GB2505715A - Sensor path optimisation method and system - Google Patents
Sensor path optimisation method and system Download PDFInfo
- Publication number
- GB2505715A GB2505715A GB1216168.3A GB201216168A GB2505715A GB 2505715 A GB2505715 A GB 2505715A GB 201216168 A GB201216168 A GB 201216168A GB 2505715 A GB2505715 A GB 2505715A
- Authority
- GB
- United Kingdom
- Prior art keywords
- path
- node
- sensor
- nodes
- optimisation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000003780 insertion Methods 0.000 claims abstract description 14
- 230000037431 insertion Effects 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims abstract description 9
- 238000004590 computer program Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 abstract 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000000926 separation method Methods 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64U—UNMANNED AERIAL VEHICLES [UAV]; EQUIPMENT THEREFOR
- B64U2101/00—UAVs specially adapted for particular uses or applications
- B64U2101/30—UAVs specially adapted for particular uses or applications for imaging, photography or videography
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (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)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
A sensor path optimisation method comprises selecting a path connecting a plurality of nodes, wherein a node j is selected and added to the path based on a density parameter which is related to a probability of detection of a target at the node j by the sensor and inversely related to a distance between the node j and a node i which is already in the path. A path improvement operation may be performed to find a shorter path each time a node is added which may comprise disconnecting two nodes to obtain multiple sub-paths. Path selection may be made using a heuristic algorithm which is a modified nearest insertion algorithm. The method may be implemented by a sensor path optimization control system in an Unmanned Airborne Vehicle (UAV).
Description
I
SENSOR PATH OPTIMISATION METHOD AND SYSTEM
Field of the Invention
The present invention relates to a method and a system for sensor path optirnisation.
Background to the Invention
An Unmanned Airborne Vehicle (UAV) used for airborne reconnaissance is equipped with a sensor that can move in different directions. The area that the sensor can cover when pointing at all mechanically possible positions is known as the "Field of Regard".
When the sensor is pointed in a particular direction, the area that the sensor can see without moving is called the "Field of View". Figure 1 illustrates the Field of View and Field of Regard. Each Field of View can be associated with a utility value which is a measure of the probability that a target exists in that area and of the probability of detection given that the target is present. Thus the Field of Regard can be represented as a set of cells, where each cell is a single Field of View with an associated utility value, as shown in Figure 2.
As UAVs are unable to hover the sensor has limited time to scan the area, and it is not possible for the whole Field of Regard to be scanned, For example, the Field of Regard may be approximately 7 km2 compared to the Field of View which may be less than 0.1km2. -Therefore the UAV must decide on a sensor path which is limited by the time available for the scanning, but which should optimise the probability of target detection ie maximise the total utility of the path. This is the reason an optimisation algorithm is required.
An additional consideration is that the algorithm must be fast. As the UAV is moving, it has a limited time to plan a scanning path. For example, the planning horizon may be only 10 seconds.
Therefore, an algorithm is required which maximises the probability of target detection within a certain time limit, wherein the algorithm must also be fast enough to plan the paths within the planning horizon of the UAV.
There are existing algorithms which gradually construct a path by adding nodes according to some heuristics, such as the Nearest Neighbour and Nearest Insertion heuristics. These algorithms are fast but do not provide good results. In particular, these methods find a solution by using only a single metric. Thus, these algorithms might generate a path which connects nodes having high utility values but the nodes may be too far apart such that in a limited time the total utility of the path is not high.
Methods that try to improve the paths with stochastic algorithms, such as Genetic Algorithms or Ant Colony Optimisation can lead to good solutions because they can run multi-objective optimisations (ie maximisation of utility and minimisation of path length or time). However, these methods are too slow and may take several minutes or even hours to generate a good solution.
Summary of the Invention
The present invention provides a sensor path optimisation method comprising: selecting a path connecting a plurality of nodes; wherein a node j is selected and added to the path based on a density parameter which is related to a probability of detection of a target at the node j by the sensor and inversely related to a distance between the node j and a node i which is already in the path.
Thus the present invention combines the utility and the node separation into a single parameter, density, in which the two component parameters are inversely related.
Preferably each node has an associated utility value, the utility value being related to a probability of detection of a target at the node by the sensor and wherein the density parameter is defined by: u/t; wherein u is the utility value of the node j, and is a measure of the distance between the node j and the node i which is already in the path.
Nodes are added to the path based on maximising the density parameter, although the parameter could equally well be inverted by defining a parameter as t1 /u and adding nodes which minimise this parameter, Preferably, each time a node is added to the path, a path improvement operation is performed to find a shorter path connecting the nodes and, if a shorter path is found, adjusting the path to match the shorter path.
In general, prior art path optimisation methods may run a path improvement algorithm at the end of the path selection process. In the present invention, path improvement operations are carried out in parallel with the path selection.
In one embodiment the path improvement operation comprises disconnecting two nodes in the path to obtain two sub-paths, reconnecting a Fast node of the first sub-path to a final node of the original path to obtain a new path and determining whether the new path is shorter than the original path.
Alternatively, the path imprQvement operation comprises disconnecting two pairs of nodes in the path, to obtain three sub-paths, reconnecting the three sub paths in a different configuration to obtain a single new path and determining whether the new path is shorter then ihe original path.
Preferably the path selection is made using a heuristic algorithm.
In one embodiment the heuristic algorithm is a modified Nearest Insertion algorithm comprising the steps of: a) selecting a node not yet included in the path based on the density parameter of the node in relation to nodes already in the path; and b) inserting the selected node in the path.
Preferably, step (a) comprises selecting a node with the highest sum of densities in relation to two nodes already in the path and step (b) comprises inserting the selected node in the path between the two nodes.
Preferably, step (b) comprises inserting the selected node in the path in a position which minimises total path length.
In another embodiment, the heuristic algorithm is a modified Nearest Neighbour algorithm which iteratively adds to the path at a current end node a node with the highest density in relation to the current end node of the path.
The present invention also provides a data collection method, in which the sensor path is optirnised using the above described method to plan a sensor path! and then a sensor is moved in accordance with the planned path.
The invention also provides a sensor path optiniisation system configured to carry out the method, a sensor control system for a vehicle including such a system, and an unmanned airborne vehicle including the control system.
The present invention also provides a computer program comprising computer readable instructions for carrying out the method according to the invention.
Brief Description of the Drawings
Embodiments of. the present invention will now be described with reference to the accompanying drawings, in which: Figure 1 illustrates a UAV and a Field of View and Field of Regard; Figure 2 illustrates the Field of Regard as a cell grid, wherein each hexagonal cell
represents a Field of View;
Figure 3 illustrates node insertion in a Nearest Insertion Heuristic method; Figure 4 illustrates a 2-opt exchange iterative improvement method; Figure 5 illustrates a 1-opt exchange iterative improvement method; and Figure 6 shows a comparison of optimisation methods.
Detailed Description
The path optimisation method of the present invention can be illustrated with reference to an optimisation problem of a sensor path of an Unmanned Airborne Vehicle (UAV).
As shown in Figure 1, the UAV has a Field of Regard, which represents all mechanically possible positions at which the sensor may point, when the UAV is at one position. The Field of View is a much smaller area which represents the area that the sensor can see when pointing in one particular direction. Each Field of View can be allocated a utility value, which is a measure of the probability that a target exists in that area and of the probability of detection given that the target is present.
As shown in Figure 2, the Field of Regard can be represented as a cell grid, where each hexagonal cell represents a Field of View with a utility value associated with it.
Darker grey values represent higher utility cells.
The time available to scan the Field of Regard is United, and thus the problem can be seen as choosing a path which keeps the path length within limits while maxirnising the total utility of the path.
Each cell can be defined as a node having a utility value, with the node location being the centre of the cell.
The present invention makes use of a parameter called density, in which the utility of a node and the distance between nodes (or the time taken between nodes) are inversely related. This parameter allows utility and time to be integrated in a single value, so the optimisation process will privilege nearer nodes with higher utilities.
The density of a node j in relation to a node i may be defined as: ui/ti1 wherein uj is the utility value of the node j, and t is a measure of the distance between the node j and the node i, The density parameter can be used by heuristic algorithms to find paths, which can then be improved by path improvement operations.
In a first embodiment, the invention uses a modified Nearest Insertion Heuristics.
The original Nearest Insertion Heuristics comprises selecting two initial nodes and connecting them to form a path. A node not in the path, which has the shortest distance to any one of the nodes in the path, is selected and added to the path such that the cost of doing this is minimal. The node selection is then repeated by selecting another node which is not in the path, which has the shortest distance to any one of the nodes in the path and adding the node to the path such that the cost of doing this is minimal. The process is repeated until the ending criteria, which may be a total number of nodes or a totar path length, are satisfied.
The modified Nearest Insertion Heuristics according to the first embodiment of the invention is illustrated in Figures 3A and 3B. First; two initial nodes 1, 2 are selected and connected to form a path. A node 3, which is not in the path and which has the maximum sum of densities between two nodes aFready in the path is seiected and added to the path such that the totaL path length is minimum compared to other possible choices.. The node selection is then repeated by selecting node S which is not in the path, which has the maximum sum of densities between two nodes already in the path and adding the node 5 to the path such that the total path length is minimum compared to other possible choices.
After each node is inserted, a path improvement operation is carried out, which will be described later.
The algorithm terminates when the time spent to cover the path is equal to or higher than the time limit ie the path length reaches a maximum limit.
The selection of nodes for insertion in the path differs from the original Nearest Insertion Algorithm because the original method is generally used for "travelling salesman" type problems, wherein all nodes will eventually be included in the path. In the present invention, the path may not include some nodes because the sensor may not have enough time to visit them all. For this reason, the heuristic used in the present invention inserts the node so that the resulting path will have minimum impact in terms of total path length.
The modified Nearest Insertion algorithm could get stuck in a local minimum if the chosen ending node is not good (for example if it is too close to the starting node or is located in an area where there are not too many neighbours with high utility). This problem could be solved by choosing different ending points and running several optimisations in parallel. The Nearest Insertion heuristic is a very fast method, so a solution can still be generated quickly even if it is repeated several times.
A second embodiment of the invention builds a path using a modified Nearest Neighbour heuristic.
Given a starting node, the unmodified Nearest Neighbour heuristic simply iteratively adds to the path the nearest node to the current node, and the algorithm terminates when the time spent to cover the path is equal to or higher than the time limit.
According to the modified algorithm of the second embodiment of the invention, the algorithm adds to the path a node which has the highest density from the current end node in the path...
As in the first embodiment, after each node is added, a path improvement operation is carried out, which will be described later. The algorithm terminates when the time spent to cover the path is equal to or higher than the time limit.
This method may be repeated a plurality of times with different starting points. The scores of all the best paths, in terms of total utility are stored together with the paths.
In scoring the paths, the time spent by the sensor to go from its current position to the start node of the path is taken into account. After obtaining a number of "best paths", the best of the "best paths" is selected as the sensor path. The Nearest Neighbour method is very fast, which means that a solution can still be generated quickly even if the method is repeated numerous times.
As discussed above, in each of the embodiments, each time a node is added to the path, a path improvement operation is carried out. Two such path improvement operations will now be described, either of which may be used with either of the two path selection methods described above.
A first possible path improvement operation is known as the "2-opt exchange" algorithm, and is illustrated in Figure 4. The 2-opt exchange algorithm comprises the following steps: (1) eliminate two "edges" from the initial path, wherein an edge" is a path directly connecting two nodes in the path, so the step involves disconnecting two pairs of nodes, to obtain three separate sub paths.
(2) reconnect the three sub paths in a different way to obtain a new continuous path. As shown in Figure 4, there is only one way to reconnect the paths so that they form a single valid path.
(3) check whether the new path is shorter than the original path and if it is, take the new path as the starting path and repeat the process. If it is not, then repeat the process disconnecting different nodes, The algorithm continues removing and reconnecting edges until no 2-opt improvements are be found in a finite time period or number of attempts.
A second possible path improvement operation has been devised and called the 1-opt exchange" algorithm. This is illustrated in Figure 5. The 1-opt exchange algorithm comprises the following steps: (1) remove one edge' from the path by disconnecting two nodes, to create two sub paths.
(2) reconnect the last node in the first sub path with the final, or goal node of the original path. This will generate a new path with a different goal node.
(3) check whether the new path is shorter than the original path and if it is, take the new path as the starting path and repeat the process. If it is not, then repeat the process disconnecting different nodes.
The algorithm continues disconnecting and reconnecting nodes until the time available is spent, or until no more nodes remain.
Figure 6 shows a comparison of the two embodiments of the invention with two conventional methods, the unmodified Nearest Neighbour and the Ant Colony Optimisation. The results are compared in terms of the total utility of the path and the time taken.
The conventional Nearest Neighbour method provided the fastest but worst results, whereas the Ant Colony Optimisation provides better results but has the worst processing time. The two methods of the invention, the modified versions of Nearest Insertion Heuristics and Nearest Neighbour, using the density parameter, provide the best results within an acceptable time scale.
It will be clear to the person skilled in the art that although the density parameter is defined as u1ft1j, the inverse of this parameter (inverse density) could equally well be used, with the parameter being minimised rather than maxirnised. The important feature of the parameter is that the utility and the time between nodes are inversely related to each other.
The present invention may be embodied in software run by a processor of a sensor control system, wherein the software comprises instructions to the processor to plan a sensor path using the optimisation method of the invention. The control system then controls a sensor to move in accordance with the optimised path to collect data. The control system may be a control system in an unmanned surveillance system, particularly an unmanned airborne vehicle used for airborne reconnaissance.
However, the present invention may be used in other routing problems where speed is essential and where the costs are on the edges (generally the distance between nodes) and also on the nodes themselves (eg the utility).
Claims (3)
- CLAIMS: 1. A sensor path optimisation method comprising: selecting a path connecting a plurality of nodes; wherein a node j is selected and added to the path based on a density parameter which is related to a probability of detection of a target at the node j by the sensor and inversely related to a distance between the node j and a node i wiich is already in the path.
- 2. A sensor path optimisation method according to claim 1, wherein each node has an associated utility value, the utility value being related to a probability of detection of a target at the node by the sensor and wherein the density parameter is defined by: uIt1 wherein u is the utility value of the node j, and t is a measure of the distance between the node j and the node i which is already in the path.
- 3. A sensor path optimisation method according to claim 1 or 2. wherein each time a node is added to the path, a path improvement operation is performed to find a shorter path connecting the nodes and, if a shorter path is found, adjusting the path to match the shorter path A sensor path optimisation method according to claim 3, wherein the path improvement operation comprises disconnecting two nodes in the path to obtain two sub-paths, reconnecting a last node of the first sub-path to a final node of the original path to obtain a new path and determining whether the new path is shorter than the original path.5. A sensor path optimisation method according to claim 3, wherein the path improvement operation comprises disconnecting two pairs of nodes in the path, to obtain three sub-paths, reconnecting the three sub paths in a dftferent configuration to obtain a single new path and determining whether the new path is shorter than the original path.6. A sensor path optimisation method according to any one of the preceding claims, wherein the path selection is made using a heuristic algorithm.7. A sensor path optimisation method according to claim 6, wherein the heuristic algorithm is a modified Nearest Insertion algorithm comprising the steps of: a) selecting a node not yet included in the path based on the density parameter of the node in relation to nodes already in the path; and b) inserting the selected node in the path.8. A sensor path optimisation method according to claim 7, wherein step (a) comprises selecting a node with the highest sum of densities in relation to two nodes already in the path and step (b) comprises inserting the selected node in the path between the two nodes.9. A sensor path optimisation method according to claim 7, wherein step (b) comprises inserting the selected node in the path in a position which minimises total path length.10. A sensor path optimisation method according to claim 6, wherein the heuristic algorithm is a modified Nearest Neighbour a]gorithm which iteratively adds to the path at a current end node a node with the highest density in relation to the current end node of the path...11. A data collection method for collecting data using a sensor, comprising using a sensor path optirnisation method according to any one of the preceding claims to plan a sensor path and moving a sensor in accordance with the planned sensor path to collect data.12. A sensor path optimisation system configured to carry out the method of any one of the preceding cLaims.13. A sensor control system for a vehicle including the sensor path optimisation system of claim 12, and further configured to control a sensor to move in accordance with an optiniised path determined by the sensor path optimisation system.14. An Unmanned Airborne Vehicle including the control system of claim 13.15. A computer program comprising computer readable instructions for carrying out the method according to any one of claims ito ii.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| GB1216168.3A GB2505715B (en) | 2012-09-11 | 2012-09-11 | Sensor path optimisation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| GB1216168.3A GB2505715B (en) | 2012-09-11 | 2012-09-11 | Sensor path optimisation method and system |
Publications (3)
| Publication Number | Publication Date |
|---|---|
| GB201216168D0 GB201216168D0 (en) | 2012-10-24 |
| GB2505715A true GB2505715A (en) | 2014-03-12 |
| GB2505715B GB2505715B (en) | 2015-12-30 |
Family
ID=47137241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| GB1216168.3A Expired - Fee Related GB2505715B (en) | 2012-09-11 | 2012-09-11 | Sensor path optimisation method and system |
Country Status (1)
| Country | Link |
|---|---|
| GB (1) | GB2505715B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256553A (en) * | 2017-12-20 | 2018-07-06 | 中国人民解放军国防科技大学 | Construction method and device for double-layer path of vehicle-mounted unmanned aerial vehicle |
| CN108681787A (en) * | 2018-04-28 | 2018-10-19 | 南京航空航天大学 | Based on the unmanned plane method for optimizing route for improving the two-way random tree algorithm of Quick Extended |
| CN110647162A (en) * | 2019-10-16 | 2020-01-03 | 厦门理工学院 | Route planning method for tour guide unmanned aerial vehicle, terminal equipment and storage medium |
| CN110856134A (en) * | 2019-10-16 | 2020-02-28 | 东南大学 | Large-scale wireless sensor network data collection method based on unmanned aerial vehicle |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219333B (en) * | 2021-12-20 | 2023-04-07 | 中国空气动力研究与发展中心计算空气动力研究所 | Sensor deployment point planning method and system in three-dimensional terrain and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006020154A1 (en) * | 2004-07-15 | 2006-02-23 | Raytheon Company | System and method for automated search by distributed elements |
| WO2007080584A2 (en) * | 2006-01-11 | 2007-07-19 | Carmel-Haifa University Economic Corp. Ltd. | Uav decision and control system |
| EP1857769A2 (en) * | 2006-05-17 | 2007-11-21 | The Boeing Company | Sensor scan planner |
-
2012
- 2012-09-11 GB GB1216168.3A patent/GB2505715B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006020154A1 (en) * | 2004-07-15 | 2006-02-23 | Raytheon Company | System and method for automated search by distributed elements |
| WO2007080584A2 (en) * | 2006-01-11 | 2007-07-19 | Carmel-Haifa University Economic Corp. Ltd. | Uav decision and control system |
| EP1857769A2 (en) * | 2006-05-17 | 2007-11-21 | The Boeing Company | Sensor scan planner |
Non-Patent Citations (1)
| Title |
|---|
| An information Potential Approach to Integrated Sensor Path Planning and Control, (G.ZHANG, W. LU, S. FERRARI), IEEE Transactions on Robotics, FEBRUARY 28, 2012, http://fred.egr.duke.edu/LISCpapers/TR_ZhangLuFerrariInformationPotential.pdf * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256553A (en) * | 2017-12-20 | 2018-07-06 | 中国人民解放军国防科技大学 | Construction method and device for double-layer path of vehicle-mounted unmanned aerial vehicle |
| CN108256553B (en) * | 2017-12-20 | 2020-03-27 | 中国人民解放军国防科技大学 | Construction method and device for double-layer path of vehicle-mounted unmanned aerial vehicle |
| CN108681787A (en) * | 2018-04-28 | 2018-10-19 | 南京航空航天大学 | Based on the unmanned plane method for optimizing route for improving the two-way random tree algorithm of Quick Extended |
| CN108681787B (en) * | 2018-04-28 | 2021-11-16 | 南京航空航天大学 | Unmanned aerial vehicle path optimization method based on improved bidirectional fast expansion random tree algorithm |
| CN110647162A (en) * | 2019-10-16 | 2020-01-03 | 厦门理工学院 | Route planning method for tour guide unmanned aerial vehicle, terminal equipment and storage medium |
| CN110856134A (en) * | 2019-10-16 | 2020-02-28 | 东南大学 | Large-scale wireless sensor network data collection method based on unmanned aerial vehicle |
| CN110856134B (en) * | 2019-10-16 | 2022-02-11 | 东南大学 | Large-scale wireless sensor network data collection method based on unmanned aerial vehicle |
| CN110647162B (en) * | 2019-10-16 | 2022-10-14 | 厦门理工学院 | Route planning method for tour guide unmanned aerial vehicle, terminal equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| GB201216168D0 (en) | 2012-10-24 |
| GB2505715B (en) | 2015-12-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ayling et al. | New approaches for metagenome assembly with short reads | |
| GB2505715A (en) | Sensor path optimisation method and system | |
| Diniz-Filho et al. | Darwinian shortfalls in biodiversity conservation | |
| JP6182196B2 (en) | Wireless LAN radio map automatic construction method and wireless LAN radio map automatic construction system | |
| König et al. | Simultaneous gene finding in multiple genomes | |
| Parks et al. | Signal, uncertainty, and conflict in phylogenomic data for a diverse lineage of microbial eukaryotes (Diatoms, Bacillariophyta) | |
| Cuevas et al. | A cuckoo search algorithm for multimodal optimization | |
| CN107480470B (en) | Known variation detection method and device based on Bayesian and Poisson distribution test | |
| JP5710572B2 (en) | Apparatus and method for generating new sequence in target genome sequence | |
| CN107403094A (en) | Distribution file is to perform the system and method for anti-virus scan between the virtual machine of distributed system is formed | |
| CN112859912A (en) | Adaptive optimization method and system for unmanned aerial vehicle path planning in relay charging mode | |
| CN112801361A (en) | UAVs and UGVs long-term multi-target path planning problem and solving algorithm | |
| US11536826B2 (en) | Tracking device with deferred activation and propagation of passive tracks | |
| CN111427341B (en) | Robot shortest expected time target searching method based on probability map | |
| CN106022936B (en) | Community structure-based influence maximization algorithm applicable to thesis cooperative network | |
| Freire et al. | Inference of viral quasispecies with a paired de Bruijn graph | |
| CN117550273B (en) | Multi-transfer robot cooperation method based on bee colony algorithm | |
| Mei et al. | Projection-Optimal Monotonic Value Function Factorization in Multi-Agent Reinforcement Learning. | |
| CN104572687B (en) | The key user's recognition methods and device that microblogging is propagated | |
| Hanoun et al. | A hybrid cuckoo search and variable neighborhood descent for single and multiobjective scheduling problems | |
| CN111180005A (en) | Multi-modal protein structure prediction method based on niche resampling | |
| Arafat et al. | Using rough set and ant colony optimization in feature selection | |
| CN113780146B (en) | Hyperspectral image classification method and system based on lightweight neural architecture search | |
| CN115542944A (en) | Multi-unmanned aerial vehicle path planning method based on power distribution network environment and related device | |
| Jaimes | Techniques to deal with many-objective optimization problems using evolutionary algorithms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PCNP | Patent ceased through non-payment of renewal fee |
Effective date: 20170911 |