CN116795122A - Path optimization method for UAV-USV collaborative exploration sea area target under energy constraint - Google Patents

Abstract

The application discloses a path optimization method for a UAV-USV collaborative exploration sea area target under energy constraint, which comprises the following steps: planning the position of a charging station based on the Voronoi diagram, so that the charging station set covers all accident high-incidence points; reducing the base number of the charging station by adopting a greedy algorithm, and selecting a set covering the most accident high-incidence points from the initial charging station set until all the accident high-incidence points are covered; the position of the charging station is intensively processed to be close to the command part; performing track planning on the USV by adopting an improved ant colony algorithm, wherein the USV carries the UAV to traverse all charging stations; and performing track planning on the UAV based on an improved Lazy Theta algorithm to ensure that the obtained path is smooth. The method adopts an optimization strategy to simultaneously solve the problems of poor flexibility and limited energy of the unmanned ship, fully plays respective advantages, and effectively searches high accident occurrence points in the sea area to achieve continuous monitoring.

Description

Path optimization method for UAV-USV collaborative exploration sea area target under energy constraint

Technical Field

The application relates to the technical field of unmanned cluster collaborative search, in particular to a path optimization method for UAV-USV collaborative exploration of sea area targets under energy constraint.

Background

In recent decades, the investment in marine exploration has increased, resulting in an increasing interest in marine technology. Various marine exploration activities involve a number of dangerous and complex tasks that are often impossible to accomplish without the assistance of specialized tools. Marine robots such as unmanned vessels (USV), unmanned Aerial Vehicles (UAV) and Autonomous Underwater Vehicles (AUV) offer different detection methods suitable for a wide range of tasks and are therefore widely used in many fields. However, the single unmanned system has limited load and visual ability, often difficult to accomplish complex tasks with a large number of detection targets, and the cooperation of multiple unmanned systems can effectively solve this problem. There is no research on the use of the USV-UAV co-system for sea target exploration and sea accident monitoring.

Disclosure of Invention

According to the problems existing in the prior art, the application discloses a path optimization method for exploring sea area targets cooperatively by UAV-USV under energy constraint, which comprises the following steps:

planning the position of a charging station based on the Voronoi diagram, so that the charging station set covers all accident high-incidence points;

reducing the base number of the charging station by adopting a greedy algorithm, and selecting a set covering the most accident high-incidence points from the initial charging station set until all the accident high-incidence points are covered;

the position of the charging station is intensively processed to be close to the command part;

performing track planning on the USV by adopting an improved ant colony algorithm, wherein the USV carries the UAV to traverse all charging stations;

and performing track planning on the UAV based on an improved Lazy Theta algorithm to ensure that the obtained path is smooth.

Further, the USV is used as a mobile charging station of the energy-limited UAV, the task sea area is monitored and searched, and T is set as an accident high-incidence point set { T) of the task sea area ₁ ,...,t _n And t _i E T (i=1,., n), plans USV charging station positions from the location of the accident high-rise, let Ω be the charging station location set { c ₀ ,...,c _m And c _j E Ω (j=0, 1,., m), since UAV energy is limited, assuming a furthest flight distance of 2R, ensuring that all accident high incidence points are covered by a circle of radius R, Φ is the set of accident high incidence points covered by ΩWherein->And->Is equal to R, charging station c is planned based on the voronoi diagram _j Regarding the accident high incidence point set as a Veno point set, obtaining a Veno diagram through a Delaunay triangle network, calculating the Veno diagram, planning the position of a charging station, and ensuring that each accident high incidence point is covered by the charging station.

Further, for planned charging station c _j Optimizing the number of the initial charging station set omega so as to reduce the base number of omega, selecting the minimum charging station set which can cover all accident high-incidence points, reducing the base number of omega by adopting a greedy algorithm, and selecting the set which covers the most accident high-incidence points from the initial charging station set omega until all accident high-incidence points are covered.

Further, the USV carries the UAV from the command unit c ₀ After the UAV arrives at the charging station, the UAV starts the navigation exploration task from the USV to take a high-rise place of the accident in the sea area, the USV continues to serve as the charging station to wait for the navigation UAV in situ, the UAV bears more exploration tasks on the premise of uniform speed according to the speed of the UAV being larger than that of the USV, the command part is used as the center, the positions of the charging stations after the quantity optimization are intensively processed, and the positions are close to the command part, so that the charging station c is updated _j The aims of shortening the navigation route of the USV and enabling the UAV to bear more exploration tasks are achieved.

Further, USV from c ₀ Starting from the optimized charging station to charging station c in Ω _j All traversalsEnd return c ₀ And (3) adopting an improved ant colony algorithm to carry out track planning on the USV, and providing the following pheromone updating strategy:

τ _ij (t+n)＝ρ ¹ ·τ _ij (t)+ρ ² ·τ _ij (t+1)+ρ ⁿ ·τ _ij (t+n-1)+Δτ _ij (t)

since ρ ε [0,1 ], when (n+t) →infinity,

wherein ρ represents the pheromone evaporation coefficient, and the value is [0, 1); Δτ _ij (t) represents the pheromone increment on path [ i, j ] in this cycle; Δτ _ij ^k (t) represents the amount of information that the kth ant leaves on the path (i, j) in the present cycle.

Further, the USV sails to each charging station c as designed _j In this case, the UAV searches for an accident high-occurrence point from the mobile charging station USV by starting the vehicle, and sets S as the UAV sub-path set, s= { S ₁ ,...,s _k If UAV power is sufficient in exploration, sub-path s _i E S is denoted as { c _j ,t _i ,t _i+1 ,...,t _q ,c _j -a }; if the UAV electric quantity is insufficient in the exploration, the USV is returned to the midway for charging and then the navigation is restarted to explore, and the sub-path s is _i E S is denoted as { c _j ,t _i ,c _j ,t _i+1 ,...,t _q ,c _j Performing track planning on the UAV exploration accident high-incidence point based on an improved Lazy Theta algorithm to enable the obtained path to be flatter, and performing improved weighting on the cost function of the Lazy Theta algorithm to obtain the estimated value of the distance from the current point to the end point, wherein the estimated value of the distance from the current point to the end point has larger influence than the distance from the departure point to the current point, namely 'more want to approach the end point':

f(m)＝g(m)+ε×h(m)

wherein g (m) is the actual cost from the initial node to the current node m; h (m) is the estimated cost from the current node m to the target node; epsilon is a heuristic weighting factor (epsilon is larger than or equal to 1).

The performance function of the UAV-USV cooperative system of the application is as follows:

efficacy function F of USV _USV 、T _USV The method comprises the following steps of:

wherein the method comprises the steps ofFor USV from c _i To c _j Distance travelled->From last charging station c for USV _m Return command unit c ₀ Distance travelled, v _USV Is USV sailing speed;

performance function F of UAV _UAV 、T _UAV The method comprises the following steps:

wherein the method comprises the steps ofFor UAV from c _j To t _i A sailing distance; />To explore the shortage of electricity in the middle of UAV, it is necessary to go from t _i Return to c _j Distance of charged sailing>Indicating that UAV charges from c _j To the next accident high incidence point t _i+1 The distance of the voyage,for the UAV to have sufficient electric quantity, the UAV does not pass through a charging station and is directly from t _i Sailing to t _i+1 Is a distance of (2); />Searching for the last accident high incidence point t for UAV _p Returning to charging station c _j The distance travelled; v _UAV Is UAV sailing speed.

To sum up, the overall efficiency function F _total ，t _total The method comprises the following steps:

F _total ＝F _USV +F _UAV

t _total ＝t _UAV +t _USV

by optimizing F _total And t _total Thereby improving the efficiency of UAV-USV collaborative exploration task sea accident high-incidence points.

Due to the adoption of the technical scheme, the UAV-USV collaborative exploration sea area target path optimization method under the energy constraint fully considers the accident high-incidence point distribution condition in the sea area and reduces the resource waste, and the minimum charging station is used for covering all accident high-incidence points; the Lazy Theta algorithm and the ant colony algorithm are improved in an online part, and the advantages of the Lazy Theta algorithm and the ant colony algorithm are combined, so that the path of the USV-UAV cooperative system is optimized, the speed of planning the path is faster, and the path is smoother; the optimization strategy can solve the problems of poor flexibility and limited energy of the unmanned ship at the same time, give full play to respective advantages, and effectively search high accident occurrence points in the sea area so as to achieve continuous monitoring.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings may be obtained according to the drawings without inventive effort to those skilled in the art.

FIG. 1 is a diagram of a mission sea area of the present application

FIG. 2 is a step diagram of the CASROA algorithm according to the application

FIG. 3 is a schematic diagram of a charging station plan according to the present application

FIG. 4 is a diagram showing the number of charging stations according to the present application

FIG. 5 is a step diagram of the charging station position optimization of the present application

FIG. 6 shows the charging station position after the position optimization according to the application

FIG. 7 is a simulation of the USV path planning of the application

Fig. 8 is an iteration diagram of the improved ant colony algorithm in the present application

FIG. 9 is a flow chart of the improved Lazy Theta algorithm of the present application

FIG. 10 is a simulation diagram of UAV path planning in accordance with the present application

Detailed Description

In order to make the technical scheme and advantages of the present application more clear, the technical scheme in the embodiment of the present application is clearly and completely described below with reference to the accompanying drawings in the embodiment of the present application:

as shown in figure 1, the task sea area comprises a reef area, a cultivation area, a small ship gathering area and the like, the reef area, the congestion, even collision and other accidents are extremely easy to occur in the areas, once the accidents occur, a cruising ship is difficult to directly drive in, and an unmanned aerial vehicle has high-efficiency exploration monitoring capability but limited energy consumption, so that the USV-UAV is adopted to search high-incidence points of the accidents on the basis of a CASROA algorithm in cooperation, and the specific method comprises the following steps:

1. planning the position of a charging station

The position of the charging station is planned based on a voronoi diagram, which is obtained through a Delaunay triangle network, which in turn is composed of a plurality of triangles, wherein each vertex of the triangles belongs to a set of voronoi points. Therefore, the accident high-occurrence point set is regarded as a Veno point set, and the Veno diagram planning charge is calculatedStation position c _j Each accident high-incidence point can be guaranteed to be in the Veno point set, namely, all the accident high-incidence points are covered by the charging station. Simulation of planning charging station positions based on a voronoi diagram is shown in fig. 3. Fig. 3 shows that the method of generating the voronoi diagram through multiple iterations can cause the situation that the same accident high-rise point is repeatedly covered by multiple charging stations, so that serious resource waste is caused, and further optimization is necessary.

2. Optimizing the number of charging stations

And adopting a greedy algorithm to reduce the omega base number, and selecting a set covering the most accident high-incidence points from the initial charging station set omega until all the accident high-incidence points are covered. The charging station simulation after the number optimization is shown in fig. 4. It can be seen from fig. 3 and 4 that the minimum number of charging stations can be used after optimization, and the exploration of all accident high points can be realized by cooperating with the UAV.

3. Optimizing charging station position

Let E be the new charging station location set { E ₀ ,...,e _m And e _j E (j=0, 1,., m) with the aim of, for any E _j All should satisfy:

(1)d(e _j ,c _o )<d(c _j ,c ₀ )；

(2) For any t _i ，

With 4 accident high incidence cases, the mapping process for solving the new charging station position is shown in fig. 5: a circle formed by taking the accident high-incidence point as the center of a circle and R as the radius is connected with a line formed by the command part and the charging station, and the intersection point of the command part and the charging station is a new charging point e _j . The optimized charging station position is shown in fig. 6.

Compared with fig. 4, the charging station position shown in fig. 6 obviously moves from the center position to the direction of the command part, the distance between the charging stations becomes smaller, and the F is increased _UAV Means of (2) reducing F _USV Is a target of (a).

4. USV track planning

And adopting an improved ant colony algorithm to carry out track planning on the USV, wherein the traditional ant colony algorithm pheromone updating strategy is as follows:

τ _ij (t+n)＝ρ·τ _ij (t)+Δτ _ij (t)

wherein ρ represents the pheromone evaporation coefficient, and the value is [0, 1); Δτ _ij (t) represents the pheromone increment on path [ i, j ] in this cycle; Δτ _ij ^k (t) represents the amount of information that the kth ant leaves on the path (i, j) in the present cycle.

The reference of the update strategy to the initial information is only limited to the time t, the actual value of the pheromone is reduced, the synergetic effect among ants is reduced, and a new pheromone update strategy is provided in the section for solving the problem:

τ _ij (t+n)＝ρ ¹ ·τ _ij (t)+ρ ² ·τ _ij (t+1)+ρ ⁿ ·τ _ij (t+n-1)+Δτ _ij (t)

and (3) recursively solving the following steps:

τ _ij (t+1)＝ρ·τ _ij (t)+Δτ _ij (t)

τ _ij (t+2)＝(ρ+ρ ³ )·τ _ij (t)+(1+ρ ² )Δτ _ij (t)

τ _ij (t+3)＝(ρ+ρ ³ +ρ ⁴ +ρ ⁶ )·τ _ij (t)+(1+ρ ² +ρ ³ +ρ ⁵ )Δτ _ij (t)

…

since ρ ε [0,1 ], when (n+t) →infinity,

ant colony algorithm based on improvementSimulation of the track planning for USV is shown in fig. 7. When the number of charging stations j=14, the iteration curve of the ant colony algorithm is as shown in fig. 8 (1), and the ant colony algorithm and the modified ant colony algorithm are respectively iterated for about 18 times and 45 times F _USV Tend to stabilize; when j=29, the iteration curve of the ant colony algorithm is as shown in fig. 8 (2), and the ant colony algorithm and the modified ant colony algorithm are respectively iterated for about 19 times and 90 times F _USV Tend to stabilize.

The improved ant colony algorithm effectively enhances the synergistic effect among ants. In a more complex task sea area, the accident points occur more times, more charging stations are needed, so that the USV sailing distance becomes longer, and the improved ant colony algorithm can show the superiority due to the higher iteration speed.

5. UAV track planning: and performing track planning on the UAV based on an improved Lazy Theta algorithm, so that the obtained path is flatter.

Assuming that m is the current node, mnbr is the neighbor node, neighbor-lists is the neighbor node list. The Lazy Theta algorithm cost function is improved herein as follows:

f(m)＝g(m)+ε×h(m)

g (m): actual cost from the initial node to the current node m;

h (m): estimating a cost from the current node m to the target node;

epsilon: heuristic weighting factor (. Epsilon. Gtoreq.1).

Weighting the heuristic means that the estimate of the distance from the current point to the end point has a greater influence (since ε. Gtoreq.1), i.e. "more desirable to approach the end point", than the distance from the departure point to the current point. A flow chart of the modified Lazy Theta algorithm for trajectory planning for UAV is shown in fig. 9 and UAV trajectory is shown in fig. 10.

The foregoing is only a preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art, who is within the scope of the present application, should make equivalent substitutions or modifications according to the technical scheme of the present application and the inventive concept thereof, and should be covered by the scope of the present application.

Claims (6)

1. A path optimization method for cooperatively exploring a sea area target by using UAV-USV under energy constraint is characterized by comprising the following steps:

planning the position of a charging station based on the Voronoi diagram, so that the charging station set covers all accident high-incidence points;

reducing the base number of the charging station by adopting a greedy algorithm, and selecting a set covering the most accident high-incidence points from the initial charging station set until all the accident high-incidence points are covered;

the position of the charging station is intensively processed to be close to the command part;

performing track planning on the USV by adopting an improved ant colony algorithm, wherein the USV carries the UAV to traverse all charging stations;

and performing track planning on the UAV based on an improved Lazy Theta algorithm to ensure that the obtained path is smooth.

2. The method for optimizing the path of the UAV-USV collaborative exploration sea area target under the energy constraint according to claim 1, wherein the method comprises the following steps: using USV as a mobile charging station of an energy-limited UAV, performing monitoring search on a mission sea area, and setting T as an accident high-incidence point set { T) of the mission sea area ₁ ,...,t _n And t _i E T (i=1,., n), plans USV charging station positions from the location of the accident high-rise, let Ω be the charging station location set { c ₀ ,...,c _m And c _j E Ω (j=0, 1,., m), since UAV energy is limited, assuming a furthest flight distance of 2R, ensuring that all accident high incidence points are covered by a circle of radius R, Φ is the set of accident high incidence points covered by ΩWherein->And->Is equal to R, charging station c is planned based on the voronoi diagram _j Regarding the accident high incidence point set as a Veno point set, obtaining a Veno diagram through a Delaunay triangle network, calculating the Veno diagram, planning the position of a charging station, and ensuring that each accident high incidence point is covered by the charging station.

3. The method for optimizing the path of the UAV-USV collaborative exploration sea area target under the energy constraint according to claim 2, wherein the method comprises the following steps: for planned charging station c _j Optimizing the number of the initial charging station set omega so as to reduce the base number of omega, selecting the minimum charging station set which can cover all accident high-incidence points, reducing the base number of omega by adopting a greedy algorithm, and selecting the set which covers the most accident high-incidence points from the initial charging station set omega until all accident high-incidence points are covered.

4. A method for path optimization of UAV-USV collaborative exploration sea area targets under energy constraint as set forth in claim 3, wherein: USV-mounted UAV slave command unit c ₀ After the UAV arrives at the charging station, the UAV starts the navigation exploration task from the USV to take a high-rise place of the accident in the sea area, the USV continues to serve as the charging station to wait for the navigation UAV in situ, the UAV bears more exploration tasks on the premise of uniform speed according to the speed of the UAV being larger than that of the USV, the command part is used as the center, the positions of the charging stations after the quantity optimization are intensively processed, and the positions are close to the command part, so that the charging station c is updated _j The aims of shortening the navigation route of the USV and enabling the UAV to bear more exploration tasks are achieved.

5. The method for optimizing the path of the UAV-USV collaborative exploration sea area target under the energy constraint according to claim 4, wherein the method comprises the following steps: USV from c ₀ Starting from the optimized charging station to charging station c in Ω _j Full traversal end return c ₀ And (3) adopting an improved ant colony algorithm to carry out track planning on the USV, and providing the following pheromone updating strategy:

τ _ij (t+n)＝ρ ¹ ·τ _ij (t)+ρ ² ·τ _ij (t+1)+ρ ⁿ ·τ _ij (t+n-1)+Δτ _ij (t)

since ρ ε [0,1 ], when (n+t) →infinity,

wherein ρ represents the pheromone evaporation coefficient, and the value is [0, 1); Δτ _ij (t) represents the pheromone increment on path [ i, j ] in this cycle; Δτ _ij ^k (t) represents the amount of information that the kth ant leaves on the path (i, j) in the present cycle.

6. The method for optimizing the path of the UAV-USV collaborative exploration sea area target under the energy constraint according to claim 1, wherein the method comprises the following steps: the USV sails to each charging station c according to the design _j In this case, the UAV searches for an accident high-occurrence point from the mobile charging station USV by starting the vehicle, and sets S as the UAV sub-path set, s= { S ₁ ,...,s _k If UAV power is sufficient in exploration, sub-path s _i E S is denoted as { c _j ,t _i ,t _i+1 ,...,t _q ,c _j -a }; if the UAV electric quantity is insufficient in the exploration, the USV is returned to the midway for charging and then the navigation is restarted to explore, and the sub-path s is _i E S is denoted as { c _j ,t _i ,c _j ,t _i+1 ,...,t _q ,c _j And performing track planning on the UAV exploration accident high-incidence point based on an improved Lazy Theta algorithm, so that the obtained path is more gentle, and performing the following improved weighting on the cost function of the Lazy Theta algorithm:

f(m)＝g(m)+ε×h(m)

wherein g (m) is the actual cost from the initial node to the current node m, h (m) is the estimated cost from the current node m to the target node, and epsilon is a heuristic weighting factor (epsilon is more than or equal to 1);

wherein the performance function of the UAV-USV cooperative system is as follows:

efficacy function F of USV _USV 、T _USV The method comprises the following steps of:

wherein the method comprises the steps ofFor USV from c _i To c _j Distance travelled->From last charging station c for USV _m Return command unit c ₀ Distance travelled, v _USV Is USV sailing speed;

performance function F of UAV _UAV 、T _UAV The method comprises the following steps:

wherein the method comprises the steps ofFor UAV from c _j To t _i A sailing distance; />To explore the shortage of electricity in the middle of UAV, it is necessary to go from t _i Return to c _j The distance of the charged navigation is set,/>indicating that UAV charges from c _j To the next accident high incidence point t _i+1 Distance travelled,/->For the UAV to have sufficient electric quantity, the UAV does not pass through a charging station and is directly from t _i Sailing to t _i+1 Is a distance of (2); />Searching for the last accident high incidence point t for UAV _p Returning to charging station c _j The distance travelled; v _UAV For the speed of the UAV's voyage,

to sum up, the overall efficiency function F _total ，t _total The method comprises the following steps:

F _total ＝F _USV +F _UAV

t _total ＝t _UAV +t _USV

by optimizing F _total And t _total Thereby improving the efficiency of UAV-USV collaborative exploration task sea accident high-incidence points.