CN114034301A - Real-time route planning method based on decision tree - Google Patents

Real-time route planning method based on decision tree Download PDF

Info

Publication number
CN114034301A
CN114034301A CN202111227109.5A CN202111227109A CN114034301A CN 114034301 A CN114034301 A CN 114034301A CN 202111227109 A CN202111227109 A CN 202111227109A CN 114034301 A CN114034301 A CN 114034301A
Authority
CN
China
Prior art keywords
planning
fly
flight
path
area
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.)
Pending
Application number
CN202111227109.5A
Other languages
Chinese (zh)
Inventor
罗喜伶
白天玥
王宇鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Hangzhou Innovation Research Institute of Beihang University
Original Assignee
Beihang University
Hangzhou Innovation Research Institute of Beihang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University, Hangzhou Innovation Research Institute of Beihang University filed Critical Beihang University
Priority to CN202111227109.5A priority Critical patent/CN114034301A/en
Publication of CN114034301A publication Critical patent/CN114034301A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20—Instruments for performing navigational calculations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Abstract

The invention discloses a real-time route planning method based on a decision tree. Firstly, establishing a low-altitude spatial domain probability model, representing a three-dimensional continuous space by using a discrete two-dimensional space, namely representing a planning space as a discrete grid diagram with a certain warp difference and weft difference, dividing a dangerous area into a no-fly area and a high-risk area, and then calculating a cost value of each node in the model; then, initial route planning is carried out, two paths of direct flight and fly-around are planned for a high-risk area, only the fly-around path is planned for a no-fly area, a route decision tree is generated, and an initial route set is determined; and selecting the planned route set according to the actual situation in the flight, and updating the planning result in real time. The invention comprehensively considers the threatening factors such as terrain barriers, severe weather and the like, and is safer; the uncertainty of the prior information of the dangerous area is considered, and a route decision tree comprising a plurality of routes is planned, so that the method is more practical; according to the actual situation in the flight, the planning result is updated in real time, and the method is more flexible.

Description

Real-time route planning method based on decision tree
Technical Field
The invention belongs to the technical field of route planning, and relates to a real-time route planning method based on a decision tree.
Background
The flight altitude of a general aviation aircraft is usually below 3000 m, wherein the activity of a low-altitude airspace below 1000 m in relative altitude is the most dominant, and the airspace is more complex compared with the flight environment of a civil aircraft. On one hand, the low-altitude atmosphere has the characteristics of strong convection, relatively severe weather change, obvious local weather and microclimate and the like, and the navigation aircraft has low speed and small volume and weak strong weather resistance; on the other hand, in a low-altitude environment, the geographic environment is complex, and the flight of the aircraft is easily influenced by terrain and obstacles.
At present, much research is carried out on the problem of planning the transportation and aviation routes, but the research on the problem of planning the navigation routes of the navigation and aviation is less. The existing research results have the following problems: (1) the flight environment of the navigation aircraft is more complex, and complex threats such as terrain obstacles and severe weather can be met; (2) uncertainty exists in prior information of dangerous areas in an airspace before route planning. For example, acquiring a severe weather region depends on weather forecast, but the possible forecasted severe weather is not as approximate; (3) the navigation flight adopts visual flight, has low speed, does not need to be limited to an airway route, has stronger flexibility and can change the route at any time.
Disclosure of Invention
In order to solve the problems of complex low-altitude environment, uncertain dangerous areas and insufficient planning flexibility in navigation flight route planning, the invention provides a real-time route planning method based on a decision tree, and the decision tree of the route with higher benefit than that of planning a single path is obtained on the premise of ensuring safety and real-time performance.
The technical scheme of the invention is as follows:
the invention provides a real-time route planning method based on a decision tree, which comprises the following steps:
1) establishing a low-altitude airspace probability model, wherein the dangerous area comprises a no-fly area and a high-risk area; representing a three-dimensional continuous space by using a discrete two-dimensional grid model, wherein each grid is a rectangular discrete grid with a certain length and width, and assigning a value to each grid point to represent a cost value of the grid point;
2) performing initial route planning, planning two paths of direct flight and fly-around for a high-risk area, planning only fly-around paths for a no-fly area, generating a route decision tree, and determining an initial route set;
3) and selecting the planned route set according to the actual situation in the flight, and updating the planning result in real time.
Preferably, in step 1), each grid point is assigned a value between 0 and 1, which represents the cost value J of the grid point,
Figure BDA0003314554280000021
in the formula pjIndicating the probability of occurrence of high-risk areas.
Preferably, the step 2) is specifically:
2.1) starting from the starting point, and planning a path towards the direction of a flight terminal point; the safety area only plans a direct flight path, and the no-flight area only plans a winding flight path;
2.2) planning two different paths of the fly-around path and the straight fly path for the first high-risk area on the path, respectively selecting a point on the fly-around path and the straight fly path as two decision-making end points of the planning, and requiring that the straight line distances between the two decision-making end points and the fly-off point are the same;
2.3) taking the decision-making terminal obtained by the previous planning as the starting point of the next planning, and planning a path towards the direction of the flight terminal respectively from each starting point; planning two different paths of the fly-around path and the straight fly path for the first high-risk area on each path, respectively selecting a point on the fly-around path and the straight fly path as two decision end points of the planning, and requiring that the straight line distances from the two decision end points to the planning start point are the same;
2.4) repeating the step 2.3) until all paths are planned to reach the flight end.
Preferably, both decision endpoints of a plan are points in the safety zone after the high risk zone for which the plan is directed.
Preferably, when two decision endpoints meeting the requirement cannot be obtained due to the existence of the next high-risk area after the high-risk area planned at a certain time, the next high-risk area is combined into the planning to be considered, and a straight flight path and a winding flight path around the high-risk area are planned and obtained.
Preferably, in step 2), the method for planning the fly-around path is an a-algorithm, and the cost function is
f(n)＝g(n)+h(n)
In the formula, f (n) represents an estimated cost function of reaching the current grid n from the starting point of each stage and then reaching the flight destination from the current grid n, g (n) is an actual cost function of reaching the current grid n from the flight starting point, and h (n) is a heuristic function and is an estimated cost function of reaching the flight destination from the current grid n; wherein the actual cost function is
Figure BDA0003314554280000031
In the formula (x)n，yn)，(x0，y0) Respectively, the two-dimensional coordinates of the current grid point n and the starting point of each stage.
Preferably, in the step 2), the heuristic function h (n) of the a method is
h(n)＝|xi-xn|+|yi-yn|
In the formula (x)n，yn)，(xt，yt) The two-dimensional coordinates of the current grid point n and the flight destination are respectively.
Preferably, the step 3) is specifically: and in the flying process, the pilot judges whether the high-risk area actually exists in real time through visual and airborne equipment, and prunes the planned route decision tree according to the judgment result.
Compared with the prior art, the real-time route planning method based on the decision tree divides the dangerous area in the low-altitude airspace into the no-fly area and the high-risk area, comprehensively considers the threat factors such as terrain barriers and severe weather, and has stronger safety; the method has the advantages that the uncertainty of the prior information of the dangerous area is considered, the air route decision tree comprising a plurality of air routes is planned, reference can be provided for the pilot before flying, the pilot can make targeted preparation in advance, and unnecessary fly-around can be reduced in actual flying; and in the flight, the planning result is updated in real time according to the actual situation, so that the characteristic of high flexibility of navigation flight can be met.
Therefore, the general aviation route planning is carried out by using the real-time route planning method based on the decision tree, and beneficial results are expected to be obtained.
Drawings
FIG. 1 is a flow chart of a decision tree based real-time route planning method of the present invention;
FIG. 2 is a schematic diagram of an embodiment of route planning;
FIG. 3 is a schematic diagram of an embodiment of a route decision tree;
fig. 4 is a schematic diagram of an embodiment of a real-time route planning method based on a decision tree.
Detailed Description
The invention will be further illustrated and described with reference to specific embodiments. The technical features of the embodiments of the present invention can be combined correspondingly without mutual conflict.
As shown in fig. 1, the real-time route planning method based on decision tree provided by the present invention includes the following steps:
the method comprises the steps of firstly, establishing a low-altitude airspace probability model, wherein a dangerous area in the model comprises a no-fly area and a high-risk area. The flight control forbidden region comprises terrain obstacles, control regions and the like; the high-risk area is mainly a severe weather area of weather forecast. And representing a three-dimensional space domain by using a two-dimensional grid model to plan a path in the horizontal direction, wherein each grid is a rectangle with a certain length and width, and each grid point is assigned with a value between 0 and 1 to represent a cost value J of the grid point.
Figure BDA0003314554280000041
In the formula pjIndicating the probability of occurrence of high-risk areas.
And step two, planning two paths of direct flight and fly-around for the high-risk area from the starting point respectively, as shown in figure 2. In the stage 1, starting from the point A, a straight flight path is planned first, and after the straight flight path passes through a first high-risk area, the straight flight path reaches the point C, wherein the straight-line distance between the point C and the point A is r1(ii) a Planning the fly-around path again, and taking the straight-line distance from the starting point A as r1Then point D is reached. The flight path passing through the high-risk area in the 1 st stage is A-C-B, and the conservative flight path completely avoiding the high-risk area is A-D-B. Points C and D are the decision endpoints of the 1 st stage and also the decision starting points of the 2 nd stage.
And step three, starting from the stage 2, planning two paths of direct flight and fly-around for the high-risk area from each initial decision point in each stage, wherein the specific method is the same as that in the step two. And repeating the steps until the direct flight path planning of the k-th stage reaches the end point, wherein the stage is the last stage.
In the above embodiment, preferably, in step two and step three, the fly-around path planning method used is an a-algorithm, and its cost function is
f(n)＝g(n)+h(n)
Wherein f (n) represents an estimated cost function from the flight starting point A to the current node n and then from the current node n to the flight terminal point B, g (n) is an actual cost function from the flight starting point A to the current node n, and h (n) is a heuristic function and is an estimated cost function from the current node n to the flight terminal point B. Wherein the actual cost function is
Figure BDA0003314554280000042
In the formula (x)n，yn)，(x0，y0) The two-dimensional coordinates of the current node n and the flight starting point A are respectively.
In the above embodiment, preferably, the heuristic function h (n) of the a method is
h(n)＝|xi-xn|+|yi-yn|
In the formula (x)n，yn)，(xt，yt) The two-dimensional coordinates of the current node n and the flight end point B are respectively.
And step four, the starting decision point of each stage has two different paths of the fly-around high-risk area and the fly-through high-risk area, so that different routes are of a binary tree structure. In general, a composition comprising 2 can be obtainedkA route decision tree for a strip path, as shown in fig. 3.
An example of a decision tree based real-time routing method is shown in fig. 4. In the figure, the upper left A is a flight starting point, and the lower right B is a flight ending point. In the figure, "squares" with different gray values represent high-risk regions (gray values from light to deep represent cost values from small to large), and "1" in the figure represents a no-fly region, and a white area in the figure represents a safety region. The "dots" in the figure represent the planned set of airlines, where the "dot" points represent the starting decision points for each phase. The number of stages k in this example is 3, where in the detour path planning of both stage 1 and stage 2, a plurality of high-risk regions are detoured. The final obtained route decision tree at most contains 23I.e. 8 lanes. In this example, since some detour path in stage 2 is the same as the end point of the traversing path, there is one branch in the route decision tree, and stage 3 has only 3 starting decision points with different positions.
Flight probability P of a route in a concentrated waynCalculated by the following method. The resulting airway decision tree contains 2kAnd (5) carrying out lane marking. In the mth route, the probability of the detour path of the ith stage is the probability p of the occurrence of the high-risk area bypassed by the stageiWhile direct and fly-around are mutually exclusive events. Thus, the probability that the path of the ith phase in the mth route will fly in actual flight is
Figure BDA0003314554280000051
Selection of routes in each phase in the mth routeSelected as an independent event. Then, the probability P of flying the m-th route in actual flightmIs composed of
Figure BDA0003314554280000052
The following table shows the flight distances and flight probabilities for each flight path in the example of fig. 4.
Parameter index Air route 1 Course 2 Air route Course 4 Air route 5 Air route 6 Route 7 Course 8
Flight distance/km 1159.5 1250.0 1347.6 1777.5 1374.1 1668.3 1434.2 1864.2
Probability of flight 0.74％ 6.68％ 7.21％ 14.57％ 5.04％ 10.18％ 18.40％ 37.18％
And fifthly, in the flying process, the pilot possibly deviates from the planned route due to the change of the operation task, or judges whether the high-risk area actually exists through means of visual observation, airborne equipment and the like, and at the moment, the route needs to be re-planned, or the planned route decision tree is pruned.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the appended claims.

Claims (8)

1. A real-time route planning method based on decision trees is characterized by comprising the following steps:
1) establishing a low-altitude airspace probability model, wherein the dangerous area comprises a no-fly area and a high-risk area; representing a three-dimensional continuous space by using a discrete two-dimensional grid model, wherein each grid is a rectangular discrete grid with a certain length and width, and assigning a value to each grid point to represent a cost value of the grid point;
2) performing initial route planning, planning two paths of direct flight and fly-around for a high-risk area, planning only fly-around paths for a no-fly area, generating a route decision tree, and determining an initial route set;
3) and selecting the planned route set according to the actual situation in the flight, and updating the planning result in real time.
2. The method as claimed in claim 1, wherein in step 1), each grid point is assigned a value between 0 and 1, which represents the cost value J of the grid point,
Figure FDA0003314554270000011
in the formula pjIndicating the probability of occurrence of high-risk areas.
3. The real-time routing method based on decision tree according to claim 1, wherein the step 2) is specifically:
2.1) planning a path from a starting point to a flight terminal direction; the safety area only plans a direct flight path, and the no-flight area only plans a winding flight path;
2.2) planning two different paths of the fly-around path and the fly-straight path for the first high-risk area on the path, respectively selecting one point on the fly-around path and the fly-straight path as two decision-making end points of the plan, and requiring that the straight line distances from the two decision-making end points to the fly-off point are the same;
2.3) taking the decision-making terminal obtained by the previous planning as the starting point of the next planning, and planning a path towards the direction of the flight terminal respectively from each starting point; planning two different paths of the fly-around path and the straight fly path for the first high-risk area on each path, respectively selecting a point on the fly-around path and the straight fly path as two decision end points of the planning, and requiring that the straight line distances from the two decision end points to the planning start point are the same;
2.4) repeating the step 2.3) until all paths are planned to reach the flight end.
4. The method of claim 3, wherein all the decision endpoints of a plan are points in a safety zone after the high risk zone for the plan.
5. The real-time routing method based on decision trees of claim 4, wherein when two decision endpoints meeting the requirements cannot be obtained due to the existence of the next high-risk area after a certain planned high-risk area, the next high-risk area is merged into the planning to be considered, and a straight flight path and a fly-around path around the high-risk area are planned.
6. The real-time routing method based on decision tree according to claim 1 or 3, wherein in the step 2), the routing method of the fly-around path is a-x algorithm, and the cost function is
f(n)＝g(n)+h(n)
In the formula, f (n) represents an estimated cost function of reaching the current grid n from the starting point of each stage and then reaching the flight destination from the current grid n, g (n) is an actual cost function of reaching the current grid n from the flight starting point, and h (n) is a heuristic function and is an estimated cost function of reaching the flight destination from the current grid n; wherein the actual cost function is
Figure FDA0003314554270000021
In the formula (x)n，yn)，(x0，y0) Respectively, the two-dimensional coordinates of the current grid point n and the starting point of each stage.
7. The method according to claim 6, wherein the heuristic function h (n) of the A method in step 2) is
h(n)＝|xi-xn|+|yi-yn|
In the formula (x)n，yn)，(xt，yt) The two-dimensional coordinates of the current grid point n and the flight destination are respectively.
8. The real-time route planning method based on decision tree according to claim 1, wherein the step 3) is specifically: and in the flying process, the pilot judges whether the high-risk area actually exists in real time through visual and airborne equipment, and prunes the planned route decision tree according to the judgment result.
CN202111227109.5A 2021-10-21 2021-10-21 Real-time route planning method based on decision tree Pending CN114034301A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111227109.5A CN114034301A (en) 2021-10-21 2021-10-21 Real-time route planning method based on decision tree

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111227109.5A CN114034301A (en) 2021-10-21 2021-10-21 Real-time route planning method based on decision tree

Publications (1)

Publication Number Publication Date
CN114034301A true CN114034301A (en) 2022-02-11

Family

ID=80135077

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111227109.5A Pending CN114034301A (en) 2021-10-21 2021-10-21 Real-time route planning method based on decision tree

Country Status (1)

Country Link
CN (1) CN114034301A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114676592A (en) * 2022-04-18 2022-06-28 北京大唐永盛科技发展有限公司 Low-altitude flight gridding management method
CN114812564A (en) * 2022-06-22 2022-07-29 北京航空航天大学杭州创新研究院 Multi-target unmanned aerial vehicle path planning method based on urban dynamic space-time risk analysis
CN116386389A (en) * 2023-03-21 2023-07-04 中国南方航空股份有限公司 Civil aviation route planning method with limit
WO2024020551A1 (en) * 2022-07-22 2024-01-25 Motional Ad Llc Corridor/homotopy scoring and validation

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102854880A (en) * 2012-10-08 2013-01-02 中国矿业大学 Robot whole-situation path planning method facing uncertain environment of mixed terrain and region
CN105629992A (en) * 2016-02-05 2016-06-01 哈尔滨工程大学 UUV navigation path planning method under threat Internet
CN107003141A (en) * 2014-10-20 2017-08-01 通腾导航技术股份有限公司 Alternative route
CN112083734A (en) * 2019-06-12 2020-12-15 中国民航大学 Collective flight path planning method using probabilistic weather forecast

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102854880A (en) * 2012-10-08 2013-01-02 中国矿业大学 Robot whole-situation path planning method facing uncertain environment of mixed terrain and region
CN107003141A (en) * 2014-10-20 2017-08-01 通腾导航技术股份有限公司 Alternative route
CN105629992A (en) * 2016-02-05 2016-06-01 哈尔滨工程大学 UUV navigation path planning method under threat Internet
CN112083734A (en) * 2019-06-12 2020-12-15 中国民航大学 Collective flight path planning method using probabilistic weather forecast

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114676592A (en) * 2022-04-18 2022-06-28 北京大唐永盛科技发展有限公司 Low-altitude flight gridding management method
CN114812564A (en) * 2022-06-22 2022-07-29 北京航空航天大学杭州创新研究院 Multi-target unmanned aerial vehicle path planning method based on urban dynamic space-time risk analysis
CN114812564B (en) * 2022-06-22 2022-09-20 北京航空航天大学杭州创新研究院 Multi-target unmanned aerial vehicle path planning method based on urban dynamic space-time risk analysis
WO2024020551A1 (en) * 2022-07-22 2024-01-25 Motional Ad Llc Corridor/homotopy scoring and validation
CN116386389A (en) * 2023-03-21 2023-07-04 中国南方航空股份有限公司 Civil aviation route planning method with limit
CN116386389B (en) * 2023-03-21 2023-12-29 中国南方航空股份有限公司 Civil aviation route planning method with limit

Similar Documents

Publication Publication Date Title
CN114034301A (en) Real-time route planning method based on decision tree
CN109814598B (en) Unmanned aerial vehicle low-altitude public navigation network design method
CN106781707B (en) A kind of path planning method for low latitude middle and long distance ferry flight
Patron et al. Flight trajectories optimization under the influence of winds using genetic algorithms
US8082102B2 (en) Computing flight plans for UAVs while routing around obstacles having spatial and temporal dimensions
CN103473956B (en) Three-dimensional arrival-departure airline network optimization method based on ant colony algorithm improvement for terminal area
CN104991895A (en) Low-altitude rescue aircraft route planning method based on three dimensional airspace grids
GB2567810A (en) Method and system for determining optimal path for drones
CN112880684A (en) Urban space unmanned aerial vehicle safety route planning method
CN103528586A (en) Sailing planning algorithm design based on grid failure
JP6947626B2 (en) UAV navigation network data generator, UAV navigation network data generation method, UAV navigation route creation device
CN109948834B (en) Unmanned aerial vehicle low-altitude public route design method
CN106774434B (en) Automatic obstacle avoidance method and system applied to unmanned aerial vehicle plant protection
CN111399543A (en) Same-region multi-collision-free air route planning method based on A-star algorithm
CN106372766A (en) UAV (Unmanned Aerial Vehicle) path planning method for electromagnetic interference environment
CN110044373A (en) A kind of fining traffic weather information on services generation method
CN1105954C (en) Route planning, terrain evading and fly environment warming system for general-purpose aviation
JP2021517290A (en) Evaluation method of congestion degree of the route based on the optimum passage route under convective weather conditions
CN112083734B (en) Method for planning integrated flight path by utilizing probabilistic weather forecast
CN112185149B (en) Path planning method and system based on urban road network data
Murrieta-Mendoza et al. Commercial aircraft lateral flight reference trajectory optimization
CN113220008B (en) Collaborative dynamic path planning method for multi-Mars aircraft
CN101982846A (en) Method for selecting optimal air line for flight under time-varying conditions
Shmelova et al. Organization of a safe and efficient system of air transportation in and around an urban area using unmanned arial vehicles
CN114659525B (en) Three-dimensional ship scene oriented extensible real-time rapid emergency path planning method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination