CN110381442B - Swarm robot target searching method based on implicit information interaction mode - Google Patents

Swarm robot target searching method based on implicit information interaction mode Download PDF

Info

Publication number
CN110381442B
CN110381442B CN201910760972.3A CN201910760972A CN110381442B CN 110381442 B CN110381442 B CN 110381442B CN 201910760972 A CN201910760972 A CN 201910760972A CN 110381442 B CN110381442 B CN 110381442B
Authority
CN
China
Prior art keywords
robot
target
grid
probability
search
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
Application number
CN201910760972.3A
Other languages
Chinese (zh)
Other versions
CN110381442A (en
Inventor
刘明雍
李赛楠
苏晗
石廷超
杨扬
李嫣然
王旭辰
黄宇轩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN201910760972.3A priority Critical patent/CN110381442B/en
Publication of CN110381442A publication Critical patent/CN110381442A/en
Application granted granted Critical
Publication of CN110381442B publication Critical patent/CN110381442B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23—Updating
    • G06F16/2379—Updates performed during online database operations; commit processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/70—Services for machine-to-machine communication [M2M] or machine type communication [MTC]

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention provides a swarm robot target searching method based on an implicit information interaction mode, which comprises the steps of firstly, gridding a searching area into a plurality of square discrete units according to fixed intervals, and establishing a searching map of each robot; secondly, for each robot, updating the probability of the target in each grid in the search map according to the detection result of the robot, introducing the active detection of the individual on the state of the neighbor robot into the search algorithm, overcoming the information interaction barrier between different individuals when the broadcast communication cannot be carried out, and thus, under the condition that the target distribution condition is unknown, mutually matching a plurality of robots to complete the search task. The invention avoids the dependence on an explicit information interaction mode, performs information fusion only according to the observation of the individual to the neighbor state, and adopts an implicit communication mode according to the observation information to solve the problem of multi-robot target search under the limited communication.

Description

Swarm robot target searching method based on implicit information interaction mode
Technical Field
The invention relates to a swarm robot target searching method under implicit information interaction, and belongs to the technical field of multi-robot target searching.
Background
The multi-robot target search is a novel search technology for completing a target search task by utilizing a plurality of robots, and the technology can effectively overcome the limitation of single robot search, such as the problems that the single robot cannot continue to search due to sudden failure, the endurance time of the single robot is insufficient, and the like.
At present, research on multi-robot target search at home and abroad mainly takes coordination control as a representative, and the method is characterized in that different robots broadcast and communicate the states of the robots, and the robots share data to perform distributed planning. However, such broadcast interactive mode is limited by channel interference, insufficient communication bandwidth, additional energy consumption for information propagation in the medium, and the individual must agree extensively on the meaning of a particular signal, which is relatively costly. Due to the existence of communication constraint, communication interaction between different robots cannot be completed normally, the robots cannot share information between individuals, each individual searches independently, the advantages of the group cannot be embodied, and the method is a simple combination of one and one.
Disclosure of Invention
In order to solve the problems in the prior art and overcome the influence of communication constraint on individual interaction, the invention provides a swarm robot target search method based on an implicit information interaction mode, which avoids the dependence on an explicit information interaction mode, performs information fusion only according to the observation of an individual on a neighbor state, and adopts an implicit communication mode to solve the problem of multi-robot target search under the condition of limited communication according to observation information.
The technical scheme of the invention is as follows:
the swarm robot target searching method based on the implicit information interaction mode is characterized in that: the method comprises the following steps:
step 1: rasterizing search regions to L at fixed intervals Δ sx×LyA square discrete unit, each grid is marked as sc＝(xs,ys) Wherein x iss∈{1,2,3,...,Lx},ys∈{1,2,3,...,Ly}; establishing a search map for each robot according to the rasterized search area, and searching each grid information structure M in the mapp＝{xs,ys,p(xs,ys)Denotes a structure representing a grid scA coordinate position within the two-dimensional plane and a probability of the presence of the target in the grid;
step 2: for each robot in the swarm robots, the following implicit information interaction-based method is adopted for searching the target:
step 2.1: initialization: for the ith robot, initializing the probability of the target existing in each grid in the search map to be
Figure GDA0002590415410000021
Step 2.2: each robot detects the grids in the detection target surface, the observation result is in binomial distribution according to the existence condition of the target in the observation grid of the airborne sensor, only two observation values of the target exist and the target do not exist, and the observation result is displayed through the color of the indicator light;
step 2.3: and updating the probability of the target existing in each grid in the self-searching map once by each robot according to the observation result and the following formula:
Figure GDA0002590415410000022
wherein
Figure GDA0002590415410000023
Grid s in search map showing robot i at time tcThe probability of existence of the object of (1),
Figure GDA0002590415410000024
grid s in search map representing robot i at time t +1cTarget existence probability of peIs the detection probability of the sensor, pfIs the false alarm probability of the sensor, tau sc1 denotes a grid scIn the presence of the actual target, τ sc0 denotes a grid scIn which there is actually no target present in the image,
Figure GDA0002590415410000025
indicating that robot i is at time t +1 for grid s c1 represents that the target is observed, 0 represents that the target is not observed, and the search map of each robot is gradually updated according to the formula;
step 2.4: if a neighbor robot appears in the detection target surface, the position of the neighbor robot and the target detection condition in the detection target surface can be known according to the color of the indicator light displayed by the neighbor robot, and the probability of the corresponding position is further fused and updated according to the observation result of the neighbor according to the following formula:
Figure GDA0002590415410000026
wherein the content of the first and second substances,
Figure GDA0002590415410000027
representing the probability distribution, p, in each grid after the robot i fuses the observations of itself and neighborsiRepresenting probability distribution in each grid after the robot i is updated according to the observed value of the robot, j representing neighbors in the perception range of the robot i, N representing the number of the neighbor robots, rijIs the position coupling coefficient, | sci-scj||2Denotes the Euclidean distance, p, between robot i and robot jijRepresenting probability distribution in each grid updated by the robot i according to the detection result of the neighbor;
step 2.5: for each robot, moving to the grid with the maximum target existence probability in the detection target surface according to the updated probability distribution in each grid; if a neighbor robot exists in the detection target surface, setting the grid where the neighbor exists and the adjacent grid of the grid where the neighbor exists as infeasible points, and moving the robot to the grid with the maximum target existence probability in the feasible region;
and step 3: for a certain robot, if the target existence probability of a certain grid in a search map of the robot reaches or exceeds a set upper limit of the target existence probability, the robot is considered to successfully search the target; and if the target existence probabilities of all the grids in the search map reach or are smaller than the set lower limit of the target existence probabilities, determining that no target exists in the search area.
Advantageous effects
The invention provides a swarm robot target searching method based on an implicit information interaction mode, which is characterized in that under the condition that broadcast communication is blocked due to limited communication, an individual updates a probability map according to a detection result of the individual based on the implicit communication mode, and introduces the individual to actively detect the state of a neighbor robot in a searching algorithm, so that the information interaction barrier between different individuals when the broadcast communication cannot be carried out is overcome, and a plurality of robots are matched with each other to complete a searching task under the condition that the target distribution condition is unknown.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a model schematic of a search environment;
FIG. 2 is a search area coverage comparison;
FIG. 3 is a search efficiency comparison;
FIG. 4 shows a probability variation curve of different grid targets of the robot; (a) a change in target presence probability of the robot 7 on the grid (10,10), (b) a change in target presence probability of the robot 7 on the grid (5,5), (c) a change in target presence probability of the robot 3 on the grid (10,10), (d) a change in target presence probability of the robot 3 on the grid (5, 5);
FIG. 5 is a diagram of different robot motion trajectories; (a) the motion trail of the robot 7, and (b) the motion trail of the robot 3.
Detailed Description
The invention relates to a swarm robot target searching method, which is characterized in that a plurality of robots are matched with each other to complete a searching task under the conditions that only the geographical position and the area size of a searching area are known and the target distribution condition is unknown.
Step 1 search for region S ∈ R2Rasterized into L at a fixed interval Δ sx×LyA square discrete unit, each grid is marked as sc＝(xs,ys) Wherein x iss∈{1,2,3,...,Lx},ys∈{1,2,3,...,Ly}; and establishing a search map for each robot according to the rasterized search area, wherein the search perception map reflects the understanding and cognition of the robot to the current search environment. Structure M for searching each grid information in mapp＝{xs,ys,p(xs,ys)Denotes a structure representing a grid scThe coordinate position within the two-dimensional plane and the probability of the presence of an object in the grid.
The robot senses the environment continuously along with the time, the search map is dynamically updated according to a set rule, and the continuously updated environment map reflects that the understanding degree of the robot on the search area is deepened continuously. And the robot carries out online search and decision according to the self environment map. The object existence probability distribution in the search map of robot i can be defined as Lx×LyIn the form of a matrix of
Figure GDA0002590415410000041
And when the target existence probability is updated to the set threshold value, ending the searching process.
Step 2: the swarm robot target searching method is based on an implicit communication mode, an individual updates a probability chart according to a detection result of the individual, and due to limited communication, broadcast communication is blocked, so that active detection of the individual on the state of a neighbor robot is introduced into a searching algorithm, and the information interaction obstacle between different individuals when broadcast communication cannot be carried out is overcome. The method comprises the following specific steps:
step 2.1: initialization: the swarm robot system does not have prior information for the target, and under the condition of considering probability normalization, for the ith robot, the probability that the target exists in each grid in the search map is initialized to be
Figure GDA0002590415410000042
Step 2.2: each robot detects the grids in the detection target surface, the observation result is in binomial distribution according to the existence condition of the target in the observation grid of the airborne sensor, only two observation values of the target exist and the target do not exist, and the observation result is displayed through the color of the indicator light.
Step 2.3: and updating the probability of the target existing in each grid in the self-searching map once by each robot according to the observation result and the following formula:
Figure GDA0002590415410000051
wherein
Figure GDA0002590415410000052
Grid s in search map showing robot i at time tcThe probability of existence of the object of (1),
Figure GDA0002590415410000053
grid s in search map representing robot i at time t +1cTarget existence probability of peIs the detection probability of the sensor, pfIs the false alarm probability of the sensor, tau sc1 denotes a grid scIn the presence of the actual target, τ sc0 denotes a grid scIn which there is actually no target present in the image,
Figure GDA0002590415410000054
indicating that robot i is at time t +1 for grid s c1 indicates that the target is observed, 0 indicates that the target is not observed, and the search map of each robot is gradually updated according to the formula.
Step 2.4: if a neighbor robot appears in the detection target surface, the position of the neighbor robot and the target detection condition in the detection target surface can be known according to the color of the indicator light displayed by the neighbor robot, and the probability of the corresponding position is further fused and updated according to the observation result of the neighbor according to the following formula:
Figure GDA0002590415410000055
wherein the content of the first and second substances,
Figure GDA0002590415410000056
representing the probability distribution, p, in each grid after the robot i fuses the observations of itself and neighborsiRepresenting probability distribution in each grid after the robot i is updated according to the observed value of the robot, j representing neighbors in the perception range of the robot i, N representing the number of the neighbor robots, rijIs the position coupling coefficient, | sci-scj||2Denotes the Euclidean distance, p, between robot i and robot jijRepresenting the probability distribution in each grid that robot i updates according to the detection results of the neighbors.
Step 2.5: for each robot, moving to the grid with the maximum target existence probability in the detection target surface according to the updated probability distribution in each grid; if a neighbor robot exists in the detection target surface, setting the grid where the neighbor exists and the adjacent grid of the grid where the neighbor exists as the infeasible points, and moving the robot to the grid with the maximum target existence probability in the feasible region.
And step 3: for a certain robot, if the target existence probability of a certain grid in a search map of the robot reaches or exceeds a set upper limit of the target existence probability, the robot is considered to successfully search the target; and if the target existence probabilities of all the grids in the search map reach or are smaller than the set lower limit of the target existence probabilities, determining that no target exists in the search area.
The invention measures the advantages and disadvantages of the search algorithm by evaluating the efficiency of target search and the coverage rate of the search area. The target search efficiency refers to the number of simulation steps required for completing the search in one search process, and the number of the simulation steps reflects the search efficiency. The coverage of the search area is calculated as follows along with the continuous increase of the simulation step length under the condition that the upper limit and the lower limit of the simulation end condition are not set: coverage is 100% of covered grid/total number of grids.
The following detailed description of embodiments of the invention is intended to be illustrative, and not to be construed as limiting the invention.
In the present embodiment, a square area with a search area of 20m × 20m and a grid map with a discrete area of 20 × 20 are selected, and the positions of static objects are arranged in grids (10,10), such as black five-pointed stars in fig. 5.
In the present embodiment, the robot system is composed of 10 isomorphic robots, and the initial positions of the robots are randomly generated within the search area, as indicated by the open circles in fig. 5.
Detection probability p of robot airborne sensoreIs 0.9, false alarm probability pfAnd 0.3, the robot does not obtain prior information of the existence of the target in the search area initially, and all grids are used as initial distribution by uniform probability distribution. In the 20 x 20 grid region, the initial probabilities are all 0.0025, the position coupling coefficient rijSet to 0.5. The upper bound of the target existence probability after the search is finished is 90%, the target exists in the search area of the embodiment, and the lower bound of the target existence probability is not set.
Fig. 4 shows the variation of the target existence probability of the grids (10,10) and (5,5) with the simulation step size, taking the robot 7 and the robot 3 as an example, the curve rising indicates that the target is observed at the moment, the curve falling indicates that the target is not observed at the moment, and the curve level indicates that the grid is not currently in the detection target surface of the robot.
According to the steps, observation, judgment, decision and execution are repeated, the broadcast communication method cannot be carried out under the condition of communication constraint, and a search efficiency comparison graph showing that interaction does not exist among individuals and implicit interaction exists is shown in fig. 3.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art without departing from the principle and spirit of the present invention.

Claims (2)

1. A swarm robot target searching method based on an implicit information interaction mode is characterized in that: the method comprises the following steps:
step 1: rasterizing search regions to L at fixed intervals Δ sx×LyA square discrete unit, each grid is marked as sc＝(xs,ys) Wherein x iss∈{1,2,3,...,Lx},ys∈{1,2,3,...,Ly}; establishing a search map for each robot according to the rasterized search area, and searching each grid information structure in the map
Figure FDA0002590415400000011
Showing that the structure represents a grid scA coordinate position within the two-dimensional plane and a probability of the presence of the target in the grid;
step 2: for each robot in the swarm robots, the following implicit information interaction-based method is adopted for searching the target:
step 2.1: initialization: for the ith robot, initializing the probability of the target existing in each grid in the search map to be
Figure FDA0002590415400000012
Step 2.2: each robot detects the grids in the detection target surface, the observation result is in binomial distribution according to the existence condition of the target in the observation grid of the airborne sensor, only two observation values of the target exist and the target do not exist, and the observation result is displayed through the color of the indicator light;
step 2.3: and updating the probability of the target existing in each grid in the self-searching map once by each robot according to the observation result and the following formula:
Figure FDA0002590415400000013
wherein
Figure FDA0002590415400000014
Grid s in search map showing robot i at time tcThe probability of existence of the object of (1),
Figure FDA0002590415400000015
grid s in search map representing robot i at time t +1cTarget existence probability of peIs the detection probability of the sensor, pfIs the probability of a false alarm for the sensor,
Figure FDA0002590415400000016
representation grid scIn which the object is actually present in the image,
Figure FDA0002590415400000017
representation grid scIn which there is actually no target present in the image,
Figure FDA0002590415400000018
indicating that robot i is at time t +1 for grid sc1 represents that the target is observed, 0 represents that the target is not observed, and the search map of each robot is gradually updated according to the formula;
step 2.4: if a neighbor robot appears in the detection target surface, the position of the neighbor robot and the target detection condition in the detection target surface can be known according to the color of the indicator light displayed by the neighbor robot, and the probability of the corresponding position is further fused and updated according to the observation result of the neighbor according to the following formula:
Figure FDA0002590415400000021
wherein the content of the first and second substances,
Figure FDA0002590415400000022
indicating that the robot i fuses itself with neighborsProbability distribution in each grid after observation, piRepresenting probability distribution in each grid after the robot i is updated according to the observed value of the robot, j representing neighbors in the perception range of the robot i, N representing the number of the neighbor robots, rijIs the position coupling coefficient, | sci-scj||2Denotes the Euclidean distance, p, between robot i and robot jijRepresenting probability distribution in each grid updated by the robot i according to the detection result of the neighbor;
step 2.5: for each robot, moving to the grid with the maximum target existence probability in the detection target surface according to the updated probability distribution in each grid;
and step 3: for a certain robot, if the target existence probability of a certain grid in a search map of the robot reaches or exceeds a set upper limit of the target existence probability, the robot is considered to successfully search the target; and if the target existence probabilities of all the grids in the search map reach or are smaller than the set lower limit of the target existence probabilities, determining that no target exists in the search area.
2. The swarm robot target searching method based on the implicit information interaction mode as claimed in claim 1, wherein: in step 2.5, if a neighbor robot exists in the detection target surface, setting the grid where the neighbor exists and the adjacent grid of the grid where the neighbor exists as the infeasible points, and moving the robot to the grid with the maximum target existence probability in the feasible region.
CN201910760972.3A 2019-08-17 2019-08-17 Swarm robot target searching method based on implicit information interaction mode Active CN110381442B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910760972.3A CN110381442B (en) 2019-08-17 2019-08-17 Swarm robot target searching method based on implicit information interaction mode

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910760972.3A CN110381442B (en) 2019-08-17 2019-08-17 Swarm robot target searching method based on implicit information interaction mode

Publications (2)

Publication Number Publication Date
CN110381442A CN110381442A (en) 2019-10-25
CN110381442B true CN110381442B (en) 2020-09-22

Family

ID=68259770

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910760972.3A Active CN110381442B (en) 2019-08-17 2019-08-17 Swarm robot target searching method based on implicit information interaction mode

Country Status (1)

Country Link
CN (1) CN110381442B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111941424A (en) * 2020-08-13 2020-11-17 上海机器人产业技术研究院有限公司 Robot moving path coverage rate measuring method and system
CN112836777B (en) * 2021-03-02 2022-09-27 同济大学 Application method of consensus initiative mechanism in group robot target search
CN113741461B (en) * 2021-09-06 2023-10-03 中国人民解放军国防科技大学 Multi-robot obstacle avoidance method oriented to limited communication under complex scene
CN114547228B (en) * 2022-04-22 2022-07-19 阿里云计算有限公司 Track generation method, device, equipment and storage medium

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080026248A (en) * 2006-09-20 2008-03-25 삼성전자주식회사 Method and apparatus for cell searching of cell reselection in portable communication system
US8442805B2 (en) * 2008-08-05 2013-05-14 Daniel Reem Efficient computation of Voronoi diagrams of general generators in general spaces and uses thereof
CN102096415B (en) * 2010-12-31 2012-09-26 重庆邮电大学 Multi-robot formation method based on Ad-Hoc network and leader-follower algorithm
CN104143021B (en) * 2014-07-17 2017-07-11 西北工业大学 A kind of space rope Web robot's approaches dynamic modeling method
CN105787998B (en) * 2016-02-25 2018-11-13 武汉大学 The two-grid of more ball particles searches for contact detecting method in a kind of emulation of discrete element
CN105700555B (en) * 2016-03-14 2018-04-27 北京航空航天大学 A kind of multiple no-manned plane collaboratively searching method based on gesture game
CN106323293B (en) * 2016-10-14 2018-12-25 淮安信息职业技术学院 Two groups of multidirectional robot path planning methods based on multiple target search
CN109299210A (en) * 2018-11-06 2019-02-01 哈尔滨工业大学（深圳） A kind of multiple no-manned plane distributed collaboration searching method based on information fusion

Also Published As

Publication number Publication date
CN110381442A (en) 2019-10-25

Similar Documents

Publication Publication Date Title
CN110381442B (en) Swarm robot target searching method based on implicit information interaction mode
CN110244733B (en) Mobile robot path planning method based on improved ant colony algorithm
CN108229719B (en) Multi-objective optimization method and device for unmanned aerial vehicle formation task allocation and flight path planning
CN110389595B (en) Dual-attribute probability map optimized unmanned aerial vehicle cluster cooperative target searching method
CN105892480B (en) Isomery multiple no-manned plane systematic collaboration, which is examined, beats task self-organizing method
CN110991972B (en) Cargo transportation system based on multi-agent reinforcement learning
CN106647744B (en) Robot path planning method and device
CN108897312A (en) Lasting supervised path planing method of more unmanned vehicles to extensive environment
Meyer-Delius et al. Using artificial landmarks to reduce the ambiguity in the environment of a mobile robot
CN101459915A (en) Wireless sensor network node coverage optimization method based on genetic algorithm
CN110418391B (en) Underwater acoustic sensor network trust model construction method based on support vector machine
CN109118064B (en) Optimal resource allocation method for probe task planning
CN112001309A (en) Target searching method, device and equipment based on unmanned aerial vehicle cluster and storage medium
CN114003059B (en) UAV path planning method based on deep reinforcement learning under kinematic constraint condition
CN110825112B (en) Oil field dynamic invasion target tracking system and method based on multiple unmanned aerial vehicles
Zhicai et al. A cooperative search algorithm based on improved particle swarm optimization decision for UAV swarm
CN114910072A (en) Unmanned aerial vehicle navigation method, device, equipment and medium based on deep reinforcement learning
US20200047330A1 (en) Robot and method for localizing robot
CN108966241B (en) Optimization method for self-adaptively improving fish swarm algorithm
Zhang et al. Unmanned ground vehicle navigation in coordinate-free and localization-free wireless sensor and actuator networks
CN115860210A (en) Multi-sensor station distribution method facing target space coverage
CN115525842A (en) Cooperative decision-making method and system for federal learning unmanned aerial vehicle based on emotional preference and migration behavior
CN115097861A (en) Multi-Unmanned Aerial Vehicle (UAV) capture strategy method based on CEL-MADDPG
CN114721435A (en) Soft constraint-based defense method for unmanned aerial vehicle cluster and related equipment
CN112465127B (en) Multi-agent collaborative target searching method based on improved biological heuristic neural network

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