CN111123921B - Method for determining autonomous level for unmanned ship system navigation task - Google Patents
Method for determining autonomous level for unmanned ship system navigation task Download PDFInfo
- Publication number
- CN111123921B CN111123921B CN201911280797.4A CN201911280797A CN111123921B CN 111123921 B CN111123921 B CN 111123921B CN 201911280797 A CN201911280797 A CN 201911280797A CN 111123921 B CN111123921 B CN 111123921B
- Authority
- CN
- China
- Prior art keywords
- unmanned ship
- user
- ship system
- autonomous
- task
- 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
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000004891 communication Methods 0.000 claims abstract description 25
- 238000012790 confirmation Methods 0.000 claims abstract description 13
- 238000013439 planning Methods 0.000 claims description 77
- 230000019771 cognition Effects 0.000 claims description 34
- 239000011159 matrix material Substances 0.000 claims description 29
- 230000007613 environmental effect Effects 0.000 claims description 19
- 230000007704 transition Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 8
- 230000001149 cognitive effect Effects 0.000 claims description 6
- 230000003993 interaction Effects 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 3
- YBJHBAHKTGYVGT-ZKWXMUAHSA-N (+)-Biotin Chemical compound N1C(=O)N[C@@H]2[C@H](CCCCC(=O)O)SC[C@@H]21 YBJHBAHKTGYVGT-ZKWXMUAHSA-N 0.000 claims 1
- FEPMHVLSLDOMQC-UHFFFAOYSA-N virginiamycin-S1 Natural products CC1OC(=O)C(C=2C=CC=CC=2)NC(=O)C2CC(=O)CCN2C(=O)C(CC=2C=CC=CC=2)N(C)C(=O)C2CCCN2C(=O)C(CC)NC(=O)C1NC(=O)C1=NC=CC=C1O FEPMHVLSLDOMQC-UHFFFAOYSA-N 0.000 claims 1
- 238000004364 calculation method Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 7
- 230000006872 improvement Effects 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/0088—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/0206—Control of position or course in two dimensions specially adapted to water vehicles
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a method for determining an autonomous level for a navigation task of an unmanned ship system, which divides the autonomous level of the unmanned ship system into four steps: full remote control, user assistance, user confirmation, and full autonomy, each level of capacity configuration includes the responsibilities of the unmanned ship system and the responsibilities of the user. The autonomous level assessment method comprises the following aspects: (1) The autonomous level definition comprises names, cooperation modes and communication modes of the autonomous level; (2) And (5) evaluating the optimal autonomous level of the unmanned ship system in real time according to the situation of the unmanned ship system. The invention has good flexibility, real-time performance and accuracy.
Description
Technical Field
The invention belongs to the field of robots, and particularly relates to an autonomous level definition and evaluation method of an unmanned system with a variable autonomous level, in particular to an autonomous level definition and evaluation method of an unmanned ship system in a navigation task.
Background
The autonomous value of unmanned systems is to compensate and enhance the ability of humans, rather than to replace humans. Many of the tasks performed by unmanned systems are co-completed by human-machine cooperation. According to the definition of the national institute of standards and technology, the autonomy of an unmanned system is the ability of the unmanned system to integrate the application of awareness, analysis, communication, planning, decision-making, and actions, etc., in order to accomplish a given task, which is given by an operator through a man-machine interface, or by other systems by way of communication. Autonomy may be described in terms of task complexity, environmental difficulty, and man-machine interaction level. Thus, the autonomy of the unmanned system may be of a level. In the task execution process, the unmanned system with the variable autonomous capability is dynamically adjusted to a proper autonomous level according to different working states, so that the working pressure of a user can be reduced, and the task execution efficiency is improved.
The existing unmanned system autonomous level definition and adjustment methods mainly comprise three types: (1) Defining multiple autonomous levels in the system design stage, evaluating the system capacity when the task starts, and selecting one autonomous level to implement according to certain rules; (2) The method comprises the steps that a definite autonomous level is not defined in a system design stage, when a task is executed, the system capacity pair is evaluated in real time, and the control authority of the task execution is completely or partially transferred; (3) Variable autonomy is implemented based on a multi-Agent System (MAS), the capabilities of which correspond to agents, and the autonomy capabilities can be adjusted by starting or stopping some agents.
Because of the persistence of the navigation task, the situation of the unmanned ship system changes in real time in the task execution process, and uncertainty is brought to the task execution. In order to cope with the uncertainty of the situation of the unmanned ship system, the unmanned ship system can efficiently complete the navigation task, and operators can participate in the navigation task in various modes. According to the invention, the autonomous level of the unmanned ship system is classified according to whether the unmanned ship system has the capability of independently completing each sub-stage task in navigation, and the autonomous level of the unmanned ship system is dynamically adjusted in the navigation task.
The invention relates to a variable autonomous level definition and evaluation method for unmanned ship system navigation tasks, which has the following basic principle:
(1) Defining an autonomous level according to whether the unmanned ship system has the capability of independently completing each sub-stage task in navigation;
(2) The real-time situation where the unmanned ship system is located is described by environmental uncertainty, task uncertainty and user uncertainty, and the autonomous level of the unmanned ship system is dynamically adjusted with the uncertainty of the situation eliminated to the greatest extent.
The invention has the advantages that: and carrying out autonomous level calculation according to the real-time situation cognition of the unmanned ship system, so that the autonomous capacity adjustment of the system has good real-time performance and dynamic performance.
Disclosure of Invention
The invention aims to provide an autonomous level definition and evaluation method of a variable autonomous unmanned ship system. According to the invention, an autonomous level decision matrix is generated according to the change of the situation of the unmanned ship system, and an autonomous level for eliminating uncertainty to the greatest extent is selected as the autonomous level for executing tasks by the unmanned ship system.
The technical scheme adopted for solving the technical problems is as follows: the autonomous level definition and implementation of the unmanned ship system comprise the names of the autonomous levels, and the collaboration mode and communication mode of the user and the unmanned ship system under different levels; an autonomous level assessment algorithm comprises situation descriptions, an autonomous level decision matrix and a comparison relation between autonomous levels.
The invention comprises the following specific contents:
(1) The definition and implementation of the autonomous level comprise the names of the autonomous level, and the collaboration mode and communication mode of the user and the unmanned system under different levels. The navigation task of the unmanned ship system mainly comprises three stages of environment cognition, task planning and task execution, and the autonomous class of the unmanned ship system is classified into four according to whether the unmanned ship system has the capability of independently completing the subtasks of the three stages and the mode of participating in the task by a user.
Autonomous level 1: the full remote control unmanned ship system only has the capability of moving according to the user instruction and does not have the capability of environmental cognition and task planning; the communication mode is that the unmanned ship system actively feeds back the environment information and the task execution condition to the user periodically, and the user sends a control instruction to the unmanned ship system.
Autonomous level 2: the unmanned ship system has task planning and task execution capabilities, but environment cognition is completed by the assistance of the user; the communication mode is that the unmanned ship system actively informs the user of uncertain information to request the user to give auxiliary cognition.
Autonomous level 3: the user confirms that the unmanned ship system has the capabilities of environment cognition, task planning and task execution, but the reliability of the planning result is not high, and the user is required to assist in confirmation; the communication mode is that the unmanned ship system informs the user of the task planning result and uncertain information, and the user processes the planning result in two cases: (1) confirming a planning result of the unmanned ship system, and enabling the unmanned ship system to execute the planning result; (2) and overruling the planning result of the unmanned ship system, providing auxiliary cognition for uncertain information, and enabling the unmanned ship system to be planned again.
Autonomous level 4: the unmanned ship system is fully autonomous and has environment cognition, task planning and task execution capacity; the communication mode is that the unmanned ship system periodically feeds back the task execution condition to the user.
(2) The invention also discloses an autonomous level evaluation algorithm for evaluating the autonomous level of the unmanned ship variable autonomous system, which specifically comprises situation description, an autonomous level decision matrix and a comparison relation between autonomous levels.
Autonomous rank assessment problem description:
A. autonomous level use a= { a for unmanned ship system 1 ,a 2 ,a 3 ,a 4 Represented by }, a 1 Indicating full remote control, a 2 Representing user assistance; a, a 3 Representing user confirmation; a, a 4 Representing full autonomy;
B. situation S of unmanned ship system is determined by environment uncertainty S e Task uncertainty S t And user uncertainty S u To describe, i.e. s= { S e ,S t ,S u Or expressed as s= { S } 1 ,s 2 ,···,s n (s is therein i Indicating uncertainty factors, respectively S e ，S t Or S u ，s i The value range of (1) can be set as [0,1 ]]The larger the value is, the smaller the uncertainty is, n is the total number of uncertainty factors, and the aim of autonomous level adjustment of the unmanned ship system is to reduce the overall uncertainty of the situation to the maximum extent;
C. the decision matrix is denoted by D (S), i.e
d i (S) represents the case where the autonomous level of the unmanned ship system is set to a when it is in situation S i New situation, i-th row of matrix and autonomous level a, which can be achieved by the system i Correspondingly, the j-th column of the matrix is associated with the uncertainty factor s j Corresponding, i.e. element d of the matrix ij (s j ) Indicating that the system is at autonomous level a i When the system is under uncertainty factor s j The value of the aspect. d, d ij (s j ) Can be abbreviated as d ij D (S) may be abbreviated as D;
D. the calculation formula of the contribution of the change of the uncertainty factor value to the system deterministic improvement is that
Wherein delta is the difference between two elements in the same column in decision matrix D, let delta ijk ＝d ik -d jk Substituting delta in (2) to obtain the slave autonomous level a i Transition to autonomous level a j At the time of uncertainty factor s k The certainty with which an aspect can be obtained is improved, where σ is determined statistically by the system designer by the distribution characteristics of each attribute. Let the autonomous level of the system be a i The system transitions to other autonomous level a j In this case, the system certainty of uncertainty factors promotes contribution vector P ij Is that
P ij ＝[p(d i1 -d j1 )p(d i2 -d j2 )···p(d in -d jn )]；
E. Calculation of contribution of autonomous level change to system deterministic improvement and settingRepresenting slave autonomous level a i Transition to autonomous level a j The certainty of the obtained is improved,
wherein 0 is<w k <1, k=1, 2, ··, n is the weight of each uncertainty factor, the value of the weight is determined by the system designer, w= [ w 1 w 2 ··· w n ]。
The method for starting the autonomous level adjustment of the unmanned ship system can be periodic or triggered based on events, and the steps of the autonomous level adjustment of the unmanned ship system are the same in either case, and the detailed steps are as follows:
A. initializing, namely initializing a data structure involved in a calculation process;
B. acquiring a real-time situation S of an unmanned ship system;
C. obtaining a decision matrix D according to the formula (1);
D. calculating the change from the current autonomous level a to the other autonomous level a according to the formula (3) j In time, certainty of unmanned system is promoted
E. According toThe values order the autonomous levels if all +.>Then no shift is made, otherwise, find ∈ ->J, which has the greatest value, is then selected to correspond to the autonomous level a j And is implemented.
Drawings
Fig. 1 is a flow chart of the autonomous level adjustment steps of the unmanned ship system.
Fig. 2 is a schematic diagram of the situation composition of an unmanned ship system.
Detailed Description
The invention is described in detail below by way of example with reference to the accompanying drawings:
(1) The definition of the autonomous level comprises the name of the autonomous level, and the cooperation and communication modes between the user and the unmanned ship system.
A. Full remote control
Definition: the unmanned ship system only has the capability of taking actions according to the user instructions and does not have the capability of environmental cognition and task planning; the result of the mission planning is given directly by the user or all the actions of the unmanned ship system are controlled by the user.
Cooperative mode: and the user performs path planning and controls the behavior of the unmanned ship system according to the environmental information returned by the unmanned ship system.
Communication mode: the unmanned ship system actively feeds back the task execution condition and the environment information to the user periodically, and the user actively sends out instructions to the unmanned ship system.
B. User assistance
Definition: the unmanned ship system has task planning and execution capacity, but environment cognition is completed by the assistance of a user; the unmanned ship system has higher uncertainty on the obtained task planning information, so that the efficiency and reliability of task planning can be reduced, and the efficiency of task planning is improved through user-assisted cognition.
Cooperative mode: the unmanned ship system sends a query to the user aiming at the parameter information with higher uncertainty, and waits for the reply of the user; the user gives out a cognitive result according to the environmental information and feeds the cognitive result back to the unmanned ship system; and the unmanned ship system performs autonomous task planning and execution after receiving the response of the user.
Communication mode: the unmanned ship system makes a query to the user for uncertain information, which the user gives assistance to cognition.
C. User confirmation
Definition: the unmanned ship system can independently perform environment cognition, task planning and task execution, but the uncertainty of the acquired information is higher, so that the uncertainty of the environment cognition is higher, and the uncertainty of a task planning result is higher; unmanned ship systems are capable of mission planning, but lack certainty of the planning results.
Cooperative mode: the unmanned ship system performs task planning fully independently, but does not execute immediately, and the planning result needs to be notified to the user to obtain the approval of the user; the communication mode is that the unmanned ship system informs the user of the task planning result and uncertain information, and the user processes the planning result in two cases: (1) confirming a planning result of the unmanned ship system, and enabling the unmanned ship system to execute the planning result; (2) and overruling the planning result of the unmanned ship system, providing auxiliary cognition for uncertain information, and enabling the unmanned ship system to be planned again.
Communication mode: the unmanned ship system informs the user of the planning result, and the user gives confirmation or provides auxiliary information.
D. Fully autonomous
Definition: the unmanned ship system can independently perform environment cognition, task planning and task execution; the unmanned ship system has a positive result on environmental cognition and task planning.
Cooperative mode: the unmanned ship system performs task planning and execution fully independently, and other interactions are not actively sent to the user except for periodically feeding back task execution situation accidents according to time slices.
Communication mode: the unmanned ship system actively feeds back the task execution condition to the user periodically.
(2) According to the autonomous level assessment algorithm disclosed by the invention, the detailed steps are as follows:
A. autonomous level use a= { a for unmanned ship system 1 ,a 2 ,a 3 ,a 4 Represented by }, a 1 Indicating full remote control, a 2 Representing user assistance; a, a 3 Representing user confirmation; a, a 4 Representing full autonomy;
B. situation S of unmanned ship system is determined by environment uncertainty S e Task uncertainty S t And user uncertainty S u To describe, i.e. s= { S e ,S t ,S u }。
S e The static environment information uncertainty s can be used 1 And dynamic environmental information uncertainty s 2 Measured by the weight of the sample. The users with different autonomous levels participate in cognition to different degrees, and the degree of uncertainty of the environment is reduced to different degrees, so that the improvement of the degree of certainty of the unmanned ship system situation is also different. User intervention can increase the degree of certainty of the environmental information.
S t Can be determined by task objective difficulty uncertainty s 3 And task achievement progress uncertainty s 4 To measure, when the unmanned system is in different autonomous levels, the different autonomous levels are not used for reducing tasks according to different participation modes of usersThe contribution of certainty is different.
S u Through user skill level uncertainty s 5 And user work preference uncertainty s 6 To measure that different users have different skill levels and work preferences, and that the improvement in overall situational certainty of the unmanned ship system is not the same for users at different levels of autonomy.
Thus S is also denoted s= { S 1 ,s 2 ,···,s 6 (s is therein i Indicating uncertainty factors, respectively S e ，S t Or S u The unmanned ship system situation is formed as shown in figure 2;
C. the decision matrix is denoted by D (S), i.e
d i (S) represents the case where the autonomous level of the unmanned ship system is set to a when it is in situation S i New situation, i-th row of matrix and autonomous level a, which can be achieved by the system i Correspondingly, the j-th column of the matrix is associated with the uncertainty factor s j Corresponding, i.e. element d of the matrix ij (s j ) Indicating that the system is at autonomous level a i When the system is under uncertainty factor s j The value of the aspect. d, d ij (s j ) Can be abbreviated as d ij D (S) may be abbreviated as D;
D. the calculation formula of the contribution of the change of the uncertainty factor value to the system deterministic improvement is that
Wherein delta is the difference between two elements in the same column in decision matrix D, let delta ijk ＝d ik -d jk Substituting delta in (5) to obtain the slave autonomous level a i Transition to autonomous level a j At the time of uncertainty factor s k A deterministic improvement in aspects can be obtained, where σ is designed by the systemThe distribution characteristics of each attribute are determined by a statistical method. Let the autonomous level of the system be a i The system transitions to other autonomous level a j In this case, the system certainty of uncertainty factors promotes contribution vector P ij Is that
P ij ＝[p(d i1 -d j1 )p(d i2 -d j2 )···p(d in -d jn )]；
E. Calculation of contribution of autonomous level change to system deterministic improvement and settingRepresenting slave autonomous level a i Transition to autonomous level a j The resulting improved performance is achieved in that,
wherein 0 is<w k <1, k=1, 2, ··, n is the weight of each uncertainty factor, the value of the weight is determined by the system designer, w= [ w 1 w 2 ··· w n ]。
The method for adjusting the autonomous level of the unmanned ship system can be periodically started or triggered by an event, and the steps for adjusting the autonomous level of the unmanned ship system are the same in either case, and the detailed steps are as follows:
A. initializing, namely initializing a data structure involved in a calculation process;
B. acquiring a real-time situation of an unmanned ship system;
C. obtaining a decision matrix D according to a formula (4);
D. calculating the change from the current autonomous level a to the other autonomous level a according to the formula (6) j In time, certainty of unmanned system is promoted
E. According toThe values order the autonomous levels if all +.>Then no shift is made, otherwise, find ∈ ->J, which has the greatest value, is then selected to correspond to the autonomous level a j And is implemented.
In one embodiment, the invention discloses a method for determining an autonomous level for an unmanned ship system navigation task, wherein the autonomous level comprises full remote control, user assistance, user confirmation and full autonomy, and the method comprises the following steps:
s1, initializing;
s2, acquiring a real-time situation of an unmanned ship system;
s3, obtaining a decision matrix;
s4, calculating the current autonomous level a of the unmanned ship system according to the decision matrix i To autonomous level a j In time, certainty of unmanned ship system is promoted
S5, improving certainty of unmanned ship systemThe autonomous levels are ordered by the value of (2) if all certainty improves +.>Then the shift is not converted, otherwise find out to let deterministic promote +>J, which has the greatest value, is then selected to correspond to the autonomous level a j And at an autonomous level a j Is implemented.
Further, the autonomyRank a= { a 1 ,a 2 ,a 3 ,a 4 }，a 1 Indicating full remote control, a 2 Representing user assistance, a 3 Indicating user confirmation, a 4 Representing full autonomy; situation s= { S of unmanned ship system e ,S t ,S u By environmental uncertainty S e Task uncertainty S t And user uncertainty S u To describe, each uncertainty has a corresponding uncertainty factor s j 。
Further, S e Including static environment information uncertainty s 1 And dynamic environmental information uncertainty s 2 ，S t Including task objective difficulty uncertainty s 3 And task achievement progress uncertainty s 4 ，S u Including user skill level uncertainty s 5 And user work preference uncertainty s 6 。
Further, the decision matrix is denoted by D (S):
d i (S) represents the case where the autonomous level is set to a when the unmanned ship system is in the real-time situation S i The ith row of the decision matrix D (S) and the autonomous rank a i Correspondingly, the j-th column of the decision matrix is associated with the uncertainty factor s j Correspondingly, element d of the decision matrix ij (s j ) Indicating that the system is at autonomous level a i When the system is under uncertainty factor s j The value of the aspect.
Further, it is provided withIndicating that the unmanned ship system is currently at autonomous level a i From the autonomous level a i Transition to autonomous level a j The obtained certainty is improved:
w k weights of each uncertainty factor, 0<w k <1, k=1, 2, the terms, n, is the weight of each uncertainty factor, w= [ w ] 1 w 2 ··· w n ]。
System deterministic boost contribution vector P for uncertainty factors ij Is that
P ij ＝[p(d i1 -d j1 )p(d i2 -d j2 )···p(d in -d jn )]；
p(d ik -d jk ) Calculated according to the following formula
Delta is the difference between two elements in the same column in decision matrix D, let delta ijk ＝d ik -d jk Will delta ijk Substituting p (delta) to obtain slave autonomous level a i Transition to autonomous level a j At the time of uncertainty factor s k The aspect can be improved by the determination that σ is the standard deviation of the normal distribution.
A. Full remote control: the unmanned ship system only has the capability of taking actions according to the user instructions and does not have the capability of environmental cognition and task planning; the result of the mission planning is given directly by the user or all the actions of the unmanned ship system are controlled by the user.
Full remote control collaboration mode: and the user performs path planning and controls the behavior of the unmanned ship system according to the environmental information returned by the unmanned ship system.
Full remote control communication mode: the unmanned ship system actively feeds back the task execution condition and the environment information to the user periodically, and the user actively sends out instructions to the unmanned ship system.
B. User assistance: the unmanned ship system has task planning and execution capacity, but environment cognition is completed by the assistance of a user; the unmanned ship system has higher uncertainty on the obtained task planning information, so that the efficiency and reliability of task planning can be reduced, and the efficiency of task planning is improved through user-assisted cognition.
User-assisted collaboration mode: the unmanned ship system sends a query to the user aiming at the parameter information with higher uncertainty, and waits for the reply of the user; the user gives out a cognitive result according to the environmental information and feeds the cognitive result back to the unmanned ship system; and the unmanned ship system performs autonomous task planning and execution after receiving the response of the user.
User-assisted communication scheme: the unmanned ship system makes a query to the user for uncertain information, which the user gives assistance to cognition.
C. User confirmation: the unmanned ship system can independently perform environment cognition, task planning and task execution, but the uncertainty of the acquired information is higher, so that the uncertainty of the environment cognition is higher, and the uncertainty of a task planning result is higher; unmanned ship systems are capable of mission planning, but lack certainty of the planning results.
User confirms the collaboration mode: the unmanned ship system performs task planning fully independently, but does not execute immediately, and the planning result needs to be notified to the user to obtain the approval of the user; the communication mode is that the unmanned ship system informs the user of the task planning result and uncertain information, and the user processes the planning result in two cases: (1) confirming a planning result of the unmanned ship system, and enabling the unmanned ship system to execute the planning result; (2) and overruling the planning result of the unmanned ship system, providing auxiliary cognition for uncertain information, and enabling the unmanned ship system to be planned again.
The user confirms the communication mode: the unmanned ship system informs the user of the planning result, and the user gives confirmation or provides auxiliary information.
D. Full autonomy: the unmanned ship system can independently perform environment cognition, task planning and task execution; the unmanned ship system has a positive result on environmental cognition and task planning.
Full autonomous collaboration mode: the unmanned ship system performs task planning and execution fully independently, and other interactions are not actively sent to the user except for periodically feeding back task execution situation accidents according to time slices.
Full autonomous communication mode: the unmanned ship system actively feeds back the task execution condition to the user periodically.
While the invention has been described with reference to the preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (3)
1. A method for determining an autonomous level for an unmanned ship system navigation task, the autonomous level comprising full remote control, user assistance, user confirmation, and full autonomy, the method comprising the steps of:
s1, initializing;
s2, acquiring a real-time situation of an unmanned ship system;
s3, obtaining a decision matrix;
wherein, the decision matrix is represented by D (S):
d i (S) represents the case where the autonomous level is set to a when the unmanned ship system is in the real-time situation S i The ith row of the decision matrix D (S) and the autonomous rank a i Correspondingly, the k column of the decision matrix is associated with the uncertainty factor s k Correspondingly, element d of the decision matrix ik (s k ) Indicating that the system is at autonomous level a i When the system is under uncertainty factor s k The value of the aspect is taken out, k=1, 2, carrying out the following steps;
s4, calculating the current autonomous level a of the unmanned ship system according to the decision matrix i To autonomous level a j In time, certainty of unmanned ship system is promoted
Wherein the autonomous level a= { a 1 ,a 2 ,a 3 ,a 4 }，a 1 Indicating full remote control, a 2 Representing user assistance, a 3 Indicating user confirmation, a 4 Representing full autonomy; the situation s= { Se, st, su } of the unmanned ship system is described by an environmental uncertainty Se, a task uncertainty St and a user uncertainty Su, each uncertainty having a corresponding uncertainty factor S k The method comprises the steps of carrying out a first treatment on the surface of the Se includes static environment information uncertainty s 1 And dynamic environmental information uncertainty s 2 St includes task target difficulty uncertainty s 3 And task achievement progress uncertainty s 4 Su includes user skill level uncertainty s 5 And user work preference uncertainty s 6 ；
S5, improving certainty of unmanned ship systemThe autonomous levels are ordered by the value of (2) if all certainty improves +.>Then the shift is not converted, otherwise find out to let deterministic promote +>J, which has the greatest value, is then selected to correspond to the autonomous level a j And at an autonomous level a j Is implemented.
2. The method for determining an autonomous level for an unmanned ship system navigation task according to claim 1, whereinIndicating that the unmanned ship system is currently at autonomous level a i From the autonomous level a i Transition to autonomous level a j The obtained certainty is improved:
w k is the weight of each uncertainty factor, 0<w k <1，k＝1,2,···,n，w＝[w 1 w 2 ···w n ],
System deterministic boost contribution vector P for uncertainty factors ij ：
P ij ＝[p(d i1 -d j1 )p(d i2 -d j2 )···p(d in -d jn )]
p(d ik -d jk ) Calculated according to the following formula
Delta is the difference between two elements in the same column in decision matrix D, let delta ijk ＝d ik -d jk Will delta ijk Substituted into p (delta)
Wherein delta is obtained from the autonomous level a i Transition to autonomous level a j At the time of uncertainty factor s k
The certainty of the aspect is improved, and sigma is the standard deviation of normal distribution.
3. The method for determining an autonomous level for an unmanned ship system navigation task according to claim 1, wherein the autonomous level defines and cooperates and communicates at each level:
A. full remote control: the unmanned ship system only has the capability of taking actions according to the user instructions, does not have the capability of environmental cognition and task planning, and is directly given to a task planning result by a user or controls all actions of the unmanned ship system by the user;
full remote control collaboration mode: the user performs path planning and controls the behavior of the unmanned ship system according to the environmental information returned by the unmanned ship system;
full remote control communication mode: the unmanned ship system actively feeds back the task execution condition and the environment information to the user periodically, and the user actively sends out an instruction to the unmanned ship system;
B. user assistance: the unmanned ship system has task planning and execution capacity, but environment cognition is completed by the assistance of a user;
user-assisted collaboration mode: the unmanned ship system sends out a query to the user aiming at the parameter information with higher uncertainty, waits for the reply of the user, gives out a cognitive result according to the environment information, feeds back the cognitive result to the unmanned ship system, and carries out autonomous task planning and execution after receiving the reply of the user;
user-assisted communication mode: the unmanned ship system makes a query to the user aiming at the uncertain information, and the user gives auxiliary cognition;
C. user confirmation: the unmanned ship system can independently perform environment cognition, task planning and task execution, and can perform task planning, but lacks certainty on planning results;
user-confirmed collaboration mode: the unmanned ship system performs task planning fully autonomously, but does not execute immediately, the planning result needs to be notified to the user, the user approval is obtained, the communication mode is that the unmanned ship system notifies the user of the task planning result and uncertain information, and the user processes the planning result in two cases:
(1) confirming a planning result of the unmanned ship system, and enabling the unmanned ship system to execute the planning result;
(2) overruling a planning result of the unmanned ship system, providing auxiliary cognition for uncertain information, and enabling the unmanned ship system to be planned again;
user-confirmed communication mode: the unmanned ship system informs the planning result to the user, and the user confirms or provides auxiliary information;
D. full autonomy: the unmanned ship system can independently perform environment cognition, task planning and task execution, and the unmanned ship system is more definite in environment cognition and task planning results;
full autonomous collaboration mode: the unmanned ship system performs task planning and execution fully independently, and other interactions are not actively sent to the user except for periodically feeding back task execution situation accidents according to time slices;
full autonomous communication mode: the unmanned ship system actively feeds back the task execution condition to the user periodically.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911280797.4A CN111123921B (en) | 2019-12-13 | 2019-12-13 | Method for determining autonomous level for unmanned ship system navigation task |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911280797.4A CN111123921B (en) | 2019-12-13 | 2019-12-13 | Method for determining autonomous level for unmanned ship system navigation task |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111123921A CN111123921A (en) | 2020-05-08 |
| CN111123921B true CN111123921B (en) | 2023-09-05 |
Family
ID=70500038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911280797.4A Active CN111123921B (en) | 2019-12-13 | 2019-12-13 | Method for determining autonomous level for unmanned ship system navigation task |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111123921B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114815594A (en) * | 2021-12-04 | 2022-07-29 | 中国船舶工业系统工程研究院 | Autonomous level transformation method for nondeterministic multi-variable factors of water surface unmanned system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807159A (en) * | 2010-03-18 | 2010-08-18 | 西北工业大学 | Self-adapting task scheduling method |
| CN103234763A (en) * | 2013-04-09 | 2013-08-07 | 北京理工大学 | System and method for quantitatively evaluating unmanned vehicles |
| CN106200679A (en) * | 2016-09-21 | 2016-12-07 | 中国人民解放军国防科学技术大学 | Single operation person's multiple no-manned plane mixing Active Control Method based on multi-modal natural interaction |
| CN109960148A (en) * | 2019-03-20 | 2019-07-02 | 中国科学院软件研究所 | A kind of the independence appraisal procedure and system of intelligent Unmanned Systems |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7765038B2 (en) * | 2005-02-16 | 2010-07-27 | Lockheed Martin Corporation | Mission planning system for vehicles with varying levels of autonomy |
-
2019
- 2019-12-13 CN CN201911280797.4A patent/CN111123921B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807159A (en) * | 2010-03-18 | 2010-08-18 | 西北工业大学 | Self-adapting task scheduling method |
| CN103234763A (en) * | 2013-04-09 | 2013-08-07 | 北京理工大学 | System and method for quantitatively evaluating unmanned vehicles |
| CN106200679A (en) * | 2016-09-21 | 2016-12-07 | 中国人民解放军国防科学技术大学 | Single operation person's multiple no-manned plane mixing Active Control Method based on multi-modal natural interaction |
| CN109960148A (en) * | 2019-03-20 | 2019-07-02 | 中国科学院软件研究所 | A kind of the independence appraisal procedure and system of intelligent Unmanned Systems |
Non-Patent Citations (1)
| Title |
|---|
| 董英茹等.采用排序组织法的USV自主等级控制算法.辽宁工程技术大学学报（自然科学版）.2018,第37卷(第1期),第210-215页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111123921A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106875090B (en) | Dynamic task-oriented multi-robot distributed task allocation forming method | |
| Zhao et al. | A heuristic distributed task allocation method for multivehicle multitask problems and its application to search and rescue scenario | |
| JP6059375B1 (en) | Production control system and integrated production control system | |
| Chapman et al. | Decentralised dynamic task allocation: A practical game–theoretic approach | |
| Pirjanian | Multiple objective behavior-based control | |
| CN109829636A (en) | emergency task scheduling planning method based on dynamic priority | |
| CN110533301A (en) | A kind of population dispatching method based on dynamic constrained matrix | |
| Chen et al. | Rule driven multi objective dynamic scheduling by data envelopment analysis and reinforcement learning | |
| CN111123921B (en) | Method for determining autonomous level for unmanned ship system navigation task | |
| Geng et al. | Learning to cooperate in decentralized multi-robot exploration of dynamic environments | |
| CN110806758B (en) | Unmanned aerial vehicle cluster autonomous level self-adaptive adjustment method based on scene fuzzy cognitive map | |
| CN110704986B (en) | Mechanical system reliability distribution fuzzy method facing element action | |
| Hou et al. | A Multiagent Cooperative Learning System With Evolution of Social Roles | |
| CN109946687B (en) | Unmanned platform integrated radio frequency system working mode decision method | |
| Pirjanian et al. | Multiple objective vs. fuzzy behavior coordination | |
| Le Ny et al. | Multi-agent task assignment in the bandit framework | |
| CN116449864A (en) | Optimal path selection method for unmanned aerial vehicle cluster | |
| CN113657718B (en) | Multi-robot dynamic alliance task allocation method and related device | |
| Mantel et al. | Trust networks in multi-robot communities | |
| Rose et al. | Comparing four metaheuristics for solving a constraint satisfaction problem for ship outfitting scheduling | |
| CN115016405A (en) | Process route multi-objective optimization method based on deep reinforcement learning | |
| CN111563596B (en) | Uncertain information reasoning target identification method based on evidence network | |
| Figueiredo et al. | Group decision-based construction of scenarios for multicriteria analysis in conditions of uncertainty on the basis of quantitative and qualitative information | |
| Zhu et al. | Consensus analysis of UAV swarm cooperative situation awareness | |
| CN112256057A (en) | Heterogeneous unmanned system cooperative control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |