滨工程大学 | More AUV flight patterns optimal control methods under a kind of obstacle environment |
CN108594638A (en) * | 2018-03-27 | 2018-09-28 | 南京航空航天大学 | The in-orbit reconstructing methods of spacecraft ACS towards the constraint of multitask multi-index optimization |
CN108733825A (en) * | 2018-05-23 | 2018-11-02 | 阿里巴巴集团控股有限公司 | A kind of objects trigger event prediction method and device |
CN110837854A (en) * | 2019-10-30 | 2020-02-25 | 东南大学 | AUV multi-source information fusion method and device based on factor graph |
CN110906936A (en) * | 2019-12-18 | 2020-03-24 | 哈尔滨工程大学 | Underwater robot path planning method |
CN110908404A (en) * | 2019-12-09 | 2020-03-24 | 自然资源部第二海洋研究所 | AUV intelligent observation motion method based on data driving |
-
2020
- 2020-04-17 CN CN202010304780.4A patent/CN111538349B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101551250A (en) * | 2008-04-02 | 2009-10-07 | 南开大学 | Mobile robot landmark dynamic configuration method and device searching facing to unknown environments |
CN101825903A (en) * | 2010-04-29 | 2010-09-08 | 哈尔滨工程大学 | Water surface control method for remotely controlling underwater robot |
WO2013154231A1 (en) * | 2012-04-09 | 2013-10-17 | 주식회사 싸이트로닉 | Method and system for static and dynamic positioning of marine structure by using real-time monitoring of mooring line |
CN103439935A (en) * | 2013-08-15 | 2013-12-11 | 青岛远创机器人自动化有限公司 | Underwater robot control system based on state machine model |
EP3107806A2 (en) * | 2014-02-19 | 2016-12-28 | CGG Services SA | Method and autonomous underwater vehicle able to maintain a planned arrangement |
CN105828287A (en) * | 2016-03-11 | 2016-08-03 | 南京航空航天大学 | Reinforcement learning collaborative tracking algorithm (RLTCA) of wireless sensor network |
CN107844460A (en) * | 2017-07-24 | 2018-03-27 | 哈尔滨工程大学 | A kind of underwater multi-robot based on P MAXQ surrounds and seize method |
CN108489491A (en) * | 2018-02-09 | 2018-09-04 | 上海交通大学 | A kind of Three-dimensional Track Intelligent planning method of autonomous underwater vehicle |
CN108594846A (en) * | 2018-03-23 | 2018-09-28 | 哈尔滨工程大学 | More AUV flight patterns optimal control methods under a kind of obstacle environment |
CN108594638A (en) * | 2018-03-27 | 2018-09-28 | 南京航空航天大学 | The in-orbit reconstructing methods of spacecraft ACS towards the constraint of multitask multi-index optimization |
CN108733825A (en) * | 2018-05-23 | 2018-11-02 | 阿里巴巴集团控股有限公司 | A kind of objects trigger event prediction method and device |
CN110837854A (en) * | 2019-10-30 | 2020-02-25 | 东南大学 | AUV multi-source information fusion method and device based on factor graph |
CN110908404A (en) * | 2019-12-09 | 2020-03-24 | 自然资源部第二海洋研究所 | AUV intelligent observation motion method based on data driving |
CN110906936A (en) * | 2019-12-18 | 2020-03-24 | 哈尔滨工程大学 | Underwater robot path planning method |
Non-Patent Citations (5)
Title |
---|
ADHAM ATYABIA等: "Current advancements on autonomous mission planning and management systems: An AUV and UAV perspective", 《ANNUAL REVIEWS IN CONTROL》 * |
BO HE等: "Autonomous navigation based on unscented-FastSLAM using particle swarm optimization for autonomous underwater vehicles", 《MEASUREMENT》 * |
XIAOKAI MUA等: "End-to-end navigation for Autonomous Underwater Vehicle with Hybrid Recurrent Neural Networks", 《OCEAN ENGINEERING》 * |
张昆玉: "基于优化理论的多AUV协同任务规划方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
薛红香: "水下机器人任务规划方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112862251A (en) * | 2021-01-12 | 2021-05-28 | 西北工业大学 | Distributed multi-AUV task allocation method based on improved PI algorithm |
CN112862251B (en) * | 2021-01-12 | 2024-02-09 | 西北工业大学 | Distributed multi-AUV task allocation method based on improved PI algorithm |
CN113836781A (en) * | 2021-05-31 | 2021-12-24 | 北京科技大学 | Large-scale robot crowd intelligent cooperative decision-making method oriented to personalized customization mode |
CN113836781B (en) * | 2021-05-31 | 2024-04-26 | 北京科技大学 | Large-scale robot crowd intelligent collaborative decision-making method oriented to personalized customization mode |
CN117111620A (en) * | 2023-10-23 | 2023-11-24 | 山东省科学院海洋仪器仪表研究所 | Autonomous decision-making method for task allocation of heterogeneous unmanned system |
CN117111620B (en) * | 2023-10-23 | 2024-03-29 | 山东省科学院海洋仪器仪表研究所 | Autonomous decision-making method for task allocation of heterogeneous unmanned system |
Also Published As
Publication number | Publication date |
---|---|
CN111538349B (en) | 2021-03-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111538349B (en) | Long-range AUV autonomous decision-making method oriented to multiple tasks | |
CN113221449B (en) | Ship track real-time prediction method and system based on optimal strategy learning | |
Xiaofei et al. | Global path planning algorithm based on double DQN for multi-tasks amphibious unmanned surface vehicle | |
CN111339229B (en) | Ship autonomous navigation aid decision-making system | |
CN116588282B (en) | AUV intelligent operation and maintenance system and method | |
Zhang et al. | Multi-AUV adaptive path planning and cooperative sampling for ocean scalar field estimation | |
CN110737267A (en) | Multi-objective optimization method for unmanned ships and intelligent comprehensive management and control system for unmanned ships | |
Bai et al. | Multi-AUV dynamic trajectory optimization and collaborative search combined with task urgency and energy consumption scheduling in 3-D underwater environment with random ocean currents and uncertain obstacles | |
Thompson et al. | Robust mission planning for autonomous marine vehicle fleets | |
Wang et al. | PSO-based method for safe sailing route and efficient speeds decision-support for sea-going ships encountering accidents | |
Yan et al. | An improved multi-AUV patrol path planning method | |
CN116625376A (en) | Unmanned aerial vehicle maritime search and rescue path planning method based on tabu bee colony algorithm | |
Gao et al. | Artificial intelligence algorithms in unmanned surface vessel task assignment and path planning: A survey | |
Al Maawali et al. | Design of intelligent thruster decision-making system for USVs | |
Zhang et al. | Asynchronous multithreading reinforcement control decision method for unmanned surface vessel | |
Yamane et al. | Prediction of ship's speed through ground using the previous voyage's drift speed | |
CN117111620B (en) | Autonomous decision-making method for task allocation of heterogeneous unmanned system | |
Wang et al. | Distributed model predictive control of inland ship platoon considering internal collision avoidance constraints | |
Zhao et al. | A Periodically Updated Adaptive Sampling Framework for Marine Mobile Observation Platforms | |
CN117215196B (en) | Ship-borne comprehensive control computer intelligent decision-making method based on deep reinforcement learning | |
Su et al. | USV Aided Reliable and Efficient Data Collection in Maritime Monitoring System | |
Zhang et al. | Risk Perception Oriented Autonomous Ship Navigation in AIS Environment | |
Yang et al. | Real-time Autonomous Glider Navigation Software | |
CN117748747B (en) | AUV cluster energy online monitoring and management system and method | |
Yan et al. | Multi-UUV collaborative search planning based on predictive control and backtracking |
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 |