LU102400B1 - Path planning method and system for unmanned surface vehicle based on improved genetic algorithm - Google Patents
Path planning method and system for unmanned surface vehicle based on improved genetic algorithm Download PDFInfo
- Publication number
- LU102400B1 LU102400B1 LU102400A LU102400A LU102400B1 LU 102400 B1 LU102400 B1 LU 102400B1 LU 102400 A LU102400 A LU 102400A LU 102400 A LU102400 A LU 102400A LU 102400 B1 LU102400 B1 LU 102400B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- unmanned surface
- surface vehicle
- genetic algorithm
- path planning
- population
- Prior art date
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 156
- 230000002068 genetic effect Effects 0.000 title claims abstract description 92
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 210000000349 chromosome Anatomy 0.000 claims description 59
- 230000035772 mutation Effects 0.000 claims description 26
- 230000008569 process Effects 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 239000012634 fragment Substances 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 5
- 230000014759 maintenance of location Effects 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 28
- 238000004364 calculation method Methods 0.000 description 14
- 238000009826 distribution Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 7
- 108090000623 proteins and genes Proteins 0.000 description 7
- 238000005457 optimization Methods 0.000 description 6
- 238000000342 Monte Carlo simulation Methods 0.000 description 3
- 230000000717 retained effect Effects 0.000 description 3
- 241000180579 Arca Species 0.000 description 2
- 108700003861 Dominant Genes Proteins 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 241000894007 species Species 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 239000004106 carminic acid Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/0206—Control of position or course in two dimensions specially adapted to water vehicles
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T70/00—Maritime or waterways transport
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a path planning method and system for an unmanned surface vehicle based on an improved genetic algorithm, which perform path planning on the unmanned surface vehicle by means of the improved genetic algorithm. The method comprises the following steps: acquiring course data and location data of the unmanned surface vehicle, and preprocessing same: collecting wave information of an environment where the unmanned surface vehicle is located, and converting same into constraint factors; performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle, to obtain an optimal path ranking; and correcting, based on the optimal path ranking, the course and speed of the unmanned surface vehicle according to the constraint factors, to complete the path planning.
Description
PATH PLANNING METHOD AND SYSTEM FOR UNMANNED SURFACE 0102400
VEHICLE BASED ON IMPROVED GENETIC ALGORITHM Field of the Invention The present disclosure relates to the field of unmanned surface vehicle control technology, and specifically to à path planning method and system for an unmanned surface vehicle based on an improved genetic algorithm. Background of the Invention The traveling salesman problem (TSP) is a typical NP-hard problem. Its purpose is to plan a shortest path for a traveling salesman to return to a starting city after going through each city once. In production and life, TSP models have been widely applied in many fields, such as vehicle path planning, machine learning. time charts. word sense disambiguation, green logistics, fuel efficiency management. and wireless charging. Therefore, solving the TSP problem is of great significance for houschold. civilian and military applications. In recent years. the research on the method of solving the TSP problem mainly tends to heuristic algorithms with adaptive control ideas, such as a genetic algorithm (GA), a simulated anncaling algorithm. an ant colony algorithm and a neural network algorithm. In contrast, GA has higher robustness and stronger global search capability, so it has been applied to the track planning of various autonomous devices such as robots. unmanned acrial vehicles (UAVs). and unmanned surface vehicles (USVs). In order to solve the problem of collision-free shortest path planning for intelligent robots, the existing methods are as follows: (1) an obstacle-based genetic algorithm is used to narrow a search arca to obtain a path with short length and low time cost. (2) A traditional genctie algorithm (CGA) applied to a mobile robot is improved to find control points of a Bezier curve and design a shortest path in a dynamic working place. (3) A parallel genetic algorithm 1s applied to a multi-UAV system in a multi-core environment, and the Bezier curve further smoothes the preliminary planned path to gencrate a final flight track. (4) GA is improved by a new cvolutionary operator for multiple UAVs. and considering the constraints of a threc-dimensional environment. the information collected from the required arca is maximized to obtain a favorable route. (5) As far as USV is concerned. three objective functions of avoiding obstacles, reaching a goal and reducing travel time are combined to evaluate the applicability of a path under a marine environmental load.
LU102400 In order to overcome the inherent problems of slow convergence, poor local search capability, premature convergence. ete. of the CGA algorithm. the performance of the algorithm is improved by means of a combination of two or more optimization algorithms based on biological evolution and mathematical ecology theory.
For example: (1) the crossover operator is improved to produce more offspring, thereby enriching the diversity of populations.
Through the test of several TSP instances. it is proved that this method has fast convergence and achieves a better planning path value than CGA. (2) With multiple goals of minimizing the fuzzy cost and fuzzy time of the total travel. the combination of ant colony optimization and genetic algorithm solves four-dimensional inaccurate TSP problems including source, destination. transportation and route. (3) In a centralized UAV layout strategy. considering the location of a ground node. an optimal value of a UAV is designed by means of an elite non-dominant sorting genetic algorithm. (4) The application of a dynamic planning navigation algorithm based on a genetic algorithm to the autonomous navigation ol a mobile ground robot in an unknown dynamic environment has better robustness and effectiveness. (5) In order to solve the problem of group trading strategy combinations in the securities market, a grouped genetic algorithm (GGA) is proposed, and its fitness function is calculated by group balance, weight balance, portfolio return and risk.
The existing USV path planning methods include traditional means such as a free space method. an artificial potential field method. and a visibility graph method, as well as intelligent optimization algorithms that have emerged with the development of artificial intelligence. such as an ant colony algorithm, a particle swarm algorithm, and a genetic algorithm.
During the research and development process, the inventors found that these algorithms have some shortcomings when applied to the path planning of unmanned surface vehicles: the free space method is difficult to apply to multi-dimensional path planning problems such as path planning of unmanned surface vehicles: the artificial potential field method and the particle swarm algorithm are prone to problems such as unreachable targets. falling into local optima. and low efficiency. causing self-crossing phenomena in unmanned surface vehicles; the visibility graph method lacks flexibility and has problems such as combinatorial explosions, while the ant colony algorithm has a large amount of calculation, and the two algorithms are time-consuming and cannot meet the timeliness requirements of
USV path planning. Although the traditional genetic algorithm cannot find the global 0102600 optimal value due to the premature phenomenon, its good parallelism and efficient scarch ability meet the needs of unmanned surface vehicles in path planning. Summary of the Invention In order to overcome the above shortcomings of the prior art. the present disclosure provides a path planning method and system for an unmanned surface vehicle based on an improved genetic algorithm. which plan a path for the unmanned surface vehicle by means of the improved genetic algorithm.
One aspect of the present disclosure provides a technical solution of a path planning method for an unmanned surface vehicle based on an improved genetic algorithm: A path planning method for an unmanned surface vehicle based on an improved genetic algorithm. including the following steps: acquiring course data and location data of the unmanned surface vehicle. and preprocessing same: collecting wave information of an environment where the unmanned surface vehicle is located. and converting same into constraint factors; performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle. to obtain an optimal path ranking: and correcting. based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factors, to complete the path planning. Another aspect of the present disclosure provides a technical solution of a path planning system for an unmanned surface vehicle based on an improved genetic algorithm: A path planning system for an unmanned surface vehicle based on an improved genetic algorithm. including: a navigation data acquisition module, configured to acquire course data and location data of the unmanned surface vehicle, and preprocess same: constraint factor determination module. configured to collect wave information of an environment where the unmanned surface vehicle is located, and convert same into constraint factors: an optimal path planning module. configured to perform path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle, to obtain an optimal path ranking: and LU102400 a track correction module, configured to correct, based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factors. to complete the path planning.
Another aspect of the present disclosure provides a technical solution of a computer-readable storage medium: A computer-readable storage medium. storing a computer program that. when executed by a processor, the following steps are implemented: acquiring course data and location data of the unmanned surface vehicle. and preprocessing same: collecting sca wave information of an environment where the unmanned surface vehicle is located. and converting same into constraint factors: performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle. to obtain an optimal path ranking: and correcting, based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factors. to complete the path planning.
Another aspect of the present disclosure provides a technical solution of a processing apparatus:
A processing apparatus, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program. the following steps are implemented: acquiring course data and location data of the unmanned surface vehicle. and preprocessing same:
collecting wave information of an environment where the unmanned surface vehicle is located, and converting same into constraint factors; performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle. to obtain an optimal path ranking: and correcting. based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factors, to complete the path planning.
Through the above technical solutions, the beneficial effects of the present disclosure are:
(1) The present disclosure uses a double-domain inversion based genetic algorithm and a multi-domain inversion based genetic algorithm for path planning of the HU102400 unmanned surface vehicle, to gencrate a feasible path with short length and no sell=intersection. which realizes the control and track correction of the unmanned surface vehicle; 5 (2) The double-domain inversion based genetic algorithm and the multi-domain inversion based genetic algorithm according to the present disclosure greatly reduce the calculation time cost. improve the robustness of the algorithms, and obtain a reasonable path that is more stable and timely and conforms to the path planning requirements of the unmanned surface vehicle.
Brief Description of the Drawings The accompanying drawings constituting a part of the present disclosure are intended to provide a further understanding of the present disclosure. and the illustrative embodiments of the present disclosure and the descriptions thercof are intended to interpret the present disclosure and do not constitute improper limitations to the present disclosure. Fig. 1 is a flowchart of a path planning method for an unmanned surface vehicle based on an improved genetic algorithm in Embodiment 1: Fig. 2 is a flowchart of a genetic algorithm in Embodiment 1: Fig. 3 is a schematic diagram of crossover in Embodiment 1; Fig. 4 is a schematic diagram of mutation in Embodiment 1: Fig. 5 is a schematic diagram of single-domain inversion in Embodiment 1; Fig. 6 is a schematic diagram of double-domain inversion in Embodiment 1; Fig. 7 is a schematic diagram of multi-domain inversion in Embodiment 1: Fig. 8(a) is a solution distribution diagram of each algorithm under P=14 planning points in Embodiment 1: Fig. 8(b) is a solution distribution diagram of each algorithm under P=22 planning points in Embodiment 1; Fig. 8(c) is a solution distribution diagram of cach algorithm under P=51 planning points in Embodiment 1: Fig. 8(d) is a solution distribution diagram ol cach algorithm under P=76 planning points in Embodiment 1: Fig. 8(e) is a solution distribution diagram of cach algorithm under P=99 planning points in Embodiment 1:
Fig. 9(a) is a solution distribution diagram of each algorithm under 5-20 population HUT02400 in Embodiment 1: Fig. 9(b) is a solution distribution diagram of each algorithm under S=40 population in Lmbodiment 1; Fig. 9(c) is a solution distribution diagram of each algorithm under S=60 population in Embodiment 1; Fig. 9(d) is a solution distribution diagram of each algorithm under S=80 population in Embodiment 1; Fig. 9(c) is a solution distribution diagram of cach algorithm under S=100 population in Embodiment 1: Fig. 10 is an optimal track diagram of five TSPLIB instances in Embodiment 1; Fig. 11 is a schematic diagram of a double-domain inversion based genetic algorithm in Embodiment 1; Fig. 12 is a schematic diagram of a multi-domain inversion based genetic algorithm in Embodiment I: Fig. 13(a) is a convergence curve diagram of each algorithm under planning points P=15 in Embodiment 1: Iig. 13(b) is a convergence curve diagram of each algorithm under planning points P=25 in Embodiment 1: Fig. 13(c) is a convergence curve diagram of cach algorithm under planning points P=35 in Embodiment 1; Fig. 13(d) is a convergence curve diagram of each algorithm under planning points P=45 in Embodiment 1; Fig. 14 is a track diagram of each algorithm under planning points P=15 in Embodiment 1: Fig. 15 is a track diagram of each algorithm under planning points P=25 in Embodiment 1: Fig. 16 is a track diagram of cach algorithm under planning points P=35 in Embodiment 1: Fig. 17 is a track diagram of each algorithm under planning points P=45 in Embodiment 1. Detailed Description of Embodiments The present disclosure will be further illustrated below in conjunction with the
„ accompanying drawings and embodiments. LU102400 It should be noted that the following detailed descriptions are exemplary and are intended to provide further descriptions of the present disclosure. All technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical filed to which the present disclosure belongs. unless otherwise indicated. It should be noted that the terms used here are merely used for describing specific embodiments, but are not intended to limit the exemplary embodiments of the present application. As used herein. the singular form is also intended to comprise the plural form unless otherwise indicated in the context. In addition. it should be understood that when the terms “contain” and/or “comprise” are used in the description, they are intended to indicate the presence of features. steps, operations. devices, components and/or combinations thereof.
Glossary: (1) CGA, traditional genetic algorithm: (2) SDIGA, single-domain inversion based genetic algorithm; (3) DDIGA, double-domain inversion based genetic algorithm: (4) MDIGA. multi-domain inversion based genetic algorithm, Embodiment | This embodiment provides a path planning method for an unmanned surface vehicle based on an improved genetic algorithm. Referring to Fig. 1. the method includes the following steps: S101. course data and location data of the unmanned surface vehicle are acquired and preprocessed.
Specifically. longitude and latitude coordinates of a navigation point required by the unmanned surface vehicle are acquired through GPS and an electronic compass. and the longitude and latitude coordinates of the navigation point are converted into abscissa and ordinate values in a rectangular coordinate system. $102. weather and wave information of an environment where the unmanned surface vehicle is located is acquired and converted into constraint factors.
Specifically, the weather and wave information, including height. velocity and wavelength of waves. of the environment where the unmanned surface vehicle is located is acquired by an ultrasonic weather sensor; and the weather and wave information is converted into constraint factors for track correction of the unmanned surface vehicle. LU102400 The constraint factors are mainly the force of waves. Since the force borne by a ship in water is mainly the force of waves, a wave force model is used as the constraint factors to correct track deflection. The functions of the constraint factors are: Fy=-0.0073V"hV;" +0.0057V/ , My = 2 Anh ne.
3 Where, h is the height of waves, Vi is the velocity of waves, and Ao 1s the wavelength of waves. $103. path planning is performed by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle, to obtain an optimal path ranking.
In this embodiment, the improved genetic algorithm includes a double-domain inversion based genetic algorithm and a multi-domain inversion based genetic algorithm. The double-domain inversion based genetic algorithm (DDIGA) involves two inversion operations between four inversion points randomly ranked. In addition, the permutation and combination of the four inversion point sequences also increase the number of inversion domains. The multi-domain inversion based genetic algorithm (MDIGA), due to the significant increase in offspring. only retains the most suitable inverse chromosomes and transfers same to a new generation, which improves local search capability.
Fig. 2 shows a calculation process of a canonical genetic algorithm (CGA). A method of real number encoding is chosen. and a string with serial numbers of visited cities is uscd to represent cach chromosome. Genetic parameters, such as population size and crossover and mutation probability, are generally defined based on experience. After an optimization problem is determined. an initial population of candidate solutions with a certain scale is randomly generated. The fitness function is 1/len (len represents the relative path length of cach chromosome), which is used to evaluate the fitness of cach individual. wherein more suitable individuals will survive in reproduction. Then, the algorithm increases the population quantity by means of iterative operations of crossover. mutation and selection. If a certain standard is satisticd or a maximum number of iterations is reached, the evolution process ends.
In CGA. crossover is mainly used to connect two parent chromosomes, and these chromosomes are separated by certain breakpoints and produce two offspring with certain crossover probability (Pc). Mutation is mainly used to exchange gene positions LU102400 on two randomly selected mutation points on the chromosome, and the occurrence of mutation has certain mutation probability (PM). It should be pointed out that the crossover causes the chromosomes to be similar and helps the convergence of the population: while the mutation increases the genetic diversity, so that the algorithm can further expand the population number under the condition of local optimization. This embodiment proposes a single-domain inversion based genetic algorithm (SDIGA). and a further inversion operation is added after the CGA mutation. Two different genes on a chromosome are defined as inversion sites, and the segment between the two genes is named an inversion domain. Then, the fragment is flipped 180 degrees (inversely) and inserted into the original position of the chromosome. The schematic diagrams of crossover. mutation and single-domain inversion arc respectively shown in Fig. 3 to Fig. 5.
Specifically. the double-domain inversion based genetic algorithm (DDIGA) is: In CGA. symbolic coding is usually used as chromosome coding, and crossover operators of partial mapping crossover (PMC) arc used to solve TSP. However. this crossover operation will cause serious damage to the parent chromosome. Only a small part of parental genes can survive. The genes of most offspring chromosomes arc generated during evolution, which is not conducive to inheriting the dominant genes from the parent chromosome. In addition, due to the limited conversion of genes. the mutation or single-domain inversion has obvious shortcomings in local search capability. Therefore, a double-domain inversion based genetic algorithm (DDIGA) is designed, as shown in Fig. 6.
The positions of four different genes are randomly defined as inversion points of a chromosome coding string. Two domains are respectively generated between the first two points and the last two points, and the fragments in the two domains are inversed at the same time to reproduce offspring. The fitnesses of offspring chromosomes and a parent chromosome are compared to determine next-generation more suitable chromosomes. The double-domain inversion is shown in Fig. 6. where | represents a parent chromosome, and l’ represents an inverted offspring chromosome.
The double-domain inversion based genetic algorithm designed in this embodiment helps to retain more dominant genes from the parent chromosome by means of double-domain inversion, and generates more adaptive coded strings for the offspring chromosomes. In addition. since reasonable fitness can ensure that the offspring evolve 10 a higher level, the capability of local scarch may be improved. LU102400 Specifically. the multi-domain inversion based genetic algorithm (MDIGA) is: In CGA. the number of offspring produced is usually the same as the number of parent chromosomes. From the basis of biological theory, the number of offspring should be greater than the number of parents to prevent species extinction and maintain species diversity in the process of biological evolution. The four randomly ranked points in the double-domain inversion based genetic algorithm create two domains for DDIGA. and only one offspring chromosome is generated after two inversions, but in fact, every two of the four inversion points can define an inversion domain. According to the permutation and combination theory, there are six domains in one inversion. Therefore, six additional offspring chromosomes will be replicated by means of single inversion of each domain in the parent chromosome, which will increase the possibility of searching more suitable offspring for cach generation to a certain extent. Accordingly, this embodiment designs a multi-domain inversion based genetic algorithm to increase the number of inversion domains and offspring chromosomes. As shown in Fig. 7, four inversion points are randomly defined in a coded string. called a-d, and six offspring chromosomes l’i-l'& are respectively generated by means of single inversion in domains a-b, a-c. a-d, b-e, b-d and ¢-d. Similar to DDIGA, I'7 is generated by double inversions in domains a-b and c-d. Then, the parent chromosome and the seven oflspring chromosomes are classified according to their fitness. Only the most dominant chromosome 1° (I's in this example) is retained for the next gencration, while the other chromosomes are completely eliminated. The multi-domain inversion based genetic algorithm proposed in this embodiment can speed up the evolution to higher fitness. and improve the convergence accuracy and robustness. In this embodiment. the effectiveness of the above-mentioned CGA. DDIGA and MDIGA algorithms is verilied by means of a Monte Carlo simulation method in terms of the number of planning points. population size, calculation efficiency. cle. (1) Comparison results of different planning points. Five model examples from TSPLIB are used: burmal4. ulysses22, ¢il51. eil76 and rat99, Correspondingly. the five planning points (P) are respectively 14, 22, 51, 76 and 99. the maximum numbers of iterations (Nmax) are respectively set to 100, 200,
1600. 2000 and 2000. and the population size (S) is 100. The values of crossover
I probability (Pc) and mutation probability (Pm) are usually determined by practical LU102400 experience. According to the recommendations of M, Flhoseny et al. 81; the recommended value range of Pc is 0.7 to |. and any value less than this value will reduce the crossover operation and is not conducive to evolution. The recommended value range of Par is 0.001 to 0.05. and any value more than this value will increase the mutation operation and cause the algorithm to jump out of the optimal solution. Therefore. according to the recommendations in the existing literature and practical experience, the crossover probability (Pc) and the mutation probability (Pm) in this embodiment are respectively defined as 0.90 and 0.10. Monte Carlo simulation is performed on each TSP instance. and a data set of optimal path distances is obtained by using the four algorithms. The comparison results are shown in five box plots in Fig. 8(a) to Fig. 8(c). For a different algorithm in cach box plot. a range bar is drawn to represent an interquartile range (IQR) of a data set, which represents the degree of dispersion of the data set. The median and average values are represented by a red line and a plus sign in the bar chart. In addition, there are borders around the data bar. and the ends of the borders represent the minimum and maximum values. À standard deviation is calculated to show the distance between the data set and the average value thereof. which reflects the robustness of the algorithm. Under the same working conditions. if the standard deviation is smaller. the robustness of the algorithm is better. When there are 14 planning points, the solution provided by CGA has a larger average distance and higher data dispersion than the other solutions in the figure, as shown in Fig. 8(a). and the results of the average optimal path distances of the other three improved algorithms are similar. and are all 30.9 m. In addition. the median values of 75 CGA and DDIGA are less than their average values, which means that in a hundred repeated simulations, the two algorithms are more likely to produce larger data than other algorithms. With the increase of P in Fig. 8(b)-Fig. 8(c). CGA has the longest path distance, the worst robustness. and a more obvious gap. MDIGA has excellent performance in reducing path distance and improving robustness. In the case of P=99. the average distance and standard deviation of MDIGA are 1341.81 m and 31.41 m. which are
49.0% and 79.6% less than those of CGA. respectively. In addition, except for the case of P=22, SDIGA performs better than DDIGA in almost all cases, which means that in this experiment, not all improvements are effective for the algorithm. Since the number of oflspring of SDIGA and DDIGA is the same as the number of parents.
HU102400 there is no essential difference between a single-domain inversion with sufficient iterations for SDIGA and a double-domain inversion with sufficient iterations for DDIGA.
The results also show that the performance of the algorithm can be significantly optimized only by increasing the number of offspring. (2) Comparison results of different population sizes.
In this embodiment, cil51 with 51 planning points in TSPLIB is selected as the working condition, The five population sizes are respectively 20. 40, 60. 80 and 100. In addition, the maximum number of iterations (Nmax) of cach algorithm is set to 1600. The crossover probability (Pc) and the mutation probability (Pm) are respectively 0.90 and 0.10, and 100 Monte Carlo simulations are performed using four algorithms with different population sizes.
Fig. %a)-Fig. 9(e) are five box plots, showing the comparison results.
As shown in Fig. 9(a), the three improved algorithms. especially SDIGA and MDIGA, effeetively reduce the optimal path distance and improve the robusiness through comparison with CGA.
In addition, the median value is almost the same as the average value in cach bar, which means that all the algorithms can generate evenly distributed data under the working condition of eslS1. As shown in Fig. 9(b)-Fig. 9(¢). when S increases, a significant impact occurs. that 1s. the overall optimal distance of each algorithm is further reduced.
Although the robustness of each algorithm slightly changes with the increase in population, but no regular change trend is found.
In addition, the double-domain inversion algorithm is not as good as SDIGA in reducing the optimal path distance and improving the robustness of the algorithm, which is inconsistent with the assumptions mentioned above.
In contrast.
MDIGA is still the most advantageous algorithm for TSP.
When S=60. the average distance of MDIGA is 451.63 m, and the standard deviation is 7.72 m, which are 25.8% and 79.2% less than those of CGA, respectively. (3) Comparison results of calculation efficiency.
In this embodiment, the calculation efficiency is compared by means of the results of five TSPLIB instances with different planning points.
Two main criteria are selected to evaluate the calculation efliciency of each algorithm: calculation time and convergence speed.
The calculation time refers to a time cost required to complete a maximum number of iterations, and the convergence speed refers to a critical number of iterations (Neri) when a solution reaches a convergence level.
Generally. it can be observed that as the number of iterations increases. the path 0102600 distance of each algorithm gradually becomes short, then converges lo a stable level at the critical number (Nj). and finally reaches global optimum.
As the number of planning points increases. the critical points and time consumption of cach algorithm have arising trend.
By contrast, in the entire calculation process. the curve of MDIGA is lower than the curves of other algorithms, the convergence speed is taster, and the critical number is lower.
For example, when P=76, MDIGA converges to Nei=5B6. which is 63% faster than CGA, and MDIGA takes 46% more time to complete the same iterations.
It is worth noting that the improved algorithms, especially SDIGA and MDIGA. greatly reduce the calculation time cost, ensure the accuracy of solutions. and avoid falling into local optima.
In addition.
Fig. 10 shows five TSPLIB instances using MDIGA, where (a) is burmal4, (b) is ulysses22, (c) is eilS1, (d) is cil76, and (e) is the best track of rat99. The abscissa and ordinate respectively represent the latitude and longitude values of cach planning point.
The red numbers are a sequence of randomly generated points, the point enclosed by the red rectangle is a starting point. and the arrows indicate the direction of the planned path.
Specifically. the specific implementation process of performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle is as follows: (1) Path planning based on double-domain inversion based genetic algorithm.
Specifically, a specific method for path planning based on a double-domain inversion-based genetic algorithm is: Step one, initialization of parameters.
A population size. a maximum number of iterations. an initial crossover probability and an initial mutation probability are set.
Step two, initialization of population.
An initial population is randomly generated as the parent in the genetic process.
Step three. calculation of fitness value.
The fitness function is defined as 1/len, where len represents the relative path length of each chromosome.
The initial population is ranked according to the calculated fitness value.
Step four. selection, crossover and mutation operations on chromosomes.
The crossover probability and the mutation probability are respectively defined as 0.90 and 0.10. Meanwhile, the fitness value of a new population is calculated. and the initial population is re-ranked according to the value to obtain the new population as a primary offspring. LU102400 Step five. double-domain inversion operation. Four serial numbers are randomly selected as inversion points of chromosome coding, two domains are respectively generated between the first two points and the last two points, the fragments in the two domains are inverted at the same time to produce new offspring, the fitness values of the offspring chromosomes and the parent chromosome are compared, the chromosomes with larger fitness values are retained, and the population is updated. The double-domain inversion is shown in Fig. 11. where S represents a parent chromosome, and S” is an inverted offspring chromosome.
Step six. judgment on iteration termination condition. The iteration termination condition is set as the requirement of a certain working condition is met or the number of iterations reaches the maximum. If the termination condition is not satisfied. the number of iterations is increased by one, and step four is performed. If the termination condition is satisfied, step seven is performed.
Step seven, an optimal individual is selected from the retention results of iterations as the optimal solution of the double-domain inversion based genetic algorithm and output. and the whole algorithm ends.
(2) Path planning based on multi-domain inversion based genetic algorithm. Specifically. a specific method for path planning based on a multi-domain inversion based genetic algorithm is: Step one, initialization of parameters. A population size. a maximum number of iterations. an initial crossover probability and an initial mutation probability are set. Step two, initialization of population. An initial population is randomly generated as the parent in the genetic process.
Step three. calculation of fitness value. The fitness function is defined as 1/len, where len represents the relative path length of each chromosome. The initial population is ranked according to the calculated fitness value.
Step four, selection, crossover and mutation operations on chromosomes. The crossover probability and the mutation probability are respectively defined as 0.90 and 0.10. Meanwhile. the fitness value of a new population is calculated, and 1he initial population is re-ranked according to the value to obtain the new population as a primary offspring.
Step five, multi-domain inversion operation: four inversion points. namely a, b, ©, d. are randomly defined in a coded string. and six offspring chromosomes S1-S6 are respectively generated by means of single inversion in domains a-b. a-c, a-d. b-c, b-d 0102400 and c-d. Similar to the double-domain inversion, S7 is generated by double inversions in the domains a-b and c-d. as shown in Fig. 12. Then, the parent chromosome and the seven offspring chromosomes arc compared according to their fitness. and the optimal chromosome is retained and passed on to the next generation. while the other chromosomes are completely eliminated. Step six, judgment on iteration termination condition. The iteration termination condition is set as the requirement of a certain working condition is met or the number of iterations reaches the maximum. If the termination condition is not satisfied, the number of iterations is increased by one, and step four 1s performed. If the termination condition is satisfied, step seven is performed.
Step seven. an optimal individual is selected from the retention results of iterations as the optimal solution of the multi-domain inversion based genetic algorithm and output. and the whole algorithm ends.
S104, path numbers are integrated according to the optimal path ranking. the speed and steering of a steering gear of the unmanned surface vehicle are adjusted. and the track of the unmanned surface vehicle is corrected. to complete the path planning. Specifically. the specific implementation process of integrating the path numbers according to the optimal path ranking, adjusting the speed and steering of the steering gear of the unmanned surface vehicle, and correcting the track of the unmanned surface vehicle, to complete the path planning is: The optimal path ranking planned by the improved genetic algorithm is combined with the longitude and latitude coordinates cotlected by a GPS navigation module, a rectangular coordinate system palh map established in the actual marine environment is drawn. and distance and deflection angle information between the current location of the unmanned surface vehicle and a target point is obtained.
Data processing is performed on the distance and deflection angle information between the current location of the unmanned surface vehicle and the target point according 10 the constraint factors Lo obtain a real-time deflection angle and a relative distance between the current location of the unmanned surface vehicle and the target point. and the steering gear is controlled to start, accelerate, deflect, or decelerate according to the obtained distance and deflection angle information.
Experimental verification The path planning method for an unmanned surface vehicle based on an improved genetic algorithm proposed in this embodiment is experimentally verified, and the 7192400 specific implementation process 1s as follows: At present, unmanned surface vehicles (USVs) have been widely used in civilian and military fields due to their advantages of reducing the risk of casualties and improving mission efficiency. As one of the core technologies, path planning is of great significance to the realization of autonomous navigation and control of the unmanned surface vehicles. The method proposed in this embodiment is applied to self-developed USV path planning. As a preliminary study, this embodiment ignores factors such as wind, current, and waves.
The USV model adopted in this embodiment is a pentamaran with a length of 1.8 meters and a width of 0.9 meters. In addition, a 48V, 45A battery provides power for a motor that drives a propeller.
A navigation, guidance and control (NGC) system is placed inside the hull to ensure a dry working environment. The NGC system is composed of three module subsystems: a navigation dala processing subsystem, a path planning subsystem and an autopilot subsystem. In the navigation data processing subsystem, a plurality of sensors including an electronic compass and a GPS are used to obtain stem direction and USV location data. An ultrasonic weather sensor produced by Airmar®, Model: WeatherStation®PB200, is used to collect real-time and site-specific weather and location information. All voltage signals from the plurality of sensors are collected by a navigation data acquisition (DAQ) system, and navigation data is stored in real time together with ship log and status information.
Then, all the information is processed and transmitted to the path planning subsystem, where GA is applied to generate the best track. According to the planned route, an autopilot determines the course and speed of the USV by means of a closed-loop controller. In addition, a GUI program compiled based on a Spring MVC framework is used to process and record all data in a personal computer. A GPRS wireless network is used as a communication unit between the USV and the personal computer. with an effective distance of 5 kilometers and a transmission speed of 1-100 Mbps. It should be noted that there are still some challenges when the path planning method proposed in this embodiment is applied to the NGC system of the USV. Due to the influence of wind, wave and current, the unmanned surface vehicle has a tendency to deviate from the planned track, so the course is required to be corrected correspondingly. At the same time, the stability of USV data transmission needs to be strengthened, especially when remote offshore operations are required. In addition, LU102400 dynamic obstacle detection and obstacle avoidance functions need to be added to the path planning subsystem, especially under severe sea conditions, the requirements for the precision of sensors arc more stringent.
In the actual environment near the Qingdao Olympic Sailing Center in Fushan Bay, 4 different planning point schemes are randomly selected according to 4 working conditions: 15, 25, 35, and 45. Each condition has the same starting point (N 36°03'22.38", E 120°22'57.06") in latitude and longitude. The above four GAs are respective used in the USV model to verily the effectiveness of their path planning.
The population size (S) is set to 100. The maximum numbers of iterations (Nmax) are 150, 250, 350 and 450, corresponding to the four planning points respectively. In addition, the crossover probability (Pc) and the mutation probability (Pw) are still 0.90 and 0.10.
Under the four working conditions, the iterative convergence curve of each algorithm is shown in Fig. 13(a)-Fig. 13(d). MDIGA has more advantages among the four comparative algorithms. As P increases, MDIGA has more obvious advantages in accelerating convergence and optimizing path distance. For example, when P=45, the curve of MDIGA converges to Neri=186, and the best path distance of 77.1 m is obtained, which is 33.1% shorter than that of CGA. In addition, the performance of DDIGA is also not as good as that of SDIGA. In most cases, the track planned by DDIGA is slightly longer than that of SDIGA, as shown in Fig. 13(a)-Fig. 13(c). Under the same Nmax, the three improved algorithms require more calculation time than CGA. However, MDIGA is not the most time-consuming algorithm. MDIGA shows the ideal ability to balance path optimization and time consumption.
Figs. 14-17 show optimal track diagrams of each algorithm under various working conditions. When there are 15 planning points in the figure, as shown in Fig. 16, SDIGA and MDIGA are better than CGA around numbers 3, 12, and 15. In addition, as the value of P increases, the track becomes more complicated, and the differences in path shape and distance are more obvious. I'rom Figs. 15 (a), 16 (a), 16 (c) and 17 (a), it can be seen that the tracks generated by CGA and DDIGA have different degrees of path crossing phenomenon, which is why a longer route distance is generated under the same condition, as compared with other algorithms. But at the same time, MDIGA is more obvious in avoiding path intersection and simplifying path shape, especially when more planning points are considered. The main reason may be that the retention of a large number of offspring and the most suitable LU102400 individuals can help avoid local optima and converge to the optimal solutions.
Embodiment 2 A path planning system for an unmanned surface vchicle based on an improved genetic algorithm. including: a navigation data acquisition module. configured to acquire course data and location data of the unmanned surface vehicle, and preprocess same: constraint factor determination module, configured to collect weather and wave information of an environment where the unmanned surface vehicle is located, and convert same into constraint factors: an optimal path planning module, configured to perform path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle. to obtain an optimal path ranking; and a track correction module, configured to correct, based on the optimal path ranking, the course and speed of the unmanned surface vehicle according to the constraint factors, to complete the path planning.
Embodiment 3 A computer-readable storage medium, storing a computer program that. when executed by a processor. the following steps arc implemented: acquiring course data and location data of the unmanned surface vehicle, and preprocessing same: collecting weather and wave information of an environment where the unmanned surface vehicle is located, and converting same into constraint factors: performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle, to obtain an optimal path ranking; and correcting, based on the optimal path ranking, the course and speed of the unmanned surface vehicle according to the constraint factors, to complete the path planning.
Embodiment 4 A processing apparatus, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the following steps arc implemented: acquiring course data and location data of the unmanned surface vehicle, and preprocessing same;
collecting weather and wave information of an environment where the unmanned LU102400 surface vehicle is located, and converting same into constraint factors; performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle, to obtain an optimal path ranking: and correcting, based on the optimal path ranking, the course and speed of the unmanned surface vehicle according to the constraint factor. to complete the path planning.
Although the specific embodiments of the present disclosure are described above in combination with the accompanying drawing. the protection scope of the present disclosure is not limited thereto. It should be understood by those skilled in the art that various modifications or variations could be made by those skilled in the art based on the technical solution of the present disclosure without any creative effort, and these modifications or variations shall fall into the protection scope of the present disclosure.
Claims (10)
1. A path planning method for an unmanned surface vehicle based on an improved genetic algorithm. comprising the following steps: acquiring course data and location data of the unmanned surface vehicle. and preprocessing same: collecting wave information of an environment where the unmanned surface vehicle is located, and converting same into constraint factors: performing path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle, to obtain an optimal path ranking: and correcting, based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factors, to complete the path planning.
2. The path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to claim 1, wherein the course data and location data of the unmanned surface vchicle comprises longitude and latitude coordinate data of a navigation point required by the unmanned surface vehicle. and the latitude and longitude coordinates of the navigation point are converted into horizontal and vertical coordinate values in a rectangular coordinate system.
3. The path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to claim 1, wherein the constraint factors are: Fo =-0.0073V/}hV,? +0.0057V/ , My = NAY 5 where. h is the height of waves, Vi is the velocity of waves. and A is the wavelength of waves. 7102400
4. The path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to claim 1. wherein the improved genetic algorithm comprises a dual-domain inversion based genetic algorithm and a multi-domain inversion based genetic algorithm.
5. The path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to claim 4, wherein a specific method for path planning based on the dual-domain inversion based genetic algorithm is: (1) initializing parameters: setting a population size. a maximum number of iterations, an initial crossover probability and an initial mutation probability; (2) initializing a population: randomly generating an initial population as the parent in the genetic process: (3) calculating fitness values: calculating the fitness value of each chromosome. and ranking the initial population according to the calculated fitness values: (4) performing selection, crossover and mutation operations on chromosomes, calculating the fitness value of a new population. and re-ranking the initial population according to the value to obtain the new population as a primary offspring: (5) performing a double-domain inversion operation: randomly selecting four serial numbers as inversion points of chromosome coding, respectively gencrating two domains between the first two points and the last two points, inverting the fragments in the two domains at the same time to produce new offspring, comparing the fitness values of the offspring chromosomes and the parent chromosome, retaining the chromosomes with larger fitness values. and updating the population; (6) judging whether an iteration termination condition is satisfied: if the termination condition is not satisfied, adding one to the number of iterations. and performing step 0102400 (4): if the termination condition is satisfied, performing step (7): and (7) selecting an optimal individual from the retention results of iterations as the optimal solution of the double-domain inversion based genetic algorithm and outputting same.
6. The path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to claim 3, wherein a specific method for path planning based on the multi-domain inversion based genetic algorithm is: (1) initializing parameters: setting a population size. a maximum number of iterations.
an initial crossover probability and an initial mutation probability: (2) initializing a population: randomly generating an initial population as the parent in the genetic process: (3) calculating fitness values: calculating the fitness value of each chromosome, and ranking the initial population according to the calculated fitness values: (4) performing selection, crossover and mutation operations on chromosomes, calculating the fitness value of a new population, and re-ranking the initial population according to the value to obtain the new population as a primary offspring; (5) performing a multi-domain inversion operation: randomly defining four inversion points in a coded string, generating six domains between any two inversion points, inverting respective fragments in the six domains to produce six new offspring chromosomes, respectively generating two domains between the first two points and the last two points, and inverting the fragments in the two domains at the same time to produce a new offspring chromosome; comparing the fitness values of the offspring chromosomes and the parent chromosome, retaining the chromosomes with larger fitness values, and updating the population: "0108600 (6) judging whether an iteration termination condition is satisfied: if the termination condition is not satisfied. adding one to the number of iterations, and performing step (4); if the termination condition is satisfied, performing step (7); and (7) selecting an optimal individual from the retention results of iterations as the optimal solution of the multi-domain inversion based genetic algorithm and outputting same.
7. The path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to claim 1, wherein a specific method for correcting. based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factors is: combining the optimal path ranking planned by the improved genetic algorithm with the longitude and latitude coordinates collected by a GPS navigation module, drawing a rectangular coordinate system path map established in the actual marine environment. and obtaining distance and deflection angle information between the current location of the unmanned surface vehicle and a target point; and performing data processing on the distance and deflection angle information between the current location of the unmanned surface vehicle and the target point according to the constraint factors, to obtain a real-time deflection angle and a relative distance between the current location of the unmanned surface vehicle and the target point.
8. A path planning system for an unmanned surface vehicle based on an improved genetic algorithm, comprising: a navigation data acquisition module, configured to acquire course data and location data of the unmanned surface vehicle, and preprocess same:
constraint factor determination module, configured to collect wave information of an 0102400 environment where the unmanned surface vehicle is located. and convert same into constraint factors: an optimal path planning module. configured to perform path planning by using the improved genetic algorithm according to the course data and location data of the unmanned surface vehicle. to obtain an optimal path ranking: and a track correction module. configured to correct, based on the optimal path ranking. the course and speed of the unmanned surface vehicle according to the constraint factor. to complete the path planning.
9. A computer-readable storage medium. storing a computer program thereon. wherein when the program is executed by a processor. the steps in the path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to any onc of claims 1-7 are implemented.
10. À processing apparatus, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program. the steps in the path planning method for an unmanned surface vehicle based on an improved genetic algorithm according to any one of claims 1-7 are implemented.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910721372.6A CN110442135B (en) | 2019-08-06 | 2019-08-06 | Unmanned ship path planning method and system based on improved genetic algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
LU102400A1 LU102400A1 (en) | 2021-02-09 |
LU102400B1 true LU102400B1 (en) | 2021-04-27 |
Family
ID=68433459
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
LU102400A LU102400B1 (en) | 2019-08-06 | 2019-09-20 | Path planning method and system for unmanned surface vehicle based on improved genetic algorithm |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN110442135B (en) |
LU (1) | LU102400B1 (en) |
WO (1) | WO2021022637A1 (en) |
Families Citing this family (88)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111145068B (en) * | 2019-12-09 | 2023-09-26 | 南京航空航天大学 | Long-distance high-timeliness economical cruising vehicle speed planning method |
CN111026135B (en) * | 2020-01-18 | 2020-11-13 | 上海大学 | High-performance sailing feedforward control system and control method for unmanned ship |
CN111610788B (en) * | 2020-06-13 | 2023-04-18 | 大连海事大学 | Hierarchical fuzzy-artificial potential field path planning method |
CN111664852B (en) * | 2020-06-15 | 2023-07-04 | 天津大学 | Unmanned aerial vehicle path planning method and device |
CN112034865B (en) * | 2020-08-12 | 2021-10-08 | 浙江大学 | Full-drive underwater vehicle track tracking control method based on optimization algorithm |
CN112381284B (en) * | 2020-11-11 | 2023-12-01 | 成都信息工程大学 | Improved genetic algorithm for multi-station path optimization of unmanned connection vehicle |
CN112529278B (en) * | 2020-12-02 | 2021-08-31 | 中国人民解放军93209部队 | Method and device for planning navigation network based on connection matrix optimization |
CN112818472B (en) * | 2021-02-25 | 2023-04-18 | 西北工业大学 | Civil aircraft flight test subject arrangement and optimization method |
CN113570022B (en) * | 2021-03-05 | 2023-05-23 | 福建工程学院 | Chargeable mobile trolley path planning method and device in sensor network |
CN113029150B (en) * | 2021-03-10 | 2023-09-08 | 辽宁工程技术大学 | Intelligent aircraft track planning method under multiple constraint conditions |
CN113175930B (en) * | 2021-03-22 | 2024-04-30 | 西安理工大学 | MMEA-based multi-unmanned aerial vehicle collaborative track planning method |
CN112947480B (en) * | 2021-03-24 | 2022-11-15 | 武汉理工大学 | Mobile robot path planning method, storage medium and system |
CN113110424A (en) * | 2021-03-26 | 2021-07-13 | 大连海事大学 | Unmanned ship collision avoidance method based on chart information |
CN113124873B (en) * | 2021-04-09 | 2022-08-16 | 青岛哈船海智科技有限公司 | UUV multi-index constraint three-dimensional route planning method based on marine environment information |
CN113093765B (en) * | 2021-04-16 | 2022-04-08 | 西南交通大学 | Universal AGV cluster control system design framework and method |
CN113159687B (en) * | 2021-04-29 | 2023-08-29 | 长安大学 | Workshop AGV-UAV (automated guided vehicle-unmanned aerial vehicle) coordinated material distribution path planning method and system |
CN113188547A (en) * | 2021-05-06 | 2021-07-30 | 国能朔黄铁路发展有限责任公司 | Unmanned aerial vehicle path planning method and device, controller and storage medium |
CN113095733A (en) * | 2021-05-10 | 2021-07-09 | 北京化工大学 | Method for solving scheduling of single mechanical arm transfer job shop based on improved genetic algorithm |
CN113177762B (en) * | 2021-05-14 | 2024-01-05 | 大连海事大学 | Multi-center electric vehicle-unmanned aerial vehicle distribution path optimization method |
CN113128788B (en) * | 2021-05-17 | 2024-03-08 | 华北电力大学 | Power emergency material conveying path optimization method, device and storage medium |
CN113222272B (en) * | 2021-05-26 | 2022-09-20 | 合肥工业大学 | Emergency material transportation and loading cooperative optimization method based on double-layer genetic coding |
CN113361753B (en) * | 2021-05-26 | 2023-07-04 | 中国电子技术标准化研究院 | Method, system and medium for determining optimal path based on quantum genetic algorithm |
CN114399043B (en) * | 2021-06-16 | 2024-06-04 | 北京理工大学 | Vehicle position routing method based on non-dominant ordered particle swarm genetic algorithm |
CN113592148B (en) * | 2021-07-01 | 2024-03-15 | 合肥工业大学 | Optimization method and system for improving delivery route of vehicle and unmanned aerial vehicle |
CN113673922A (en) * | 2021-07-09 | 2021-11-19 | 合肥工业大学 | Fishbone type warehouse layout-based multi-vehicle picking path problem optimization method and system |
CN113595774B (en) * | 2021-07-19 | 2023-08-01 | 广西大学 | IAGA algorithm-based high-speed train networking topology optimization method |
CN113762594B (en) * | 2021-07-23 | 2023-07-07 | 合肥工业大学 | Route planning method and device for vehicle-machine collaborative distribution post-disaster rescue materials |
CN113624235A (en) * | 2021-07-31 | 2021-11-09 | 武夷科技信息(北京)有限公司 | Method for dynamically adjusting navigation path in real time by unmanned aerial vehicle |
CN113687657B (en) * | 2021-08-26 | 2023-07-14 | 鲁东大学 | Method and storage medium for multi-agent formation dynamic path planning |
CN113887782A (en) * | 2021-08-27 | 2022-01-04 | 西安电子科技大学 | Genetic-firework mixing method and system for maintenance resource distribution scheduling |
CN113821029B (en) * | 2021-08-31 | 2022-05-10 | 南京天溯自动化控制***有限公司 | Path planning method, device, equipment and storage medium |
CN113741477B (en) * | 2021-09-14 | 2022-10-21 | 中国船舶重工集团公司第七0七研究所九江分部 | Under-actuated ship berthing path planning method |
CN113885555A (en) * | 2021-09-14 | 2022-01-04 | 安徽送变电工程有限公司 | Multi-machine task allocation method and system for power transmission line dense channel routing inspection |
CN113780839B (en) * | 2021-09-15 | 2023-08-22 | 湖南视比特机器人有限公司 | Evolutionary sorting job scheduling method and system based on deep reinforcement learning |
CN113779885B (en) * | 2021-09-16 | 2024-04-30 | 南京航空航天大学 | Tolerance optimization method based on genetic algorithm |
CN113778128B (en) * | 2021-09-18 | 2023-03-31 | 中国人民解放军国防科技大学 | Unmanned aerial vehicle reconnaissance task planning method and device based on genetic algorithm |
CN113778127B (en) * | 2021-09-18 | 2023-02-28 | 中国人民解放军国防科技大学 | Unmanned aerial vehicle reconnaissance path planning method and device based on genetic algorithm |
CN113703463B (en) * | 2021-09-23 | 2023-12-05 | 大连海事大学 | Ship collision avoidance path planning method based on elite multi-population evolution algorithm |
CN113819919B (en) * | 2021-09-26 | 2022-07-01 | 陕西科技大学 | Robot optimal path planning method based on Boolean constraint |
CN113945796B (en) * | 2021-09-29 | 2023-10-24 | 集美大学 | Power distribution network fault positioning method, terminal equipment and storage medium |
CN113867358B (en) * | 2021-10-13 | 2023-08-29 | 中国人民解放军陆军工程大学 | Intelligent path planning method for multi-unmanned vehicle collaborative traversal task |
CN114049243B (en) * | 2021-10-26 | 2023-09-29 | 中山大学 | Enteromorpha accumulation amount estimation method, system and storage medium |
CN114020022B (en) * | 2021-11-04 | 2023-08-18 | 中国人民解放军陆军工程大学 | Heterogeneous unmanned aerial vehicle collaborative hit task planning method and device |
CN113885533B (en) * | 2021-11-12 | 2022-08-09 | 江苏海洋大学 | Unmanned driving method and system of unmanned boat |
CN114169828B (en) * | 2021-12-08 | 2024-06-04 | 南通大学 | Multi-machine type multi-material type unmanned aerial vehicle emergency distribution path planning method and planning system thereof |
CN114355895B (en) * | 2021-12-14 | 2023-11-07 | 南京航空航天大学 | Vehicle active collision avoidance path planning method |
CN114281104B (en) * | 2021-12-16 | 2024-02-20 | 成都戎星科技有限公司 | Multi-unmanned aerial vehicle cooperative regulation and control method based on improved ant colony algorithm |
CN114281108B (en) * | 2021-12-28 | 2024-05-03 | 北京神星科技有限公司 | Delivery task planning method based on hierarchical improvement genetic algorithm |
CN114265355B (en) * | 2021-12-30 | 2024-02-20 | 上海大学 | Remote controller with autonomous unmanned ship hovering control function and control method thereof |
CN114397896B (en) * | 2022-01-10 | 2023-07-14 | 贵州大学 | Dynamic path planning method for improving particle swarm optimization |
CN114610027A (en) * | 2022-03-07 | 2022-06-10 | 吉林大学 | Ship navigation path planning method |
CN114608596B (en) * | 2022-03-09 | 2024-04-12 | 大连大学 | Path planning method based on DSSA algorithm |
CN114647541B (en) * | 2022-03-18 | 2024-04-26 | 西安微电子技术研究所 | Circuit self-repairing method based on improved evolution hardware |
CN114781863B (en) * | 2022-04-21 | 2024-06-07 | 南京航空航天大学 | Multi-robot collaborative machining efficiency and quality double-objective optimization task planning method |
CN114897246A (en) * | 2022-05-16 | 2022-08-12 | 北京航空航天大学 | Airship multi-task point route planning method based on genetic algorithm and greedy strategy |
CN114924593B (en) * | 2022-06-16 | 2024-04-26 | 华中科技大学 | Quick planning method for vehicle and multi-unmanned aerial vehicle combined route |
CN115079995B (en) * | 2022-07-22 | 2022-11-11 | 中南大学 | Method and system for sequencing measured sample measuring points of neutron spectrometer |
CN115079704B (en) * | 2022-08-01 | 2024-02-27 | 中国电信股份有限公司 | Path planning method and device, storage medium and electronic equipment |
CN115421508B (en) * | 2022-08-01 | 2024-06-18 | 西安工业大学 | Particle swarm optimization method suitable for multi-unmanned aerial vehicle multi-target track planning |
CN115345549B (en) * | 2022-08-11 | 2024-01-23 | 上海维祥信息技术有限公司 | Vehicle path adjustment method and system combined with loading scheme |
CN115357021B (en) * | 2022-08-25 | 2023-10-10 | 齐齐哈尔大学 | Water surface unmanned ship path planning method based on improved Lingri search algorithm |
CN115438860A (en) * | 2022-09-06 | 2022-12-06 | 西安电子科技大学广州研究院 | Multi-agent path planning method based on evolutionary algorithm |
CN115495859B (en) * | 2022-09-19 | 2023-11-03 | 上海交通大学 | Warehouse net planning method based on genetic algorithm |
CN115469671B (en) * | 2022-10-21 | 2024-05-28 | 江苏海洋大学 | Robot path planning method based on aggregation-replacement strategy |
CN115562074A (en) * | 2022-10-29 | 2023-01-03 | 福州大学 | Simulation system and method for rapidly generating optimal planned path of unmanned aerial vehicle |
CN115451974B (en) * | 2022-11-09 | 2023-04-07 | 广东电网有限责任公司湛江供电局 | Power equipment routing inspection path planning method, system, equipment and medium |
CN115879603B (en) * | 2022-11-17 | 2024-05-14 | 武汉大学 | Multi-target-point-oriented multi-unmanned aerial vehicle collaborative data acquisition method and device |
CN115801665B (en) * | 2022-11-22 | 2024-05-24 | 山西大学 | Intra-domain route protection method based on forwarding graph |
CN116047909B (en) * | 2023-01-13 | 2023-09-05 | 大连海事大学 | Unmanned plane-ship cooperative robust self-adaptive control method for maritime parallel search |
CN115857516B (en) * | 2023-03-02 | 2023-07-14 | 之江实验室 | Full-coverage path planning method and device combining cow ploughing type movement and genetic algorithm |
CN115952942B (en) * | 2023-03-13 | 2023-06-02 | 季华实验室 | Multi-target inspection scheduling planning method and device, electronic equipment and storage medium |
CN116090357A (en) * | 2023-04-07 | 2023-05-09 | 中国人民解放军96901部队 | Multi-target hitting task feasibility rapid assessment method |
CN116088542B (en) * | 2023-04-12 | 2023-08-18 | 中国水产科学研究院南海水产研究所 | Fishing boat operation safety early warning method and system based on remote sensing technology |
CN116678409B (en) * | 2023-04-24 | 2024-03-12 | 大连海事大学 | North polar ice area ship path planning method based on improved wolf algorithm |
CN116542468B (en) * | 2023-05-06 | 2023-10-20 | 中国人民解放军32370部队 | Unmanned aerial vehicle cluster task planning method |
CN116954239B (en) * | 2023-06-12 | 2024-03-19 | 成都丰千达科技有限公司 | Unmanned aerial vehicle track planning method and system based on improved particle swarm optimization |
CN117213484B (en) * | 2023-06-21 | 2024-06-18 | 深圳大学 | Navigation point connection method and device based on genetic algorithm and intelligent terminal |
CN116519003B (en) * | 2023-06-28 | 2023-09-01 | 广汽埃安新能源汽车股份有限公司 | Path planning method and device, electronic equipment and storage medium |
CN116989797B (en) * | 2023-09-26 | 2023-12-15 | 北京理工大学 | Unmanned aerial vehicle track optimization method and device, electronic equipment and storage medium |
CN117196271B (en) * | 2023-11-08 | 2024-02-13 | 北京理工大学 | Double-effect scheduling method for heterogeneous robots in flexible workshops |
CN117494919B (en) * | 2023-11-13 | 2024-04-19 | 广州力生机器人技术有限公司 | Path planning method and device based on multi-robot collaborative stacking operation |
CN117742373A (en) * | 2023-11-13 | 2024-03-22 | 广东工业大学 | Unmanned aerial vehicle path planning method based on improved genetic algorithm |
CN117356546B (en) * | 2023-12-01 | 2024-02-13 | 南京禄口国际机场空港科技有限公司 | Control method, system and storage medium of spraying vehicle for airport lawn |
CN117419731B (en) * | 2023-12-19 | 2024-03-12 | 青岛哈尔滨工程大学创新发展中心 | Path planning method based on multi-strategy artificial bee colony algorithm in marine environment |
CN117608200B (en) * | 2024-01-23 | 2024-05-10 | 山东科技大学 | Ocean aircraft path planning method |
CN117601136B (en) * | 2024-01-24 | 2024-04-02 | 济宁鲁鑫工程机械有限公司 | Automatic welding robot path planning method and system |
CN117689677B (en) * | 2024-02-01 | 2024-04-16 | 山东大学日照研究院 | Grinding wheel abrasion state identification method, system, equipment and medium |
CN117951851B (en) * | 2024-02-19 | 2024-06-25 | 重庆大学 | Road large drainage path optimization design method for coping with extreme rainfall |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6629037B1 (en) * | 2000-06-26 | 2003-09-30 | Westerngeco, L.L.C. | Optimal paths for marine data collection |
CN102323820A (en) * | 2011-04-19 | 2012-01-18 | 江苏科技大学 | Intelligent rudder system and control method thereof |
CN102830707A (en) * | 2012-09-21 | 2012-12-19 | 中国航空无线电电子研究所 | Aircraft take-off, landing and pitch attitude control system |
JP2014193706A (en) * | 2013-02-27 | 2014-10-09 | Furuno Electric Co Ltd | Automatic steering device, automatic steering method and automatic steering program |
CN105867383A (en) * | 2016-05-16 | 2016-08-17 | 哈尔滨工程大学 | Automatic collision preventing control method of USV |
EP3287862A1 (en) * | 2016-08-25 | 2018-02-28 | Imotec Holding B.V. | Method for steering an underactuated ship |
CN107741744B (en) * | 2017-11-15 | 2020-09-25 | 哈尔滨工程大学 | Optimal heading control method for ocean observation platform |
CN108983774B (en) * | 2018-07-16 | 2021-06-01 | 哈尔滨工程大学 | Single-jet-pump-propelled unmanned surface vehicle self-adaptive course control method based on fuzzy state observer |
CN108919675B (en) * | 2018-09-10 | 2021-03-19 | 上海阜有海洋科技有限公司 | Multi-point mooring system control model optimization algorithm |
CN109062058B (en) * | 2018-09-26 | 2021-03-19 | 大连海事大学 | Ship course track tracking design method based on self-adaptive fuzzy optimal control |
CN109828570B (en) * | 2019-02-18 | 2021-10-01 | 哈尔滨工程大学 | Control and guide method for self-adaptive boundary layer water surface unmanned ship |
-
2019
- 2019-08-06 CN CN201910721372.6A patent/CN110442135B/en active Active
- 2019-09-20 LU LU102400A patent/LU102400B1/en active IP Right Grant
- 2019-09-20 WO PCT/CN2019/107109 patent/WO2021022637A1/en active Application Filing
Also Published As
Publication number | Publication date |
---|---|
WO2021022637A1 (en) | 2021-02-11 |
LU102400A1 (en) | 2021-02-09 |
CN110442135B (en) | 2020-08-21 |
CN110442135A (en) | 2019-11-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
LU102400B1 (en) | Path planning method and system for unmanned surface vehicle based on improved genetic algorithm | |
CN111780777B (en) | Unmanned vehicle route planning method based on improved A-star algorithm and deep reinforcement learning | |
Ai et al. | Coverage path planning for maritime search and rescue using reinforcement learning | |
CN108459503B (en) | Unmanned surface vehicle track planning method based on quantum ant colony algorithm | |
Roberge et al. | Comparison of parallel genetic algorithm and particle swarm optimization for real-time UAV path planning | |
Tsou et al. | An Ant Colony Algorithm for efficient ship routing | |
CN113110520A (en) | Robot path planning method based on multiple intelligent optimization parallel algorithms | |
Lan et al. | Path planning for underwater gliders in time-varying ocean current using deep reinforcement learning | |
Du et al. | An optimized path planning method for coastal ships based on improved DDPG and DP | |
Zeng et al. | Robotic global path-planning based modified genetic algorithm and A* algorithm | |
Bai et al. | USV path planning algorithm based on plant growth | |
Dong et al. | Double ant colony algorithm based on dynamic feedback for energy-saving route planning for ships | |
Tang | Coverage path planning of unmanned surface vehicle based on improved biological inspired neural network | |
Gao et al. | An optimized path planning method for container ships in Bohai bay based on improved deep Q-learning | |
Tao et al. | A study on path planning of unmanned aerial vehicle based on improved genetic algorithm | |
Ma et al. | Solving heterogeneous USV scheduling problems by problem-specific knowledge based meta-heuristics with Q-learning | |
Tan et al. | Adaptive adjustable fast marching square method based path planning for the swarm of heterogeneous unmanned surface vehicles (USVs) | |
Zhao et al. | Global path planning of unmanned vehicle based on fusion of A∗ algorithm and Voronoi field | |
Li et al. | Dynamic route planning for a USV-UAV multi-robot system in the rendezvous task with obstacles | |
Cui et al. | Multi-strategy adaptable ant colony optimization algorithm and its application in robot path planning | |
Liu et al. | The path planning study of autonomous patrol robot based on modified Astar algorithm and genetic algorithm | |
Xi et al. | Power-aware path planning for vehicle-assisted multi-UAVs in mobile crowd sensing | |
Song et al. | A novel path planning algorithm for ships in dynamic current environments | |
CN114943168B (en) | Method and system for combining floating bridges on water | |
Shih et al. | A parallel genetic approach to path-planning with upstream-current avoidance for multi-AUG deployment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FG | Patent granted |
Effective date: 20210427 |