CN103697895A - Method for determining optimal path of flight vehicle based on self-adaptive A star algorithm - Google Patents

Abstract

The invention discloses a method for determining an optimal path of a flight vehicle based on a self-adaptive A star algorithm, which is mainly used for solving the problems of excessively long length of the divided path, excessively large threat value of an obstacle to the flight vehicle and excessively large number of waypoints of the prior art. The method comprises the following implementation steps: (1) setting a start point SP and an end point EP of the flight vehicle in a map; (2) generating an obstacle region; (3) performing grid division on the obstacle region to obtain a grid map; (4) setting a grid where the start point SP of the flight vehicle is located as SG and setting the grid where the end point EP is located as EG in the grid map; (5) performing path planning according to the grid SG where the start point of the flight vehicle is located and the grid EG where the end point is located. The method disclosed by the invention has the advantages of short path length, low threat value and a small number of waypoints, and can be used for determining the optimal path of the flight vehicle.

Description

Aircraft optimal path based on self-adaptation A star algorithm is determined method

Technical field

The invention belongs to spatial data processing technology field, the particularly path planning of aircraft, the method can be used for carrying out the path planning in intelligent navigation, the field such as unmanned.

Background technology

Aircraft is to rely on terrain information and enemy's situation information to carry out path planning, and, under certain constraint condition, aircraft is from starting point to terminating point, and search can preferentially meet the flight path of specific purpose optimum, and wherein flight path represents with a series of way points.

Along with the development of military technology, actual war demand, the path planning modeling of aircraft is from single System Development to system complicated and changeable.The path planning problem of aircraft presents real-time, multi-dimensional nature, uncertainty and the complex characteristic such as non-linear.Be based upon the model under legacy system, still get used to the angle known and that can not change from barrier terrain information, with static, be idealized as the modeling problem that main mode is carried out exploratory flight device path planning, ignored from actual angle and treated whole flight course.Most paths planning method is poor for the barrier landform scene effect of pressing close to true environment of complexity.This is to be simple static planing method because traditional algorithm adopts, at practical flight device, before flight, according to barrier landform scene, carry out path planning, in flight course, the immutable and barrier in path meets specific rule, and the gap of this and real scene is larger.Obvious this planing method has been carried out Utopian processing by barrier landform scene, and in fact the terrain information of flying scene is irregular random.

Mainly contain at present three types of technology and solve path planning problem: 1. traditional algorithm, as Grid Method, Voronoi figure method; 2. intelligent optimization algorithm, as genetic algorithm; 3. other algorithms, as dynamic programming algorithm.

Traditional algorithm is to calculate according to classical mathematical equation, the path that more desirable barrier map calculates with this type of algorithm is more excellent in the overall situation, but comparatively idealized to the requirement of barrier, for the planning effect of actual landform map, be largely subject to the impact of algorithm itself and the idealized degree of barrier.Intelligent optimization algorithm, if genetic algorithm is to be proposed by John Holland in 1975, it is the evolution laws that a class is used for reference organic sphere, the randomization searching method that utilizes the survival of the fittest, survival of the fittest genetic mechanism to develop and come, its principal feature is directly structure objects to be operated, and does not have the restriction of differentiate and continuous; There is inherent Implicit Parallelism and better global optimizing ability; Adopt the optimization method of randomization, the search volume that energy automatic acquisition and guidance are optimized, adjusts the direction of search adaptively, does not need the rule of determining.In robustness, compared with traditional algorithm, be significantly improved, the Utopian degree of dependence of barrier is lowered, but the computing time of intelligent optimization algorithm be generally long, cannot adapt to dynamically, the variable map that more approaches real situation carries out path planning.For other algorithms, as dynamic programming algorithm, due to algorithm self restriction, on local path, can reach optimum and adapt to dynamic map, but cannot guarantee in overall situation planning, may there is the unstable situation with can not calculating overall path of route result.

Summary of the invention

The object of the invention is to the deficiency for prior art, propose a kind of aircraft optimal path based on self-adaptation A star algorithm and determine method, to improve the correctness in the path that multiple different maps are cooked up, improve the stability of same width map being carried out repeatedly to the route of path planning simultaneously, strengthen the robustness to different map path plannings.

Technical thought of the present invention is: required Various types of data is carried out to pre-service, obtain terminal and barrier region coordinate, by terminal coordinate difference ratio, calculate ordinate direction size of mesh opening size, the path that draws its path planning by traditional A star algorithm, and first way point when the direction of search changes is as new starting point, redefine size of mesh opening size, by A star algorithm, search for, when meeting terminating point condition, path planning finishes, and exports all effective way points.Its concrete steps comprise as follows:

(1) starting point SP and the terminating point EP of aircraft are set in map;

(2) dyspoiesis object area:

2a), according to the position of the starting point SP of aircraft and terminating point EP, set up rectangular coordinate system;

2b) establish p _i, q _ibe respectively the starting point of each barrier and the coordinate of terminal that aircraft need be walked around, to p _ipoint is done respectively the straight line perpendicular to rectangular coordinate system X-axis and Y-axis, to q _ipoint is done respectively the straight line perpendicular to rectangular coordinate system X-axis and Y-axis, and the rectangle surrounding with these four straight lines represents barrier region T _i, the number that wherein i is barrier;

(3) barrier region is carried out to grid division, obtain grid map;

(4), in grid map, the grid of establishing the starting point SP place of aircraft is that the grid at SG, terminating point EP place is EG;

(5) according to the starting point place grid SG of aircraft and terminating point place grid EG, carry out path planning:

5a) making the grid SG at the starting point SP place of aircraft is father node D _j, establish general plan displacement and be: E _j+1=L _b+ E _j, L wherein _bfor father node D _jto the required path of walking of each the grid b being adjacent, i.e. father node D _jto the length of each the grid b line of centres being adjacent, the label that b is adjacent mesh, 1≤b≤8, E _jfor arriving father node D from starting point place grid SG _jthe distance moving, j is father node D _jlabel; Set father node D _jwhen the grid SG of starting point place, initial value is: j=0, L=0, E _j=0;

5b) calculating respectively the pre-estimation that first adjacent mesh arrives terminating point place grid EG by this two paths of level-vertical translation and vertical-horizontal translation expends: H _n=R (k) * m, and the minimum value that these two pre-estimations are expended are designated as HM, wherein R (k) by translation process the grid number of process, k by translation process the grid label of process, n is path label, n is integer, 1≤n≤2, m is map amplification coefficient, 1≤m≤20;

5c) to father node D _jeach adjacent grid repeating step (5b), obtains the minimum value HM that pre-estimation separately expends _b;

5d) calculate and father node D _jeach adjacent grid b is to the total estimates F (b) of the grid EG at terminating point place:

F(b)＝G(b)+HM _b；

Wherein G (b) is total mobile expending, and its value equates with general plan displacement, always moves and expends: G (b)=E _j+1;

5e) using the corresponding grid x of the minimum value in total estimates F (b) as new father node D _j+1, general plan displacement is: E _j+1=L _x+ E _j, L wherein _xfor father node D _jto the length of the grid x line of centres being adjacent, current plan mobile route is father node D _jwith new father node D _j+1the line of the grid element center at place;

5f) judge new father node D _j+1with father node D _jthe direction of search whether identical, if the direction of search is identical, return step (5b) proceed search; Otherwise, execution step (5g);

5g) by the new father node D in step (5f) _j+1as new starting point SP, and judge whether the distance of this point and terminating point EP is less than 2, if distance is less than 2, stop path planning, now, in grid map, the grid SG at the starting point SP place of aircraft arrives the path of the grid EG at terminating point EP place, is all father node D in grid map _twith D _t+1the line segment that the grid element center line at place forms, 0≤t≤j wherein, otherwise, return to step (3).

The present invention has the following advantages compared with prior art:

The first, the present invention adopts self-adaptation A star algorithm, compares traditional A star algorithm, aircraft is carried out to the drawn path of path planning and in length, shorten to some extent, and the air route minimizing of counting, has improved the correctness in the path of cooking up.

The second, the present invention compares with traditional algorithm, and aircraft can carry out real-time update to current map according to the variation of surrounding environment in flight course, has shortened path, and in this process, has reduced the threat value of barrier to aircraft;

The 3rd, the present invention compares other algorithms, owing to barrier topographic map not being carried out to any idealized processing, has overcome the shortcoming of the robustness deficiency in the path of cooking up for different obstruct thing topographic map.

Accompanying drawing explanation

Fig. 1 is realization flow figure of the present invention;

Fig. 2 is that the present invention utilizes self-adaptation A star algorithm to carry out the sub-process figure of path planning;

Fig. 3 is the barrier topographic map that the present invention tests;

Fig. 4 is the route result figure that the inventive method is cooked up;

Fig. 5 is the route result figure that traditional A star algorithm is cooked up.

Embodiment

Below in conjunction with accompanying drawing, step of the present invention is described in further detail.

With reference to Fig. 1, performing step of the present invention is as follows:

Step 1. arranges the starting point and ending point position of aircraft, in map, the starting point SP of aircraft and terminating point EP is carried out to mark.

Step 2. dyspoiesis object area.

2a), according to the position of the starting point SP of aircraft and terminating point EP, set up rectangular coordinate system;

2b) establish p _i, q _ibe respectively the starting point of each barrier and the coordinate of terminal that aircraft need be walked around, to p _ipoint is done respectively the straight line perpendicular to rectangular coordinate system X-axis and Y-axis, to q _ipoint is done respectively the straight line perpendicular to rectangular coordinate system X-axis and Y-axis, and the rectangle surrounding with these four straight lines represents barrier region T _i, the number that wherein i is barrier;

Step 3. is carried out grid division by barrier region, obtains grid map.

3a) setting each grid directions X length of side is 2, according to the coordinate difference ratio of starting point SP and terminating point EP, the length of side Y of computing grid Y-direction _l:

$Y_l=2*\left|\frac{y_{\mathrm{ep}}-y_{\mathrm{sp}}}{x_{\mathrm{ep}}-x_{\mathrm{sp}}}\right|;$

(x wherein _sp, y _sp) be the coordinate of starting point SP, (x _ep, y _ep) be the coordinate of terminating point EP;

3b), according to the size of each grid, barrier region is divided into and contains 100 * 100 grids;

3c) in all grids of barrier region, the grid value that barrier is covered is all made as 1, and other grid values are made as 0, obtain grid map.

Step 4. is in grid map, and the grid of establishing the starting point SP place of aircraft is that the grid at SG, terminating point EP place is EG.

Step 5., according to the starting point place grid SG of aircraft and terminating point place grid EG, is carried out path planning.

With reference to Fig. 2, the concrete steps of this step are:

5a) making the grid SG at the starting point SP place of aircraft is father node D _j, establish general plan displacement and be: E _j+1=L _b+ E _j, L wherein _bfor father node D _jto the required path of walking of each the grid b being adjacent, i.e. father node D _jto the length of each the grid b line of centres being adjacent, the label that b is adjacent mesh, 1≤b≤8, E _jfor arriving father node D from starting point place grid SG _jthe distance moving, j is father node D _jlabel; Set father node D _jwhen the grid SG of starting point place, initial value is: j=0, L=0, E _j=0;

5b) calculating respectively the pre-estimation that first adjacent mesh arrives terminating point place grid EG by this two paths of level-vertical translation and vertical-horizontal translation expends: H _n=R (k) * m, and the minimum value that these two pre-estimations are expended are designated as HM, wherein R (k) by translation process the grid number of process, k by translation process the grid label of process, n is path label, n is integer, 1≤n≤2, m is map amplification coefficient, 1≤m≤20;

5c) to father node D _jeach adjacent grid repeating step (5b), obtains the minimum value HM that pre-estimation separately expends _b;

5d) calculate and father node D _jeach adjacent grid b is to the total estimates F (b) of the grid EG at terminating point place:

F(b)＝G(b)+HM _b；

Wherein G (b) is total mobile expending, and its value equates with general plan displacement, always moves and expends: G (b)=E _j+1;

5e) using the corresponding grid x of the minimum value in total estimates F (b) as new father node D _j+1, general plan displacement is: E _j+1=L _x+ E _j, L wherein _xfor father node D _jto the length of the grid x line of centres being adjacent, current plan mobile route is father node D _jwith new father node D _j+1the line of the grid element center at place;

5f) judge new father node D _j+1with father node D _jthe direction of search whether identical, if the direction of search is identical, return step (5b) proceed search; Otherwise, execution step (5g);

5g) by the new father node D in step (5f) _j+1as new starting point SP, and judge whether the distance of this point and terminating point EP is less than 2, if distance is less than 2, stop path planning, now, in grid map, the grid SG at the starting point SP place of aircraft arrives the path of the grid EG at terminating point EP place, is all father node D in grid map _twith D _t+1the line segment that the grid element center line at place forms, 0≤t≤j wherein, otherwise, return to step (3).

Effect of the present invention further illustrates by following emulation.

One, simulated conditions

Hardware platform is: Intel Core2Duo CPU E6550@2.33GHZ, 2GB RAM

Software platform is: VC++6.0

As shown in Figure 3, wherein the square region in Fig. 3 (a) is barrier region to the barrier topomap that experiment is used, the starting point SP that the annulus in the upper left corner is aircraft, and the terminating point EP that the annulus in the lower right corner is aircraft, other regions are clear region.Two square region in Fig. 3 (b) are barrier region, the starting point SP that the annulus in the upper left corner is aircraft, and the terminating point EP that the annulus in the lower right corner is aircraft, other regions are clear region.Two border circular areas in Fig. 3 (c) are barrier region, the starting point SP that the annulus in the upper left corner is aircraft, and the terminating point EP that the annulus in the lower right corner is aircraft, other regions are clear region.Circle and square region in Fig. 3 (d) are barrier region, the starting point SP that the annulus in the upper left corner is aircraft, and the terminating point EP that the annulus in the lower right corner is aircraft, other regions are clear region.

Two, emulation content

Emulation one, by the inventive method respectively to Fig. 3 (a)～3(d) carry out path planning, obtain route programming result as Fig. 4, wherein in Fig. 4 (a), the solid line from starting point SP to terminating point EP is that Fig. 3 (a) is planned to the path obtaining, solid line from starting point SP to terminating point EP in Fig. 4 (b) is that Fig. 3 (b) is planned to the path obtaining, solid line from starting point SP to terminating point EP in Fig. 4 (c) is that Fig. 3 (c) is planned to the path obtaining, solid line from starting point SP to terminating point EP in Fig. 4 (d) is that Fig. 3 (d) is planned to the path obtaining.

Emulation two, by traditional A star algorithm respectively to Fig. 3 (a)～3(d) carry out path planning, the route programming result obtaining is as Fig. 5, wherein in Fig. 5 (a), the solid line from starting point SP to terminating point EP is that Fig. 3 (a) is planned to the path obtaining, solid line from starting point SP to terminating point EP in Fig. 5 (b) is that Fig. 3 (b) is planned to the path obtaining, solid line from starting point SP to terminating point EP in Fig. 5 (c) is that Fig. 3 (c) is planned to the path obtaining, solid line from starting point SP to terminating point EP in Fig. 5 (d) is that Fig. 3 (d) is planned to the path obtaining.

Three, interpretation of result

Respectively the path of cooking up in Fig. 4 and Fig. 5, barrier are counted and arranged the threat value of aircraft and air route, the result obtaining is as table 1, and wherein threat value is the threaten degree of barrier region to aircraft, according to radar equation D=1/d ⁴calculate, wherein d is that aircraft is apart from the short lines distance of barrier.

Table 1

As can be seen from Table 1, no matter the present invention is to count and be all better than traditional A star algorithm in path, threat value or air route, has proved fully superiority of the present invention and robustness.

Claims (2)

1. the aircraft optimal path based on self-adaptation A star algorithm is determined a method, comprises the steps:

(1) starting point SP and the terminating point EP of aircraft are set in map;

(2) dyspoiesis object area:

2a), according to the position of the starting point SP of aircraft and terminating point EP, set up rectangular coordinate system;

2b) establish p _i, q _ibe respectively the starting point of each barrier and the coordinate of terminal that aircraft need be walked around, to p _ipoint is done respectively the straight line perpendicular to rectangular coordinate system X-axis and Y-axis, to q _ipoint is done respectively the straight line perpendicular to rectangular coordinate system X-axis and Y-axis, and the rectangle surrounding with these four straight lines represents barrier region T _i, the number that wherein i is barrier;

(3) barrier region is carried out to grid division, obtain grid map;

(4), in grid map, the grid of establishing the starting point SP place of aircraft is that the grid at SG, terminating point EP place is EG;

(5) according to the starting point place grid SG of aircraft and terminating point place grid EG, carry out path planning:

5a) making the grid SG at the starting point SP place of aircraft is father node D _j, establish general plan displacement and be: E _j+1=L _b+ E _j, L wherein _bfor father node D _jto the required path of walking of each the grid b being adjacent, i.e. father node D _jto the length of each the grid b line of centres being adjacent, the label that b is adjacent mesh, 1≤b≤8, E _jfor arriving father node D from starting point place grid SG _jthe distance moving, j is father node D _jlabel; Set father node D _jwhen the grid SG of starting point place, initial value is: j=0, L=0, E _j=0;

5b) calculating respectively the pre-estimation that first adjacent mesh arrives terminating point place grid EG by this two paths of level-vertical translation and vertical-horizontal translation expends: H _n=R (k) * m, and the minimum value that these two pre-estimations are expended are designated as HM, wherein R (k) by translation process the grid number of process, k by translation process the grid label of process, n is path label, n is integer, 1≤n≤2, m is map amplification coefficient, 1≤m≤20;

5c) to father node D _jeach adjacent grid repeating step (5b), obtains the minimum value HM that pre-estimation separately expends _b;

5d) calculate and father node D _jeach adjacent grid b is to the total estimates F (b) of the grid EG at terminating point place:

F(b)＝G(b)+HM _b；

Wherein G (b) is total mobile expending, and its value equates with general plan displacement, always moves and expends: G (b)=E _j+1;

5e) using the corresponding grid x of the minimum value in total estimates F (b) as new father node D _j+1, general plan displacement is: E _j+1=L _x+ E _j, L wherein _xfor father node D _jto the length of the grid x line of centres being adjacent, current plan mobile route is father node D _jwith new father node D _j+1the line of the grid element center at place;

5f) judge new father node D _j+1with father node D _jthe direction of search whether identical, if the direction of search is identical, return step (5b) proceed search; Otherwise, execution step (5g);

5g) by the new father node D in step (5f) _j+1as new starting point SP, and judge whether the distance of this point and terminating point EP is less than 2, if distance is less than 2, stop path planning, now, in grid map, the grid SG at the starting point SP place of aircraft arrives the path of the grid EG at terminating point EP place, is all father node D in grid map _twith D _t+1the line segment that the grid element center line at place forms, 0≤t≤j wherein, otherwise, return to step (3).

2. the aircraft optimal path based on self-adaptation A star algorithm according to claim 1 is determined method, and what wherein step (3) was described carries out grid division by barrier region, obtains grid map, carries out as follows:

3a) setting the grid directions X length of side is 2, according to the coordinate difference ratio of starting point SP and terminating point EP, the length of side Y of computing grid Y-direction _l:

$Y_l=2*\left|\frac{y_{\mathrm{ep}}-y_{\mathrm{sp}}}{x_{\mathrm{ep}}-x_{\mathrm{sp}}}\right|;$

(x wherein _sp, y _sp) be the coordinate of starting point SP, (x _ep, y _ep) be the coordinate of terminating point EP;

3b) in barrier region, the grid value that barrier is covered is all made as 1, and other grid values are all made as 0, obtain grid map.

Images