CN102929286B - Rapid planning method for surface global path of planet - Google Patents
Rapid planning method for surface global path of planet Download PDFInfo
- Publication number
- CN102929286B CN102929286B CN201210487551.6A CN201210487551A CN102929286B CN 102929286 B CN102929286 B CN 102929286B CN 201210487551 A CN201210487551 A CN 201210487551A CN 102929286 B CN102929286 B CN 102929286B
- Authority
- CN
- China
- Prior art keywords
- path
- node
- communicated
- algorithm
- line segment
- 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.)
- Expired - Fee Related
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a rapid planning method for a surface global path of a planet, and belongs to the technical field of deep space exploration. The method comprises the following steps of: firstly obtaining an obstacle distribution information graph in a region to be subjected to path planning, carrying out analysis process on the graph, selecting a plurality of feasible nodes in the feasible region capable of avoiding obstacles and carrying out path connection in the feasible region according to the selected nodes; obtaining the coordinate information of the nodes, creating a network topology which the path planning needs by adopting a Dijkstra algorithm and planning an initial optimal path through taking the length of the path as the constraint condition; and taking the path length function as a fitness function, taking the mathematical function relationship followed in the process of selecting the nodes and the constraint range of the coordinates as a to-be-optimized object and the constraint condition of a genetic algorithm, optimizing the initial optimal path by adopting the genetic algorithm and outputting the optimization result used as the final planning path. The method has the advantages of simplicity in algorithm, high efficiency, good generality and strong expandability.
Description
Technical field
The present invention relates to the quick planing method of a kind of surface global path of planet, belong to field of deep space exploration.
Background technology
Along with progress and the development of science and technology, planetary detection has become one of focus of 21st century solar-system operation, such as, but planetary surface has extremely rugged environment condition, high vacuum, microgravity, high temperature difference, strong radiation environment etc., the mankind are difficult to existence in such a situa-tion.Therefore, planetary detecting robot must be adopted in this case to replace cosmonaut, land in planetary surface, do actual exploration and tasks of science exploration at planetary surface.
Planetary detecting robot is when planetary surface work, except being undertaken except remote control by ground, sniffing robot is also needed to have the ability of certain autonomous searching target, this just needs sniffing robot can by self-contained computer system, complete autonomous path planning to a certain degree according to the terrain information gathered, and the core of path planning plans the algorithm applied.
Apply in algorithm field mobile robot, lot of domestic and foreign scholar conduct in-depth research, also emerge many theoretical and outstanding algorithms be applied of having begun to take shape, as Artificial Potential Field Method, dijkstra's algorithm, ant group algorithm, neural network algorithm, genetic algorithm etc. simultaneously.Although often kind of algorithm has respective advantage, be also difficult to overcome the defect self existed.
Such as Dijkstra is a kind of typical shortest path first, and for calculating the shortest path of a node to other all nodes, advantage is that algorithm is simple, efficient, universal and gets well; Its shortcoming is then that planning field is only applicable to network topology structure, obtains the optimal path in existing topological structure.Genetic algorithm is based on the biological evolution theory such as natural genetic mechanism and natural selection, the random search algorithm of structure, and advantage is to have extensibility, is easy to used in combination etc. with other technology; Its shortcoming be then calculated amount large, encode lack of standardization and coding exists the inaccuracy represented, single use genetic algorithm is difficult to accurately design fitness function, what cause problem constraint expression is inaccurate, realizes complicated etc.
Summary of the invention
The object of the invention is, for solving the problem that existing planetary detecting robot global path planning method calculated amount is large, realization is complicated, range of application is limited, to propose a kind of quick planing method of surface global path of planet combined with genetic algorithm by dijkstra's algorithm.
The quick planing method of a kind of surface global path of planet, its specific implementation step is as follows:
Step 1, obtains the obstacle distributed intelligence figure in pending path planning region, and carries out analyzing and processing to it, in the feasible region of energy avoiding obstacles, choose multiple feasible node, and is communicated with according to the path that the node chosen carries out in feasible region.
The selection principle of node is: each node transverse and longitudinal coordinate meets same mathematical function relationship.
The mathematical function relationship that described node coordinate meets includes but not limited to linearly, quadratic function.
Path in feasible region is communicated with the principle followed adjacent node and be communicated with, and the path be communicated with can not be passed barrier, can not cross over multiple feasible node.
Step 2, to the node serial number be communicated with, obtains node coordinate information, and is stored in data information matrix U
1and U
2in, wherein U
1storage node coordinate information, U
2store the number information being communicated with line segment two end node.
In formula, n is the numbering sum of node, (x
k, y
k) for kth (k=1,2 ..., the n) coordinate information of individual node; M is the connection line segment sum formed after all feasible nodes are communicated with, (i
l, j
l) be l (l=1,2 ..., m) bar is communicated with node serial number corresponding to line segment two end points, and (i
l, j
l) ∈ { (i
l, j
l) | i
l, j
l=1,2 ..., n, i
l≠ j
l.
Step 3, the data matrix U that step 2 is obtained
1and U
2as starting condition, create and adopt dijkstra's algorithm to carry out network topology structure needed for path planning, using the length in path as constraint condition, cook up an initial optimal path, that is:
f(s)=min(length(s
start→s
goal))
Wherein: s
startfor starting point, s
goalfor impact point, f (s) is the shortest paths of the network topology structure middle distance created.
After obtaining initial optimal path, set up matrix U 3 and arrive from starting point the connection line segment sum experienced impact point process in order to storing initial optimal path, and be communicated with node serial number corresponding to line segment two ends.
Wherein: v represents the connection line segment sum that initial optimal path experiences from starting point arrival impact point process, (e
q, f
q) be the numbering that q article of connection line segment two end node is corresponding, (e
q, f
q) meet following relation:
(e
q,f
q)∈{(e
q,f
q)|q=1,2,…v，e
q≠f
q}。
Step 4, according to the matrix U that step 3 obtains
3in node serial number, and matrix U
1the world coordinates information of middle corresponding node, sets up fitness function, the present invention using path function f (S) as fitness function:
f(S)=min(length(e
1→f
1)+length(e
2→f
2)+…+length(e
q→f
q))
Wherein: q=1,2 ... v.
According to the mathematical function relationship followed when choosing node, to determine from starting point to impact point experience the horizontal ordinate of node and relation that ordinate meets, and the restriction range of corresponding coordinate:
f(x
k,y
k)=0,(a
k<x
k<b
k)
Wherein: y
k=f (x
k) be coordinate x in step 1
kwith y
kthe funtcional relationship that meets, (a
k, b
k) be horizontal ordinate x
kthe bound of restriction range, two the nearest barrier summit horizontal ordinates corresponding to respective nodes position are determined.
Step 5, the fitness function that step 4 is determined and the restriction range of corresponding coordinate, respectively as target to be optimized and the constraint condition of genetic algorithm, adopt genetic algorithm to be optimized process to the initial optimal path that step 3 obtains, export optimum results as final path planning.
Beneficial effect
Dijkstra's algorithm is effectively combined with genetic algorithm and forms a kind of hybrid algorithm by the inventive method, overcome single use dijkstra's algorithm and plan that the shortest path obtained is only applicable to network topology structure, obtain the inherent shortcoming of the optimal path in topological structure, and single use genetic algorithm calculated amount is large, the lack of standardization and inaccuracy of encoding, be difficult to accurately design fitness function, cause problem constraint expression inaccurate, realize the inherent shortcomings such as complicated.Also inherit dijkstra's algorithm simultaneously and there is the advantage that algorithm is simple, efficient, versatility is good, and the extensibility of genetic algorithm, be easy to the feature with other technology grade used in combination.
Accompanying drawing explanation
Fig. 1 is the quick planing method process flow diagram of surface global path of planet of the present invention;
Fig. 2 is the schematic diagram chosen, be communicated with and number of barrier feasible region interior nodes in embodiment;
Fig. 3 is the analog simulation contrast schematic diagram that in embodiment, path planning obtains optimal path.
Embodiment
In order to better objects and advantages of the present invention are described, below in conjunction with embodiment and accompanying drawing, technology contents is further described.
In present embodiment, the mixed path planing method specific implementation step that dijkstra's algorithm combines with genetic algorithm is as follows:
(1) obstacle distributed intelligence figure in pending path planning region is obtained, analyzing and processing is carried out to acquired obstacle distributed intelligence figure, some feasible nodes can be chosen in the feasible region of avoiding obstacles, and the connection in feasible region is carried out to these nodes, the present embodiment interior joint choose and connection in feasible region meets the following conditions:
A. choosing of node meets linear math constraint condition y
k=cx
k+ d (a
k<x
k<b
k), (x
k, y
k) be the world coordinates of a kth node; I.e. barrier scope interior zone, node is chosen on the line of barrier summit, barrier scope perimeter, and the border line according to horizontal ordinate direction corresponding thereto, barrier summit (or ordinate direction) obstacle information distribution plan is chosen.
When b. the connection in feasible region being carried out to node, can not barrier be passed, and multiple node can not be crossed over be communicated with.
(2) node be communicated with is numbered, obtains node world coordinates information, and create the matrix U of store digital information
1and U
2, wherein U
1in order to storage node coordinate information, U
2in order to save as the number information being communicated with line segment two end node, carrying out initial path planning for adopting dijkstra's algorithm provides feasible starting condition.
Wherein: n is the sum after nodes encoding, n=28 in the present embodiment, (x
k, y
k) for kth (k=1,2 ..., n) the world coordinates information of individual node; M is the sum being communicated with line segment after node is communicated with, m=47 in the present embodiment, (i
l, j
l) be l (l=1,2 ..., m) bar is communicated with the numbering of line segment two end points, and (i
l, j
l) ∈ { (i
l, j
l) | i
l, j
l=1,2 ..., n, i
l≠ j
l.
(3) by data matrix U
1and U
2carry out the network topology structure required for path planning as employing dijkstra's algorithm, using the length in path as constraint condition, cook up an initial optimal path:
f(s)=min(length(s
start→s
goal))
As shown in Figure 2, starting point s in the present embodiment
startbe numbered 1, impact point s
goalbe numbered 28.F (s) is at the shortest paths of the network topology structure middle distance created.
After obtaining initial optimal path, set up matrix U
3arrive from starting point the connection line segment sum lived through impact point process in order to storing initial optimal path, and be communicated with node serial number corresponding to line segment two ends.U in the present embodiment
3the data stored are:
Matrix U
3first row represent from starting point and arrive the numbering that impact point optimal path institute lives through connection line segment, matrix U
3second and third list show the numbering of the connection line segment two ends corresponding node lived through.
(4) according to matrix U
3middle the stored node serial number experienced from starting point arrival impact point process, and matrix U
1the world coordinates information of the corresponding node of middle storage, for genetic algorithm sets up fitness function, path selection length function as fitness function is:
According to when choosing node the linear math constraint condition that meets, determine from starting point arrive impact point live through node horizontal ordinate and ordinate the relational expression that meets, and the restriction range of corresponding coordinate:
y
k=cx
k+d(a
k<x
k<b
k)
Wherein c, d are the coefficient meeting linear relationship, a
k, b
kfor the bound of corresponding horizontal ordinate restriction range, k is through the numbering through node.
(5) to the above-mentioned fitness function determined and corresponding restriction range, genetic algorithm is adopted to be optimized process to the initial optimal path obtained with dijkstra's algorithm, in the present embodiment, use genetic algorithm time, the size of initial population is 20, retain in genetic iteration process per generation in 2 optimum individuals be genetic to the next generation, crossover probability is set to 0.8, mutation probability is set to 0.2, and evolutionary generation was set to for 50 generations, exports the final optimal path after optimizing through process.
Fig. 3 is the linear math constraint condition meeting node selection, and after being communicated with in node feasible region and requiring, the choosing of node, is communicated with and the method schematic diagram of node serial number; Fig. 3 is the analog simulation schematic diagram that path planning obtains optimal path in 100 × 100 (unit) scope, and wherein dotted line represents that the single dijkstra's algorithm of employing plans the initial optimal path obtained; Heavy line represents the path optimizing obtained after dijkstra's algorithm and genetic algorithm combine method planning.
Known through calculating, single use dijkstra's algorithm carries out the length d of the initial path of path planning acquisition in the present embodiment
1=129.4076, and be optimized through genetic algorithm the optimal path length d=118.8003 obtained after process, can see that outbound path obtains significant optimization, illustrate that the optimal path obtained is further close to theoretic optimal path, this also overcomes single use dijkstra's algorithm and plans that the shortest path obtained is only applicable to network topology structure, obtains the inherent shortcoming of the optimal path in topological structure; In above process, the initial path obtained is planned according to employing dijkstra's algorithm, carry out searching process for adopting genetic algorithm and establish complete fitness function, and the restriction range in fitness function corresponding to variable, which overcome the single use genetic algorithm of single employing genetic algorithm to be difficult to accurately design fitness function, what cause problem constraint expression is inaccurate, realizes complicated etc. inherent shortcoming.The mixed path planing method that dijkstra's algorithm combines with genetic algorithm can complete the task of path planning well.
Claims (3)
1. the quick planing method of surface global path of planet, is characterized in that: comprise the following steps:
Step 1, obtains the obstacle distributed intelligence figure in pending path planning region; After analyzing and processing, in the feasible region of energy avoiding obstacles, choose multiple feasible node, and be communicated with according to the path that the node chosen carries out in feasible region;
The selection principle of node is: each node transverse and longitudinal coordinate meets same mathematical function relationship;
Path in feasible region is communicated with the principle followed adjacent node and be communicated with, and the path be communicated with can not be passed barrier, can not cross over multiple feasible node;
Step 2, to the node serial number be communicated with, obtains node coordinate information, and is stored in data information matrix U
1and U
2in, wherein U
1storage node coordinate information, U
2store the number information being communicated with line segment two end node;
In formula, n is the numbering sum of node, (x
k, y
k) be the coordinate information of a kth node; M is the connection line segment sum formed after all feasible nodes are communicated with, (i
l, j
l) be the node serial number that l article of connection line segment two end points are corresponding, and (i
l, j
l) ∈ { (i
l, j
l) | i
l, j
l=1,2 ..., n, i
l≠ j
l; K=1,2 ..., n, l=1,2 ..., m;
Step 3, the data matrix U that step 2 is obtained
1and U
2as starting condition, create and adopt dijkstra's algorithm to carry out network topology structure needed for path planning, using the length in path as constraint condition, cook up an initial optimal path:
f(s)＝min(length(s
start→s
goal))
Wherein: s
startfor starting point, s
goalfor impact point, f (s) is the shortest paths of the network topology structure middle distance created;
Set up matrix U
3arrive from starting point the connection line segment sum experienced impact point process in order to storing initial optimal path, and be communicated with node serial number corresponding to line segment two ends;
Wherein: v represents the connection line segment sum that initial optimal path experiences from starting point arrival impact point process, (e
q, f
q) be the numbering that q article of connection line segment two end node is corresponding, (e
q, f
q) meet following relation:
(e
q,f
q)∈{(e
q,f
q)|q＝1,2,…v,e
q≠f
q}；
Step 4, according to the matrix U that step 3 obtains
3in node serial number, and matrix U
1the world coordinates information of middle corresponding node, using path function f (S) as fitness function:
f(S)＝min(length(e
1→f
1)+length(e
2→f
2)+…+length(e
q→f
q))
Determine coordinates restriction scope:
f(x
k,y
k)＝0,(a
k＜x
k＜b
k)
Wherein: y
k=f (x
k) be coordinate x
kwith y
kthe funtcional relationship that meets, (a
k, b
k) be horizontal ordinate x
kthe bound of restriction range;
Step 5, the fitness function that step 4 is determined and the restriction range of corresponding coordinate, respectively as target to be optimized and the constraint condition of genetic algorithm, adopt genetic algorithm to be optimized process to the initial optimal path that step 3 obtains, export optimum results as final path planning.
2. the quick planing method of a kind of surface global path of planet according to claim 1, is characterized in that: horizontal ordinate x
knearest two the barrier summit horizontal ordinates of bound corresponding to respective nodes position of restriction range are determined.
3. the quick planing method of a kind of surface global path of planet according to claim 1, is characterized in that: the mathematical function relationship that node coordinate described in step 1 meets comprises linearly, quadratic function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210487551.6A CN102929286B (en) | 2012-11-26 | 2012-11-26 | Rapid planning method for surface global path of planet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210487551.6A CN102929286B (en) | 2012-11-26 | 2012-11-26 | Rapid planning method for surface global path of planet |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102929286A CN102929286A (en) | 2013-02-13 |
| CN102929286B true CN102929286B (en) | 2015-05-27 |
Family
ID=47644114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210487551.6A Expired - Fee Related CN102929286B (en) | 2012-11-26 | 2012-11-26 | Rapid planning method for surface global path of planet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102929286B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103528586B (en) * | 2013-10-31 | 2016-06-01 | 中国航天时代电子公司 | Path Planning based on fault grid designs |
| CN105354393A (en) * | 2015-12-02 | 2016-02-24 | 上海核工程研究设计院 | System and method for realizing nuclear plant three-dimensional digital map |
| CN105607646B (en) * | 2016-02-05 | 2018-06-26 | 哈尔滨工程大学 | There are the UUV Route planners of necessary point under a kind of obstacle environment |
| CN105841704B (en) * | 2016-04-13 | 2019-01-18 | 京信通信系统（中国）有限公司 | A kind of determination method and device of movement routine |
| CN105867383A (en) * | 2016-05-16 | 2016-08-17 | 哈尔滨工程大学 | Automatic collision preventing control method of USV |
| CN106295164B (en) * | 2016-08-05 | 2018-12-04 | 中国兵器科学研究院 | A kind of paths planning method and electronic equipment |
| CN109357685B (en) * | 2018-11-05 | 2020-10-20 | 飞牛智能科技（南京）有限公司 | Method and device for generating navigation network and storage medium |
| CN109784526B (en) | 2018-12-05 | 2023-02-28 | 阿波罗智能技术(北京)有限公司 | Method, device and equipment for planning traffic path and readable storage medium |
| CN109459052B (en) * | 2018-12-28 | 2022-10-11 | 上海大学 | Full-coverage path planning method for sweeper |
| CN109799820B (en) * | 2019-01-22 | 2020-12-22 | 智慧航海（青岛）科技有限公司 | Unmanned ship local path planning method based on comparative random road map method |
| CN109931943B (en) * | 2019-03-25 | 2020-09-01 | 智慧航海（青岛）科技有限公司 | Unmanned ship global path planning method and electronic equipment |
| CN111060103B (en) * | 2019-12-11 | 2021-12-10 | 安徽工程大学 | Path planning method for local dynamic optimization and obstacle avoidance |
| CN112214031B (en) * | 2020-09-25 | 2021-08-20 | 北京理工大学 | Multi-node collaborative landing position planning method based on genetic particle swarm optimization |
| CN113282089B (en) * | 2021-05-26 | 2022-07-05 | 浙江大学 | Global path planning method for mobile robot in high-temperature scene |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1851598A (en) * | 2006-06-02 | 2006-10-25 | 哈尔滨工业大学 | Deep space detector big-angle flexible path autonomic generating method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005316614A (en) * | 2004-04-27 | 2005-11-10 | Univ Nihon | Optimization method and optimization program |
-
2012
- 2012-11-26 CN CN201210487551.6A patent/CN102929286B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1851598A (en) * | 2006-06-02 | 2006-10-25 | 哈尔滨工业大学 | Deep space detector big-angle flexible path autonomic generating method |
Non-Patent Citations (3)
| Title |
|---|
| 徐瑞等.深空探测器自主任务规划系统设计.《中国宇航学会深空探测技术专业委员会第一届学术会议》.2005,第277-283页. * |
| 程小军等.具有非凸约束的航天器姿态机动预测控制.《宇航学报》.2011,第32卷(第5期),第1070-1076页. * |
| 行为控制月球车路径规划技术;居鹤华等;《自动化学报》;20040731;第30卷(第4期);第572-576页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102929286A (en) | 2013-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102929286B (en) | Rapid planning method for surface global path of planet | |
| CN110502006B (en) | Full-coverage path planning method for mobile robot in abandoned land of mining area | |
| CN109839110B (en) | Multi-target point path planning method based on rapid random search tree | |
| Barca et al. | Swarm robotics reviewed | |
| Moorehead et al. | Autonomous exploration using multiple sources of information | |
| Panda et al. | An effective path planning of mobile robot using genetic algorithm | |
| Wang et al. | Rank-driven salp swarm algorithm with orthogonal opposition-based learning for global optimization | |
| Dogru et al. | Energy efficient coverage path planning for autonomous mobile robots on 3D terrain | |
| Yan et al. | A two-stage optimization method for unmanned aerial vehicle inspection of an oil and gas pipeline network | |
| Larki et al. | Solving the multiple traveling salesman problem by a novel meta-heuristic algorithm | |
| CN107422734B (en) | Robot path planning method based on chaotic reverse pollination algorithm | |
| Shen et al. | Simulation study on coverage path planning of autonomous tasks in hilly farmland based on energy consumption model | |
| Pathak et al. | Traveling salesman problem using bee colony with SPV | |
| Chen et al. | 2D multi-area coverage path planning using L-SHADE in simulated ocean survey | |
| Cheng et al. | Learning an optimal sampling distribution for efficient motion planning | |
| Parque et al. | Optimization of route bundling via differential evolution with a convex representation | |
| CN102324059A (en) | Target assignment method based on evolution | |
| Rivadeneyra et al. | Probabilistic estimation of multi-level terrain maps | |
| Yi et al. | Informative path planning with a human path constraint | |
| Nair | Efficient Path Planning Algorithm for Mobile Robots Performing Floor Cleaning Like Operations | |
| Xu et al. | Trajectory planning of Unmanned Aerial Vehicle based on A* algorithm | |
| Paz et al. | Data Association in O (n) for Divide and Conquer SLAM. | |
| Wang et al. | Welding robot path optimization based on hybrid discrete PSO | |
| Gasser et al. | Voxplan: A 3d global planner using signed distance function submaps | |
| Hao et al. | Path planning for aircraft based on MAKLINK graph theory and multi colony ant algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150527 Termination date: 20171126 |