CN112488359B - Multi-agent static multi-target trapping method based on RRT and OSPA distance - Google Patents
Multi-agent static multi-target trapping method based on RRT and OSPA distance Download PDFInfo
- Publication number
- CN112488359B CN112488359B CN202011202424.8A CN202011202424A CN112488359B CN 112488359 B CN112488359 B CN 112488359B CN 202011202424 A CN202011202424 A CN 202011202424A CN 112488359 B CN112488359 B CN 112488359B
- Authority
- CN
- China
- Prior art keywords
- target
- trapping
- distance
- agent
- captured
- 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
- 230000003068 static effect Effects 0.000 title claims abstract description 12
- 238000001926 trapping method Methods 0.000 title claims abstract description 7
- 101001086191 Borrelia burgdorferi Outer surface protein A Proteins 0.000 title claims abstract 6
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 14
- 239000003795 chemical substances by application Substances 0.000 claims description 35
- 238000005312 nonlinear dynamic Methods 0.000 claims description 6
- 230000001788 irregular Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 abstract description 2
- 230000002776 aggregation Effects 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000005457 optimization Methods 0.000 abstract description 2
- 238000005070 sampling Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 3
- 238000000059 patterning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- 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"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman 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
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Feedback Control In General (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a multi-agent static multi-target trapping method based on the distance between RRT and OSPA. According to the method, firstly, an agent keeps a leader-follower formation, a path planning path is conducted on a leader, the leader moves, and after entering a trapping area, OSPA distance calculation is conducted to allocate trapping points. And finally, each intelligent agent respectively performs route planning, moves to a trapping point and forms trapping on the static target. In the aspect of path planning, path planning is carried out between an intelligent agent and a surrounding point through an RRT algorithm, the path planning time is shortened through reducing the sampling range of the intelligent agent, and finally, path optimization is carried out through a Bezier curve. The method can allocate the trapping points to the intelligent body and plan the path by itself, and can effectively shorten the trapping time under the condition of multi-target relative aggregation by adding the mode judgment.
Description
Technical Field
The invention belongs to the technical field of intelligent systems, in particular to the technical field of intelligent body trapping and path planning, and relates to a multi-intelligent body static multi-target trapping method based on RRT (Rapid-exploring Random Trees, fast explored random tree algorithm) and OSPA (Optimal Sub-patten Assignment, optimal Sub-mode allocation) distance.
Background
The path planning problem is a collision-free safety path problem for a target object to find a collision-free safety path from a starting point to an end point in a range of a designated area, and the collision-free safety path problem has wide application in the fields of airport towing, logistics storage, traffic control, robots, electronic games and the like. There are many well-established algorithms such as: the potential field method, the algorithms D and A, the double-side convex hull expansion model algorithm, the ant colony algorithm and the like, wherein the rapid search algorithm (RRT) is one of the algorithms, and is a data structure and algorithm for searching the non-convex high-dimensional space efficiently. In the distributed system, mutual collaboration among multiple agents is established, so that the distributed agent system can efficiently complete complex tasks. The trapping control problem application mainly comprises unmanned vehicle trapping, aerial target trapping, water surface/underwater target trapping and the like in the military field.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a multi-agent static multi-target trapping method based on the distance between RRT and OSPA.
The method comprises the following steps:
(1) Multi-intelligent system formation:
set S= { S of N-agent-composed enclosure bodies 1 ,s 2 ,…,s N },s 1 S is the leader n For follower, n=2, 3, …, N;
the first order Lipschitz nonlinear dynamic equation of the leader is:
the first order Lipschitz nonlinear dynamic equation of the follower is:λ n ∈R N the position vector is the position vector of the nth agent; u (u) n ∈R N Is the control input parameter of the nth agent, R N Representing an N-dimensional vector; t is time, A, B, C is system parameter, f (lambda n T) is a continuously differentiable function and satisfies the lipschz condition.
For each follower, its control protocol is:
u n (t)=-Z(r(t))∑a nm [λ n (t)-λ m (t-τ(t))]-Z(r(t))∑a nm [λ n (t)-λ 1 (t-τ(t))],m=2,3,…,N,m≠n;
wherein lambda is m (t- τ (t)) represents the nth follower s n From the mth follower s m Acquired information lambda 1 (t- τ (t)) means that the nth follower is from the leader s 1 The acquired information; τ (t) is a time-varying time-lag function and satisfiesAndindicating the upper bound of time lag%>For the time lag derivative, h is the upper bound of the time lag derivative, Z represents the control gain, r (t) is the mode at the moment of t, a nm For follower s n Sum s m And connecting the weight values.
(2) Path planning:
(2.1) within the feasible region, relative to a randomly generated point q rand Select the closest q in the tree rand Node q of (2) near Will q near Extending to q with specified step size rand Generating a new node q new Connection q near And q new Adding an extended tree, and performing iterative computation by adopting an RRT algorithm until q new Arrive at target point q goal Or a target point area, find a base path in the extended tree.
(2.2) Bessel-contouring is performed on the base path: when the K-order Bezier curve is in the K-dimensional space, the curve is composed of k+1 control points P e Composition e=0, 1, …, k; the bezier curve of the k-order is expressed as:B k (t) is bezier curve k-order output, gamma is control input parameter of path, and gamma is more than or equal to 0 and less than or equal to 1; f (F) k,e (gamma) is the k-th order mixing function of the Bernstein polynomial,/for>Is a two-term coefficient>Representing factorial, H K ∈R K×K Is an identity matrix.
(3) Determining a trapping area:
target coordinate set to be captured g= { (x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x J ,y J )},(x j ,y j ) J=1, 2, …, J being the number of targets to be captured; surrounding J targets to be captured into an irregular polygonCentroid O coordinateDistance between each object to be captured and centroid O>The radius r=max { D of the circle j -a }; the centroid O is used as the center of a circle, and the area with the radius r is the trapping area.
(4) N intelligent bodies move to the trapping area, and when the leader s 1 When reaching the boundary of the trapping area, task allocation is started:
in (1) the->Representing the real target state of the sensor w at time v,/->Sensor w, representing moment v, estimates the target state, X v,w For a true multi-target state set,to estimate a multi-objective state set; p is the norm, μ and ρ are the real target number and the estimated target number, respectively; pi (II) N Represents the set of permutations, pi (l) represents the first permutation, l=n-! The method comprises the steps of carrying out a first treatment on the surface of the c is a horizontal distance parameter representing the average circle distance error of each error potential, d is +.>And->Distance.
Taking the last moment position before each agent enters the trapping area as a starting point set eta= { eta 1 ,η 2 ,…,η N And takes the target as the final point set g= { G 1 ,G 2 ,…,G J The OSPA distance is OSPA (p, c, η, G).
(5) Composing an acquisition formation:
n intelligent agents form a static circumference formation by taking a target to be captured as a circle center, the distances between adjacent intelligent agents are equal,the intelligent body keeps the same distance with the object to be captured and has the same interval angle theta n Distributed around the object to be captured, 0 < theta n < pi, number of traps required for a single target[]Represents a rounding, and the trapping point sigma= { sigma 1,1 ,…,σ 1,β ,σ 2,1 ,…,σ 2,β ,…,σ α,β },α=1,2,…,J,β=1,2,…,ε。
(6) The targets closest to the leader are captured preferentially, and then capturing is carried out sequentially from near to far according to the distance.
When the method is faced with the problem of multi-static target trapping, the method combining path planning and an OSPA algorithm can allocate the optimal trapping point to the intelligent agent and can self-route planning, and under the condition of multi-target relative aggregation, the trapping time can be effectively shortened.
Drawings
FIG. 1 is a flow chart of an implementation of the present invention;
FIG. 2 is a schematic diagram of a multi-agent system formation;
fig. 3 is a schematic diagram of RRT path planning;
FIG. 4 is a schematic view of a trapping region;
FIG. 5 is a diagram of a desired Capture formation.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, the multi-agent static multi-target trapping method based on RRT and OSPA distance is characterized in that an agent is firstly enabled to keep leading-following formation, a path planning is carried out on a leader, the leader leading follower moves, and after entering a trapping area, OSPA distance calculation is carried out to allocate trapping points. And finally, each intelligent agent respectively performs route planning, moves to a trapping point and forms trapping on the static target. In the aspect of path planning, path planning is carried out between an intelligent agent and a surrounding point through an RRT algorithm, the path planning time is shortened through reducing the sampling range of the intelligent agent, and finally, path optimization is carried out through a Bezier curve.
The method comprises the following steps:
(1) Multi-intelligent system formation:
as shown in fig. 2, in a multi-agent system composed of sixteen agents, the set of trapping agents s= { S 1 ,s 2 ,…,s 16 (s is therein 1 S is the leader 2 ~s 16 For the follower, the follower maintains an angle and speed to track the leader.
The first order Lipschitz nonlinear dynamic equation of the leader is:
the first order Lipschitz nonlinear dynamic equation of the follower is:n=2,3,…,16;λ n ∈R 16 the position vector is the position vector of the nth agent; u (u) n ∈R 16 The control input parameter is the control input parameter of the nth agent; t is time, A, B, C is system parameter, f (lambda n T) is a continuous and differentiable function and satisfies the lipschz condition;
for each follower, its control protocol is:
u n (t)=-Z(r(t))∑a nm [λ n (t)-λ m (t-τ(t))]-Z(r(t))∑a nm [λ n (t)-λ 1 (t-τ(t))],m=2,3,…,16,m≠n;
wherein lambda is m (t- τ (t)) represents the nth follower s n From the mth follower s m Acquired information lambda 1 (t- τ (t)) means that the nth follower is from the leader s 1 The acquired information; τ (t) is a time-varying time-lag function and satisfiesAnd indicating the upper bound of time lag%>For the time lag derivative, h is the upper bound of the time lag derivative, Z represents the control gain, r (t) is the mode at the moment of t, a nm For follower s n Sum s m And connecting the weight values.
(2) Path planning:
(2.1) As shown in FIG. 3, within the feasible region, a randomly generated point q is opposed rand Select the closest q in the tree rand Node q of (2) near Will q near Extending to q with specified step size rand Generating a new node q new Connection q near And q new Adding an extended tree, and performing iterative computation by adopting an RRT algorithm until q new Arrive at target point q goal Or a target point area at a start point q init And target point q goal Finds the base path.
(2.2) since the path generated by the RRT algorithm is not smooth enough, bezier patterning is performed on the base path. The bezier curve is one of typical parametric curves for smooth path generation.
Bessel-like patterning is performed on the base path: when the K-order Bezier curve is in the K-dimensional space, the curve is composed of k+1 control points P e Composition e=0, 1, …K; the bezier curve of the k-order is expressed as:B k (t) is bezier curve k-order output, gamma is control input parameter of path, and gamma is more than or equal to 0 and less than or equal to 1; f (F) k,e (gamma) is the k-th order mixing function of the Bernstein polynomial,/for>Is a two-term coefficient>
(3) Determining a trapping area:
the target to be captured is four G 1 、G 2 、G 3 And G 4 Its coordinate set g= { (x) 1 ,y 1 ),(x 2 ,y 2 ),(x 3 ,y 3 ),(x 4 ,y 4 ) }. As shown in FIG. 4, due to uncertainty of the trapping region, a trapping circular region is formed by the centroids of the irregular polygons, four objects to be trapped are trapped into the irregular polygons, and the coordinates of the centroids of the polygons are ODistance between each object to be captured and centroid O>The radius r=max { D of the circle j -a }; the centroid O is used as the center of a circle, and the area with the radius r is the trapping area.
(4) N intelligent bodies move to the trapping area, and when the leader s 1 When the boundary of the trapping area is reached, task allocation is started.
Due to the number of agents and targets, task allocation is required to ensure that the agents can reach the optimal trapping point, and the algorithm is adopted because the distance consistency of OSPA is good. The following OSPA distance is introduced to solve the problem that the Wasserstein distance is undefined when the set is empty.
In the method, in the process of the invention,representing the real target state of the sensor w at time v,/->Sensor w, representing moment v, estimates the target state, X v,w For a true multi-objective state set,/->To estimate a multi-objective state set; p is the norm, μ and ρ are the real target number and the estimated target number, respectively; pi (II) N Represents the set of permutations, pi (l) represents the first permutation, l=n-! The method comprises the steps of carrying out a first treatment on the surface of the c is a horizontal distance parameter representing the average circle distance error of each error potential, d is +.>And->Distance.
Taking the last moment position before each agent enters the trapping area as a starting point set eta= { eta 1 ,η 2 ,…,η N And takes the target as the final point set g= { G 1 ,G 2 ,…,G J The OSPA distance is OSPA (p, c, eta, G);
(5) Composing an acquisition formation:
sixteen intelligent agents take a target to be captured as a circle center to form a static circumference formation, the distances between adjacent intelligent agents are equal,(x n ,y n ) And (5) the coordinates of the first agent. Desirably, the formation of the capture is as shown in FIG. 5, with the agent maintained at the same distance from the target to be captured and at an equal angular separation θ n Distributed around the object to be captured, 0 < theta n < pi, number of traps required for a single target +.>
The capture point σ= { σ 1,1 ,…,σ 1,β ,σ 2,1 ,…,σ 2,β ,…,σ α,β },α=1,2,3,4,β=1,2,3,4。
(6) And (3) enclosing and catching: and (3) preferentially trapping the target closest to the leader, and then sequentially trapping according to the distance from the near to the far, wherein each intelligent agent plans a path according to the given starting point and the trapping point.
Claims (1)
1. The multi-agent static multi-target trapping method based on the distance between RRT and OSPA is characterized by comprising the following steps:
(1) Multi-intelligent system formation:
set S= { S of N-agent-composed enclosure bodies 1 ,s 2 ,…,s N },s 1 S is the leader n For follower, n=2, 3, …, N;
the first order Lipschitz nonlinear dynamic equation of the leader is:
the first order Lipschitz nonlinear dynamic equation of the follower is:λ n ∈R N the position vector is the position vector of the nth agent; u (u) n ∈R N Is the control input parameter of the nth agent, R N Representing an N-dimensional vector; t is time, A, B, C is system parameter, f (lambda n T) is continuously differentiableAnd satisfies the lipschz condition;
for each follower, its control protocol is:
u n (t)=-Z(r(t))∑a nm [λ n (t)-λ m (t-τ(t))]-Z(r(t))∑a nm [λ n (t)-λ 1 (t-τ(t))],m=2,3,…,N,m≠n;
wherein lambda is m (t- τ (t)) represents the nth follower s n From the mth follower s m Acquired information lambda 1 (t- τ (t)) means that the nth follower is from the leader s 1 The acquired information; τ (t) is a time-varying time-lag function and satisfiesAnd indicating the upper bound of time lag%>For the time lag derivative, h is the upper bound of the time lag derivative, Z represents the control gain, r (t) is the mode at the moment of t, a nm For follower s n Sum s m Connecting weight values;
(2) Path planning:
(2.1) within the feasible region, relative to a randomly generated point q rand Select the closest q in the tree rand Node q of (2) near Will q near Extending to q with specified step size rand Generating a new node q new Connection q near And q new Adding an extended tree, and performing iterative computation by adopting an RRT algorithm until q new Arrive at target point q goal Or a target point area, finding a basic path in the expansion tree;
(2.2) Bessel-contouring is performed on the base path: when the K-order Bezier curve is in the K-dimensional spaceThe curve is composed of k+1 control points P e Composition e=0, 1, …, k; the bezier curve of the k-order is expressed as:B k (t) is bezier curve k-order output, gamma is control input parameter of path, and gamma is more than or equal to 0 and less than or equal to 1; f (F) k,e (gamma) is the k-th order mixing function of the Bernstein polynomial,/for> Is a two-term coefficient>The following is carried out Representing factorial, H K ∈R K×K Is an identity matrix;
(3) Determining a trapping area:
target coordinate set to be captured g= { (x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x J ,y J )},(x j ,y j ) J=1, 2, …, J being the number of targets to be captured; surrounding J targets to be captured into an irregular polygon, and forming the O coordinates of the polygon centroidDistance between each object to be captured and centroid O>The radius r=max { D of the circle j -a }; taking the centroid O as the circle center, and taking the area with the radius r as the trapping area;
(4) N intelligent bodies move to the trapping area, and when the leader s 1 When reaching the boundary of the trapping area, task allocation is started:
in the method, in the process of the invention,representing the real target state of the sensor w at time v,/->Sensor w, representing moment v, estimates the target state, X v,w For a true multi-objective state set,/->To estimate a multi-objective state set; p is the norm, μ and ρ are the real target number and the estimated target number, respectively; pi (II) N Represents the set of permutations, pi (l) represents the first permutation, l=n-! The method comprises the steps of carrying out a first treatment on the surface of the c is a horizontal distance parameter representing the average circle distance error of each error potential, d is +.>And->A distance;
taking the last moment position before each agent enters the trapping area as a starting point set eta= { eta 1 ,η 2 ,…,η N And takes the target as the final point set g= { G 1 ,G 2 ,…,G J The OSPA distance is OSPA (p, c, eta, G);
(5) Composing an acquisition formation:
n intelligent agents form a static circumference formation by taking a target to be captured as a circle center, the distances between adjacent intelligent agents are equal,the intelligent body keeps the same distance with the object to be captured and has the same interval angle theta n Distributed around the object to be captured, 0 < theta n < pi, number of traps required for a single target[]Represents a rounding, and the trapping point sigma= { sigma 1,1 ,…,σ 1,β ,σ 2,1 ,…,σ 2,β ,…,σ α,β },α=1,2,…,J,β=1,2,…,ε;
(6) The targets closest to the leader are captured preferentially, and then capturing is carried out sequentially from near to far according to the distance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011202424.8A CN112488359B (en) | 2020-11-02 | 2020-11-02 | Multi-agent static multi-target trapping method based on RRT and OSPA distance |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011202424.8A CN112488359B (en) | 2020-11-02 | 2020-11-02 | Multi-agent static multi-target trapping method based on RRT and OSPA distance |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112488359A CN112488359A (en) | 2021-03-12 |
CN112488359B true CN112488359B (en) | 2023-11-17 |
Family
ID=74927906
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011202424.8A Active CN112488359B (en) | 2020-11-02 | 2020-11-02 | Multi-agent static multi-target trapping method based on RRT and OSPA distance |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112488359B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113759929B (en) * | 2021-09-22 | 2022-08-23 | 西安航天动力研究所 | Multi-agent path planning method based on reinforcement learning and model predictive control |
CN114089748B (en) * | 2021-11-09 | 2024-05-03 | 浙江工业大学 | Formation capturing method based on track prediction |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107943053A (en) * | 2017-12-15 | 2018-04-20 | 陕西理工大学 | A kind of paths planning method of mobile robot |
CN110517286A (en) * | 2019-08-12 | 2019-11-29 | 杭州电子科技大学 | Single goal dynamically track based on MAS control and surround and seize method |
CN110705803A (en) * | 2019-10-11 | 2020-01-17 | 福州大学 | Route planning method based on triangle inner center guide RRT algorithm |
CN111141304A (en) * | 2019-12-30 | 2020-05-12 | 福州大学 | Path planning method based on concentric circle sampling and RRT guiding algorithm |
-
2020
- 2020-11-02 CN CN202011202424.8A patent/CN112488359B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107943053A (en) * | 2017-12-15 | 2018-04-20 | 陕西理工大学 | A kind of paths planning method of mobile robot |
CN110517286A (en) * | 2019-08-12 | 2019-11-29 | 杭州电子科技大学 | Single goal dynamically track based on MAS control and surround and seize method |
CN110705803A (en) * | 2019-10-11 | 2020-01-17 | 福州大学 | Route planning method based on triangle inner center guide RRT algorithm |
CN111141304A (en) * | 2019-12-30 | 2020-05-12 | 福州大学 | Path planning method based on concentric circle sampling and RRT guiding algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN112488359A (en) | 2021-03-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110456823B (en) | Double-layer path planning method aiming at unmanned aerial vehicle calculation and storage capacity limitation | |
CN112488359B (en) | Multi-agent static multi-target trapping method based on RRT and OSPA distance | |
CN113495578B (en) | Digital twin training-based cluster track planning reinforcement learning method | |
Nguyen et al. | Formation control and obstacle avoidance of multiple rectangular agents with limited communication ranges | |
Zhilenkov et al. | System of autonomous navigation of the drone in difficult conditions of the forest trails | |
CN110632941A (en) | Trajectory generation method for target tracking of unmanned aerial vehicle in complex environment | |
CN113345008B (en) | Laser radar dynamic obstacle detection method considering wheel type robot position and posture estimation | |
CN113485371B (en) | Underwater multi-AUV path planning method based on improved sparrow search algorithm | |
CN108334080A (en) | A kind of virtual wall automatic generation method for robot navigation | |
CN109871031A (en) | A kind of method for planning track of fixed-wing unmanned plane | |
EP3690756B1 (en) | Learning method and learning device for updating hd map by reconstructing 3d space by using depth estimation information and class information on each object, which have been acquired through v2x information integration technique, and testing method and testing device using the same | |
Pan et al. | GPU accelerated real-time traversability mapping | |
Wei et al. | Underwater dynamic target tracking of autonomous underwater vehicle based on MPC algorithm | |
CN112651486A (en) | Method for improving convergence rate of MADDPG algorithm and application thereof | |
CN115143970A (en) | Obstacle avoidance method and system of underwater vehicle based on threat degree evaluation | |
Zhilenkov et al. | The use of convolution artificial neural networks for drones autonomous trajectory planning | |
CN111833381A (en) | Unmanned aerial vehicle target tracking trajectory generation method, unmanned aerial vehicle and storage medium | |
CN114397887A (en) | Group robot aggregation control method based on three-layer gene regulation and control network | |
Cheng et al. | Cooperative control of UAV swarm via information measures | |
CN112799393B (en) | Parking scene oriented map simplifying system | |
Zhang et al. | Enhanced fiducial marker based precise landing for quadrotors | |
CN111007848B (en) | Multi-agent cooperative operation control method based on bounded space | |
He et al. | A Moving Object Detection and Predictive Control Algorithm Based on Deep Learning | |
CN116520852A (en) | Method, device and equipment for capturing multiple targets by group robots under local information | |
CN113447948B (en) | Camera and multi-laser-radar fusion method based on ROS robot |
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 |