CN113552881B - Multipath planning data set generation method for neural network training - Google Patents
Multipath planning data set generation method for neural network training Download PDFInfo
- Publication number
- CN113552881B CN113552881B CN202110802240.3A CN202110802240A CN113552881B CN 113552881 B CN113552881 B CN 113552881B CN 202110802240 A CN202110802240 A CN 202110802240A CN 113552881 B CN113552881 B CN 113552881B
- Authority
- CN
- China
- Prior art keywords
- matrix
- points
- point
- map
- generating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 title claims abstract description 29
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 27
- 239000011159 matrix material Substances 0.000 claims abstract description 163
- 238000003491 array Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 238000006467 substitution reaction Methods 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 9
- 125000004122 cyclic group Chemical group 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Abstract
The invention relates to a multi-path planning data set generation method for neural network training, which is characterized in that a random matrix generation algorithm is used for generating a 01 matrix map, the accuracy of the matrix map is improved through secondary circulation traversal, then a starting point and an end point position are set by using a random function, and the multi-path generation is carried out on the basis of a single path generated by a traditional path searching A algorithm by introducing a time dimension, so that collision avoidance is realized. Considering the richness of the data set, respectively manufacturing corresponding data sets according to three conditions of different starting points, different ending points, the same ending points and the same starting points, and finally setting the circulation times to finish the manufacturing of the data sets.
Description
Technical Field
The present invention relates to a path planning method, and in particular, to a method for generating a multipath planning data set based on neural network training.
Background
With the rapid development of modern robots, path planning technology has been widely focused and applied as an important branch and focus in the field of robot research. The existing path planning method comprises a genetic algorithm, an artificial potential field method, a random expansion tree algorithm, a free space method, a grid method and other intelligent heuristic algorithms, and the path planning and the path searching are easy to sink into a local optimal value due to inherent defects of the algorithm, so that a global optimal solution is not obtained.
In recent years, with the great development of artificial intelligence industry and machine computing, deep learning has been widely applied in computer vision, image recognition, speech recognition, image segmentation, natural language processing, etc., and how to utilize neural networks for path planning has become a research hotspot. The success of deep learning depends largely on whether there are a sufficient number of and excellent data set samples to train, but there are currently no large number of data sets unified for robot path planning worldwide available for students to study and evaluate, and the goal of multipath simultaneous prediction can be achieved.
Therefore, it is necessary to propose a method for generating a multipath planning data set for training a neural network, so as to achieve the purpose of path planning of the neural network.
Disclosure of Invention
The invention solves the problems existing in the prior art and provides a multi-path planning data set generation method for neural network training.
The technical scheme adopted by the invention is that the method for generating the multipath planning data set for the neural network training comprises the following steps:
step 1: generating a map matrix only containing 0 and 1 by using a random matrix generation algorithm, wherein the size of the map matrix in the transverse direction and the size of the map matrix in the longitudinal direction are set to be positive integers n, and the probability of 0 generation in the map matrix is p,0< p <1; in the matrix, 1 is represented as an obstacle, 0 is represented as an unobstructed, and the finally generated result matrix is stored as n x n and marked as D1;
step 2: performing secondary traversal on the generated map matrix map by using a circulation method to ensure the accuracy of the data set;
step 3: setting positions of a starting point and an ending point in a generated map matrix by using a random function, respectively constructing a starting point matrix n x n and an ending point matrix n x n which are the same as the map matrix in size, and marking the starting point matrix as D2 and the ending point matrix as D3;
step 4: carrying out single-path generation by combining a traditional A-phase routing algorithm with a starting point matrix, an end point matrix and map matrix information;
step 5: introducing a time dimension to perform multipath generation, performing secondary traversal on the single-path information generated in the step 4 to prevent collision, and obtaining final path matrix information, thereby completing multipath generation and saving the multipath information as n;
step 6: aiming at three conditions of different starting points and ending points, the same ending point and the same starting point, adding corresponding constraint conditions in the step 3, and continuing to carry out the steps 4 and 5 to obtain and store multi-path matrixes under the three conditions, wherein the different starting points and the ending points are marked as D4, the same ending point is marked as D5 and the same starting point is marked as D6;
step 7: and (3) circularly performing M times on the steps 1-6, storing the result of each operation, finally performing dimension integration on the matrixes D1,2 and 3 to obtain three-dimensional input M (n) and n 3, selecting any two-dimensional matrix in the D4,5 and 6 as output M (n) and transmitting the output M (n) and n 1 into a neural network for training according to specific training requirements.
Preferably, in the step 1, in the random matrix generation algorithm in the step 1, a numpy expansion library of python is utilized to perform all-1 matrix generation, then matrix selection is utilized, a part of values of 1 are changed to 0 based on a set probability p generated by 0, and finally, matrix random scrambling is performed through a shuffle function in the numpy library, so as to obtain an initial map matrix. In the step 1, n is a multiple of 5, and p is 0.6;
preferably, in the step 2, the loop method adopts a while loop, and if the condition of the second traversal is not satisfied, the loop is continued, and the second traversal is satisfied by using a matrix block traversal pair of 2x2Matrix of->And meet the following requirementsMatrix of->Performing substitution of matrix elements from 0 to 1 or from 1 to 0; wherein i and j refer to the x-axis coordinate position and the y-axis coordinate position of the matrix, respectively;
preferably, in the step 3, a method for generating a multipath planning data set for training a neural network is provided. The method specifically comprises the following steps:
step 3.1: the number of the starting points and the ending points is determined, the starting points and the ending points are one or more groups, the starting points and the ending points can be set automatically according to training requirements, a single path can be set into one group, and multiple paths are set into multiple groups;
step 3.2: randomly generating one or more different points in the generated matrix map by using a random function, setting the points as starting points, generating a starting point matrix with the same size as the map matrix, setting the position of each starting point as 1, setting the other positions as 0, storing the matrix as n x n, and marking the matrix as D2;
step 3.3: creating a matrix which is identical to the map matrix, and setting the corresponding position values of the one or more different starting points in the matrix to be 1 to obtain a matrix N;
step 3.4: randomly generating one or more different points in the matrix N generated in the step 3.3 by using a random function, setting the points as end points, generating a matrix with the same size as a map matrix, setting the positions of the one or more different end points as 1, setting the other positions as 0, storing the matrix as N x N, and marking the matrix as D3;
preferably, in the step 4, a method for generating a multipath planning data set for training a neural network is provided. The method specifically comprises the following steps:
step 4.1: analyzing and running the map matrix information, any starting point and corresponding end point information by using an A-algorithm to generate a single path;
step 4.2: storing the position information of each path point under a single path into an array;
step 4.3: returning to the step 4.1 to select another group of starting points and ending points until all starting points and ending points under the current map matrix find corresponding single paths between the starting points and the ending points and store the single paths into the corresponding array;
preferably, in the step 5, a method for generating a multipath planning data set for training a neural network is provided. The method specifically comprises the following steps:
step 5.1: introducing a time dimension, and performing secondary traversal comparison on all the arrays stored in the single path generated in the step 4.3 under the condition of uniform speed or variable speed;
step 5.2: if two or more single-path position information overlap states that collision occurs at the same moment, returning to the step 3 to select the starting point and the end point again;
step 5.3: generating a matrix with the same size as a matrix map until all the single-path information does not overlap at any time under the condition of a specified speed, setting the generated position information corresponding to all the single paths as 1 in the matrix, setting other positions as 0, and storing the matrix as n;
in the step 5, the time dimension is introduced, the array length intercepted by traversing the single-path information is changed, collision avoidance at different speeds can be set, the aim of speed change is realized, and the method is more suitable for actual application scenes.
Preferably, in the step 6, for better extending the type and practicality of the data set, in the step 6, considering three cases of different starting points and ending points, the same ending point and the same starting point, constraints are added in the step 3: randomly generating an end point and a plurality of start points for the same end point; randomly generating a starting point and a plurality of end points for the same starting point; aiming at different starting points and end points, each point is ensured not to be repeated;
preferably, in step 7, the cyclic generation is performed in steps 1 to 6, the cyclic number is set to M, the result of each operation is stored, the matrices of D1,2 and 3 are dimensionally integrated to obtain a three-dimensional input M (n×n×3), wherein one-dimensional n×n×1 corresponds to the matrix of D1, two-dimensional n×n×2 corresponds to the matrix of D2, three-dimensional n×n×3 corresponds to the matrix of D3, any two-dimensional matrix of D4,5 and 6 is taken as output M (n×n), and is transmitted into the neural network for training, and the cycle is ended to obtain M (n×n×3) matrices as inputs and M (n×n) matrices as outputs.
The invention provides a multi-path planning data set generation method for neural network training, which is characterized in that a random matrix generation algorithm is used for generating a 01 matrix map, the accuracy of the matrix map is improved through secondary circulation traversal, and then a starting point and an end point are set by using a random function. And introducing a time dimension to carry out multipath generation on the basis of a single path generated by a traditional path finding A-x algorithm, so as to realize collision avoidance. Considering the richness of the data set, respectively manufacturing corresponding data sets according to three conditions of different starting points, different ending points, the same ending points and the same starting points, and finally setting the circulation times to finish the manufacturing of the data sets.
Drawings
Fig. 1 is a matrix corresponding to the behavior of the through obstacle according to the present invention.
Fig. 2 is a flow chart of a multi-path planning data set generation method of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples, but the scope of the present invention is not limited thereto.
The invention relates to a multi-path planning data set generation method for neural network training, which comprises the following steps.
Step 1: generating a map matrix only containing 0 and 1 by using a random matrix generation algorithm, wherein the size of the map matrix in the transverse direction and the size of the map matrix in the longitudinal direction are set to be positive integers n, and the probability of 0 generation in the map matrix is p,0< p <1; in the matrix, 1 is represented as an obstacle, 0 is represented as an unobstructed, and the finally generated result matrix is stored as n x n and marked as D1;
in the step 1, the random matrix generation algorithm firstly uses the numpy expansion library of python to generate a matrix with the total number of 1, then uses matrix selection, based on the set probability p generated by 0, makes a part of values with 1 changed into 0, and finally uses the shuffle function in the numpy expansion library to randomly shuffle the matrix to obtain the initial map matrix. In the step 1, n is a multiple of 5, and p is 0.6;
step 2: performing secondary traversal on the generated map matrix map by using a circulation method to ensure the accuracy of the data set;
in the step 2, the loop method adopts a while loop, and if the condition of the secondary traversal is not satisfied, the loop is continued, and the secondary traversal is that the matrix block traversal pair of 2x2 satisfiesMatrix of->Satisfy-> Matrix Ii, je of matrix elements from 0 to 1 or from 1 to 0; wherein i and j refer to the x-axis coordinate position and the y-axis coordinate position of the matrix, respectively;
step 3: setting positions of a starting point and an ending point in a generated map matrix by using a random function, respectively constructing a starting point matrix n x n and an ending point matrix n x n which are the same as the map matrix in size, and marking the starting point matrix as D2 and the ending point matrix as D3;
the step 3 comprises the following steps:
step 3.1: the number of the starting points and the ending points is determined, the starting points and the ending points are one or more groups, the starting points and the ending points can be set automatically according to training requirements, a single path can be set into one group, and multiple paths are set into multiple groups;
step 3.2: randomly generating one or more different points in the generated matrix map by using a random function, setting the points as starting points, generating a starting point matrix with the same size as the map matrix, setting the position of each starting point as 1, setting the other positions as 0, storing the matrix as n x n, and marking the matrix as D2;
step 3.3: creating a matrix which is identical to the map matrix, and setting the corresponding position values of the one or more different starting points in the matrix to be 1 to obtain a matrix N;
step 3.4: randomly generating one or more different points in the matrix N generated in the step 3.3 by using a random function, setting the points as end points, generating a matrix with the same size as a map matrix, setting the positions of the one or more different end points as 1, setting the other positions as 0, storing the matrix as N x N, and marking the matrix as D3;
step 4: carrying out single-path generation by combining a traditional A-phase routing algorithm with a starting point matrix, an end point matrix and map matrix information;
the step 4 includes the following steps:
step 4.1: analyzing and running the map matrix information, any starting point and corresponding end point information by using a path searching algorithm A to generate a single path;
step 4.2: storing the position information of each path point under a single path into an array;
step 4.3: returning to the step 4.1 to select another group of starting points and ending points until all starting points and ending points under the current map matrix find corresponding single paths between the starting points and the ending points and store the single paths into the corresponding array;
step 5: introducing a time dimension to perform multipath generation, performing secondary traversal on the path information generated in the step 4 to prevent collision, and obtaining final path matrix information, thereby completing multipath generation and saving the path information as n;
the step 5 comprises the following steps:
step 5.1: introducing a time dimension, and performing secondary traversal comparison on all the arrays stored in the single path generated in the step 4.3 under the condition of uniform speed or variable speed;
step 5.2: if two or more single-path position information overlap at the same moment, indicating that collision occurs, and returning to the step 3 to select the starting point and the end point again;
step 5.3: generating a matrix with the same size as a matrix map until all the single-path information does not overlap at any time under the condition of a specified speed, setting the generated position information corresponding to all the single paths as 1 in the matrix, setting other positions as 0, and storing the matrix as n;
in the step 5, the time dimension is introduced, the array length intercepted by traversing the single-path information is changed, collision avoidance at different speeds can be set, the aim of speed change is realized, and the method is more suitable for actual application scenes.
Step 6: aiming at three conditions of different starting points and ending points, the same ending point and the same starting point, adding corresponding constraint conditions in the step 3, and continuing to carry out the steps 4 and 5 to obtain and store multi-path matrixes under the three conditions, wherein the different starting points and the ending points are marked as D4, the same ending point is marked as D5 and the same starting point is marked as D6;
in the step 6, in order to better expand the type and the practicability of the data set, in the step 6, considering three situations of different starting points and ending points, the same ending point and the same starting point, adding a constraint in the step 3: randomly generating an end point and a plurality of start points for the same end point; randomly generating a starting point and a plurality of end points for the same starting point; different starting points and end points ensure that each point is not repeated;
step 7: and (3) circularly performing M times on the steps 1-6, storing the result of each operation, and finally performing dimension integration on the matrixes D1,2 and 3 to obtain three-dimensional input M (n) and transmitting a group of two-dimensional matrixes D4,5 and 6 as output M (n) to a neural network for training according to specific training requirements.
In the step 7, the cyclic generation is performed in the steps 1 to 6, the cyclic times are set to N times, the result of each operation is stored, the matrices of D1,2 and 3 are dimensionally integrated to obtain a three-dimensional input M (n×n×3), wherein one-dimensional n×n×1 corresponds to the matrix of D1, two-dimensional n×n×2 corresponds to the matrix of D2, three-dimensional n×3 corresponds to the matrix of D3, any two-dimensional matrix in D4,5 and 6 is taken as an output m×n, and is transmitted into the neural network for training, and the cycle is ended to obtain M (n×n×3) matrices as inputs and M (n×n) matrices as outputs.
Claims (7)
1. A method for generating a multipath planning data set for neural network training, characterized by: the method comprises the following steps:
step 1: generating a map matrix only containing 0 and 1 by using a random matrix generation algorithm, wherein the size of the map matrix in the transverse direction and the longitudinal direction is set to be a positive integer n, the probability of 0 generation in the map matrix is p,0< p <1, 1 represents an obstacle in the matrix, 0 represents no obstacle, and the finally generated matrix is stored as n x n and marked as D1;
step 2: performing secondary traversal on the generated map matrix by using a circulation method to ensure the accuracy of the data set;
step 3: setting positions of a starting point and an ending point in a generated map matrix by using a random function, respectively constructing a starting point matrix n x n and an ending point matrix n x n which are the same as the map matrix in size, and marking the starting point matrix as D2 and the ending point matrix as D3;
step 4: the method for generating the single path by combining the traditional A-phase routing algorithm with the starting point matrix, the end point matrix and the map matrix information comprises the following steps:
step 4.1: analyzing and running the map matrix information, any starting point and corresponding end point information by using a path searching algorithm A to generate a single path;
step 4.2: storing the position information of each path point under a single path into an array;
step 4.3: returning to the step 4.1 to select another group of starting points and ending points until all starting points and ending points under the current map matrix find corresponding single paths between the starting points and the ending points and store the single paths into the corresponding array;
step 5: introducing a time dimension to perform multipath generation, performing secondary traversal on the path information of the single path generated in the step 4 to prevent collision, and obtaining final path matrix information, thereby completing multipath generation and saving the path information as n x n, and comprising the following steps:
step 5.1: introducing a time dimension, and performing secondary traversal comparison on all the arrays stored in the single path generated in the step 4.3 under the condition of uniform speed or variable speed;
step 5.2: if two or more single-path position information overlap at the same moment, indicating that collision occurs, and returning to the step 3 to select the starting point and the end point again; step 5.3: generating a matrix with the same size as a matrix map until all the single-path information does not overlap at any time under the condition of a specified speed, setting the generated position information corresponding to all the single paths as 1 in the matrix, setting other positions as 0, and storing the matrix as n;
step 6: aiming at three conditions of different starting points and ending points, the same ending point and the same starting point, adding corresponding constraint conditions in the step 3, and continuing to carry out the steps 4 and 5 to obtain and store multi-path matrixes under the three conditions, wherein the different starting points and the ending points are marked as D4, the same ending point is marked as D5 and the same starting point is marked as D6;
step 7: and (3) circularly performing M times on the steps 1-6, storing the result of each operation, finally performing dimension integration on the matrixes D1, D2 and D3 to obtain three-dimensional input M (n) and taking any two-dimensional matrix in the matrixes D4, D5 and D6 as output M (n) and transmitting the output M (n) and the output M (n) to a neural network for training.
2. A method of generating a multipath planning data set for neural network training as claimed in claim 1, wherein: in the step 1, the random matrix generation algorithm firstly uses the numpy expansion library of python to generate a matrix with the total number of 1, then uses matrix selection, based on the set probability p generated by 0, makes a part of values with 1 changed into 0, and finally uses the shuffle function in the numpy expansion library to randomly shuffle the matrix to obtain the initial map matrix.
3. A method of generating a multipath planning data set for neural network training as claimed in claim 1, wherein: in the step 1, n is a multiple of 5, and p is 0.6.
4. A method of generating a multipath planning data set for neural network training as claimed in claim 1, wherein: in the step 2, the second traversal is that the second traversal is performed by 2x2Matrix block traversal, pair of satisfaction Matrix of->And meet the following requirements Matrix of->The substitution of matrix elements from 0 to 1 or from 1 to 0 is performed, where i and j refer to the x-axis coordinate position and the y-axis coordinate position of the matrix, respectively.
5. A method of generating a multipath planning data set for neural network training as claimed in claim 1, wherein: the step 3 comprises the following steps:
step 3.1: determining the number of starting points and end points, wherein the number of the starting points and the end points is one or more;
step 3.2: randomly generating one or more different points in the generated matrix map by using a random function, setting the points as starting points, generating a starting point matrix with the same size as the map matrix, setting the position of each starting point as 1, setting the other positions as 0, storing the starting point matrix as n x n, and marking the starting point matrix as D2;
step 3.3: creating a matrix which is identical to the map matrix, and setting the corresponding position values of the one or more different starting points in the matrix to be 1 to obtain a matrix N;
step 3.4: randomly generating one or more different points in the matrix N generated in the step 3.3 by using a random function, setting the points as end points, generating an end point matrix with the same size as the map matrix, setting the position of each end point as 1, setting the other positions as 0, storing the end point matrix as N x N, and marking the end point matrix as D3.
6. A method of generating a multipath planning data set for neural network training as claimed in claim 1, wherein: in the step 6, for three cases of different starting points and ending points, the same ending point and the same starting point, adding constraints in the step 3;
randomly generating an end point and a plurality of start points for the same end point;
randomly generating a starting point and a plurality of end points for the same starting point;
for different starting points and end points, each point is guaranteed not to be repeated.
7. A method of generating a multipath planning data set for neural network training as claimed in claim 1, wherein: in the step 7, the result of each operation is stored, the matrices of D1,2 and 3 are dimensionally integrated to obtain a three-dimensional input M (n×n×3), wherein one-dimensional n×n×1 corresponds to the matrix of D1, two-dimensional n×n×2 corresponds to the matrix of D2, three-dimensional n×n×3 corresponds to the matrix of D3, any two-dimensional matrix of D4,5 and 6 is taken as output M (n×n) and transmitted into the neural network for training, and the cycle is ended to obtain M (n×n×3) matrices as input and M (n×n) matrices as output.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110802240.3A CN113552881B (en) | 2021-07-15 | 2021-07-15 | Multipath planning data set generation method for neural network training |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110802240.3A CN113552881B (en) | 2021-07-15 | 2021-07-15 | Multipath planning data set generation method for neural network training |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113552881A CN113552881A (en) | 2021-10-26 |
| CN113552881B true CN113552881B (en) | 2024-03-26 |
Family
ID=78131918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110802240.3A Active CN113552881B (en) | 2021-07-15 | 2021-07-15 | Multipath planning data set generation method for neural network training |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113552881B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114779780B (en) * | 2022-04-26 | 2023-05-12 | 四川大学 | Path planning method and system in random environment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103402235A (en) * | 2013-07-22 | 2013-11-20 | 上海交通大学 | Directional antenna-based multi-rate multi-path route optimization method |
| CN109443363A (en) * | 2018-11-09 | 2019-03-08 | 厦门大学 | Certainty of dividing and ruling path optimizing algorithm |
| CN110333739A (en) * | 2019-08-21 | 2019-10-15 | 哈尔滨工程大学 | A kind of AUV conduct programming and method of controlling operation based on intensified learning |
| CN110345960A (en) * | 2019-06-13 | 2019-10-18 | 福建工程学院 | Route planning intelligent optimization method for avoiding traffic obstacles |
| CN113093787A (en) * | 2021-03-15 | 2021-07-09 | 西北工业大学 | Unmanned aerial vehicle trajectory planning method based on velocity field |
-
2021
- 2021-07-15 CN CN202110802240.3A patent/CN113552881B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103402235A (en) * | 2013-07-22 | 2013-11-20 | 上海交通大学 | Directional antenna-based multi-rate multi-path route optimization method |
| CN109443363A (en) * | 2018-11-09 | 2019-03-08 | 厦门大学 | Certainty of dividing and ruling path optimizing algorithm |
| CN110345960A (en) * | 2019-06-13 | 2019-10-18 | 福建工程学院 | Route planning intelligent optimization method for avoiding traffic obstacles |
| CN110333739A (en) * | 2019-08-21 | 2019-10-15 | 哈尔滨工程大学 | A kind of AUV conduct programming and method of controlling operation based on intensified learning |
| JP2021034050A (en) * | 2019-08-21 | 2021-03-01 | 哈爾浜工程大学 | Auv action plan and operation control method based on reinforcement learning |
| CN113093787A (en) * | 2021-03-15 | 2021-07-09 | 西北工业大学 | Unmanned aerial vehicle trajectory planning method based on velocity field |
Non-Patent Citations (1)
| Title |
|---|
| 基于激光雷达的多路径地形匹配算法;孟海东;王祺;慕春棣;陈丽萍;;计算机仿真(第02期);第69页-71页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113552881A (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110929873A (en) | Quantum program processing method and device, storage medium and electronic device | |
| CN108334805A (en) | The method and apparatus for detecting file reading sequences | |
| CN113552881B (en) | Multipath planning data set generation method for neural network training | |
| CN105320764A (en) | 3D model retrieval method and 3D model retrieval apparatus based on slow increment features | |
| CN106202224B (en) | Search processing method and device | |
| KR102305575B1 (en) | Method and system for highlighting similar areas using similarity between images | |
| CN112115879A (en) | Shielding-sensitive self-supervision pedestrian re-identification method and system | |
| CN112464611A (en) | Automatic PCB wiring system based on cloud-end collaborative intelligent processing | |
| CN110264392B (en) | Strong connection graph detection method based on multiple GPUs | |
| CN112465123A (en) | Processing device and method for optimizing neural network model | |
| Chen et al. | Neuroevolution of augmenting topologies with learning for data classification | |
| CN113886226B (en) | Test data generation method of confrontation generation model based on twin network | |
| CN115861841A (en) | SAR image target detection method combined with lightweight large convolution kernel | |
| CN113658236B (en) | Incomplete point cloud registration method based on graph attention mechanism | |
| CN104462503A (en) | Method for determining similarity between data points | |
| Kaur et al. | On solving partition driven standard cell placement problem using firefly-based metaheuristic approach | |
| CN114969279A (en) | Table text question-answering method based on hierarchical graph neural network | |
| CN110309139B (en) | High-dimensional neighbor pair searching method and system | |
| Xu et al. | A novel object detection algorithm based on enhanced R-FCN and SVM | |
| CN114833461A (en) | Free steering method and sequencing device for non-closed contour of laser cutting path | |
| Lim et al. | Analyzing deep neural networks with noisy labels | |
| CN113761243A (en) | Online retrieval method and system | |
| Nallaperuma et al. | Parameterized complexity analysis and more effective construction methods for ACO algorithms and the euclidean traveling salesperson problem | |
| Haojun et al. | A collective entity linking method based on graph embedding algorithm | |
| CN110474844B (en) | Training method and chip for learning type index data structure of high-performance intelligent router |
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 |