CN116560408A - A sequential convex optimization method and system for time-optimal trajectory of UAV - Google Patents

Abstract

The invention discloses a sequence convex optimization method and a system for an unmanned aerial vehicle time optimal track, which relate to the technical field of unmanned aerial vehicle path planning and comprise the following steps: acquiring unmanned plane parameter information, path constraint information, task environment information and algorithm parameter information; obtaining barrier information in an actual environment, constructing a grid chart of the actual environment, and uniformly dividing the map into grids with equal proportion; obtaining an unmanned aerial vehicle track optimization configuration space according to the real environment grid diagram; searching in the unmanned aerial vehicle track optimization configuration space by adopting a path planning algorithm based on A to obtain a feasible unmanned aerial vehicle navigation path; and obtaining a safe flight track with optimal time based on the SFC-SCP algorithm. The invention has the advantages of short navigation time, strong obstacle avoidance capability and short algorithm time consumption.

Description

A sequential convex optimization method and system for time-optimal trajectory of UAV

technical field

The present invention relates to the technical field of UAV path planning, in particular to a sequential convex optimization method and system for UAV time optimal trajectory.

Background technique

With the rapid development of science and technology, UAVs are applied to various fields, such as exploration, reconnaissance, search and rescue. However, due to the complexity of the real environment and the limitations of the UAV itself, traditional trajectory planning is difficult to ensure the safety and success rate of the UAV; A feasible path that meets and optimizes a specific performance index.

For the UAV time optimal trajectory optimization problem, in the prior art, the non-convex problem is usually transformed into a convex problem, and then the sequential convex optimization method is applied to achieve a fast solution. However, it is difficult to quickly obtain a feasible solution in a dense and complex environment with many obstacles, and sequential convex optimization is sensitive to the initial value, and may fall into a local minimum in a complex environment, resulting in the failure of trajectory iteration to converge; therefore, it is urgent to design a new UAV time optimal trajectory optimization method to solve the above technical problems.

Contents of the invention

In order to solve the above problems, the present invention provides a sequential convex optimization method for the time optimal trajectory of the UAV, comprising the following steps:

Obtain the obstacle information of the UAV in the actual environment, construct the real environment grid map, perform path planning through the A* algorithm, and generate the path planning configuration space of the UAV;

The trajectory sequence convex optimization algorithm based on SFC-SCP is used, and the flight time between trajectory points is used as the optimization variable to search in the path planning configuration space to obtain a time-optimal safe flight trajectory.

Preferably, in response to the construction process of the real environment grid map, by obtaining the initial state and terminal state of the UAV, state and control constraint boundaries, threat location information, path discrete points, safe flight corridor boundary constraints, initial trust region radius, Convergence threshold, establish an environment map for the UAV time optimal trajectory optimization problem, and randomly generate obstacles in the environment map;

Based on the environment map and obstacle information, the real environment grid map is constructed, and the map is evenly divided into grids of equal size.

Preferably, in response to the path planning process through the A* algorithm, based on the environment map, through the A* algorithm, obtain a shortest path from the starting point to the end point, and generate an initial reference trajectory of the UAV, wherein the initial reference trajectory represents for:

Among them, X ₀ represents the obstacle avoidance path using the A* algorithm as the reference trajectory, and the reference trajectory X ^q in subsequent iterations is the solution result of the previous iteration; q represents the qth iteration in sequential convex optimization, and X ^q represents the qth The solution result of sequence iteration;

According to the initial reference trajectory and obstacle information, the path planning configuration space is constructed.

Preferably, in response to the construction process of the path planning configuration space, based on the initial reference trajectory, combined with surrounding obstacles, a set of half-plane constraints corresponding to each path point is obtained and assigned to each trajectory point, wherein the half-plane constraint Collections are represented as:

l(i, m)=-H(i, m)×J(i, m), i=1, 2,..., N; m=1, 2,..., M

In the formula, H(i, m) is the semi-plane unit normal vector generated between the i-th path point and the m-th cylindrical obstacle, and J(i, m) is the nearest path point on the m-th circular obstacle to the i-th path point A point coordinate of and /> Indicates the position and radius of the center point and radius of the horizontal plane of the mth cylinder;

The semi-plane constraint set of all path points on the planned path generated by the A* algorithm constitutes a convex domain set, and the path planning configuration space is constructed.

Preferably, in response to the obstacle acquisition process, the cylindrical threat and the prismatic threat are used as obstacles, and the height is set to be infinitely high. The obstacle is expressed as:

Among them, ||·|| ₂ is the 2-norm, and /> Indicates a cylindrical threat and a diamond-shaped threat, p _xy indicates the current position of the drone, /> and /> Indicates the position and radius of the center point and radius of the horizontal plane of the mth cylinder, /> and b _{m, i} represent the half-space coefficient of the m-th rhombus threatening the i-th side, /> Indicates the number of sides of the mth rhombus threat.

Preferably, in response to the construction process of the convex domain set, a convex polygonal area is constructed by combining the cube trust region constraint and the half-plane constraint as a constraint area of a trajectory point, and the convex domain set is constructed according to the constraint area, wherein the constraint area Expressed as:

Among them, B(p _i ) represents a fixed-size cube trust region constraint; Denotes waypoint p _i with cylindrical threat /> half-space; /> represents the half-space of waypoint p _i and prismatic threat p _m ;

The set of convex fields is expressed as: Γ＝{SFC ₁ ∪SFC ₂ ∪...∪SFC _N }.

Preferably, in response to the acquisition process of the safe flight trajectory, the trapezoidal integration method is used to discretize the kinematics model of the UAV, and the right-hand term of the kinematics model of the discretized UAV is linearly processed according to the reference state trajectory. The linear constraints of position and velocity are obtained respectively, and the safe flight trajectory is obtained through the trajectory sequence convex optimization algorithm based on SFC-SCP, where the linear constraints are expressed as:

Among them, p=[p _x , p _y , p _z ] is the three-dimensional space position, v=[v _x , v _y , v _z ] is the velocity, a=[a _x , a _y , a _z ] is the acceleration, Δt is the time interval from the kth point to the k+1th point.

Preferably, in response to constructing a trajectory sequence convex optimization algorithm based on SFC-SCP, the front-end construction of SFC alleviates the problem of initial value sensitivity and result non-convergence when solving the trajectory of SCP, and generates a trajectory sequence convex optimization algorithm based on SFC-SCP.

Preferably, in response to the process of obtaining the safe flight trajectory through the trajectory sequence convex optimization algorithm based on SFC-SCP, it is judged whether the acquisition process satisfies the convergence condition, and if so, the current trajectory planning result is output; if not satisfied, the reference trajectory is updated, Continue to iteratively solve, where the convergence condition: all constraints are satisfied, and the trajectory results of two consecutive iterations are the same within a certain error range, expressed as:

Among them, q represents the qth iteration in sequential convex optimization, and X ^q represents the solution result of the qth iteration.

The invention discloses a sequential convex optimization system for the time-optimized trajectory of an unmanned aerial vehicle, comprising:

The data collection module is used to obtain the obstacle information of the UAV in the actual environment;

The data processing module is used to construct the real environment grid map according to the obstacle information, perform path planning through the A* algorithm, and generate the path planning configuration space of the UAV;

The flight trajectory planning module is used to adopt the trajectory sequence convex optimization algorithm based on SFC-SCP, and use the flight time between trajectory points as the optimization variable to search in the path planning configuration space to obtain a safe flight trajectory with optimal time.

The invention discloses the following technical effects:

The invention can generate a feasible path for the UAV that meets the requirements by using the strong robustness and convergence of the safe flight corridor according to the mission requirements of the UAV to reach a given target point, and has the advantages of short flight time, strong obstacle avoidance ability, The advantage of the algorithm is that it takes a short time.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

Fig. 1 is the flowchart of the UAV time optimal trajectory optimization method provided by the embodiment of the present invention;

Fig. 2 is a schematic diagram of a half-plane of a cylindrical threat provided by an embodiment of the present invention;

FIG. 3 is a schematic diagram of a half-plane of a prismatic threat provided by an embodiment of the present invention;

4 is a schematic diagram of a multi-plane safe flight corridor of a single track point provided by an embodiment of the present invention;

FIG. 5 is a schematic diagram of a multi-plane safe flight corridor of a track point set provided by an embodiment of the present invention;

FIG. 6 is a schematic diagram of corridor constraints of 1=7 discrete points provided by an embodiment of the present invention;

Fig. 7 is a two-dimensional diagram of comparison of scene-algorithm planning results provided by the embodiment of the present invention;

Fig. 8 is a three-dimensional diagram of comparison of scene-algorithm planning results provided by the embodiment of the present invention;

Fig. 9 is a two-dimensional diagram of comparison of planning results of scene two provided by the embodiment of the present invention;

Fig. 10 is a three-dimensional diagram of comparison of planning results of scene two provided by the embodiment of the present invention;

Fig. 11 is the trajectory time diagram of 100 random scene algorithms provided by the embodiment of the present invention;

Fig. 12 is a time-consuming diagram for solving 100 random scene algorithms provided by the embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only It is a part of the embodiments of this application, not all of them. The components of the embodiments of the application generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present application.

As shown in Figure 1-12, the present invention provides a time-optimal trajectory sequence convex optimization method for UAVs constrained by a safe flight corridor. The time-optimal trajectory sequence convex optimization method includes:

Obtain the UAV's initial state and terminal state, state and control constraint boundaries, threat location information, path discrete points, safe flight corridor boundary constraints, initial trust region radius, and convergence threshold;

Obtain obstacle information in the actual environment, construct a real environment grid map, and evenly divide the map into grids of equal proportions;

The UAV parameter information includes the size of the UAV, the path constraint information includes the starting position and the target point location of the UAV, and the mission environment information includes the map size, the number of obstacles, and the location of obstacles.

Convex optimization of UAV time-optimal trajectory sequence construction based on the real environment grid map specifically includes:

In the real environment grid map, a shortest safe path from the starting point to the ending point is obtained through path planning algorithms such as A*;

Convex polyhedron constraints for constructing trajectory points for threat areas and waypoints in the real environment raster map;

The UAV trajectory optimization model constrained by the safe flight corridor is constructed, and the flight time between trajectory points is used as the optimization variable, and the sequential convex optimization method is used to solve the problem.

Optionally, a path planning algorithm based on the A* algorithm is used to search in the convex optimization configuration space of the time-optimal trajectory sequence of the UAV to obtain a feasible flight path of the UAV.

Threats are divided into cylindrical threats and prismatic threats, and the half-plane sets are obtained for the two threats respectively, and the planning convex domain of each path point is obtained. The cylindrical threat half-plane can be expressed as:

l(i,m)=-H(i,m)×J(i,m), i=1,2,...,N; m=1,2,...,M (3)

Among them, H(i,m) is the semi-plane unit normal vector generated by the i-th path point and the m-th cylindrical obstacle, and J(i,m) is the nearest path point on the m-th circular obstacle to the i-th path point A point coordinate, l(i,m) is the distance constraint between the i-th path point and the m-th cylindrical obstacle, using formula (4) to describe the half-plane space of the cylindrical obstacle:

Η(i,m)×p _i ≥ J(i,m) (4)

The prismatic threat half-plane first calculates the horizontal center point of the prismatic threat. According to the formation of multiple rays from this point and each vertex, it is judged which two rays the path point is located in. As shown in formula (8), the two vertices are used The edge of is used as a half-plane, as shown in formula (5).

in, b _{m and j} are the half-plane space coefficients formed by the jth vertex and the central point of the mth prismatic threat, respectively.

Among them, a_p _m and b_p _m are the space coefficients of the half-plane on the side where the path point is located, respectively, using represents the half-space of waypoint p _i and prismatic threat p _m .

In order to increase the calculation efficiency of the flight corridor, when obtaining the half-plane, only the half-plane of obstacles within a certain range of the current waypoint is taken, that is, a fixed-size cube trust region constraint is added to each waypoint, which can be expressed as formula (7), η is half the side length of the square.

Combining cube trust region constraints and half-plane constraints can construct a convex polygonal area, which is the constraint area of a trajectory point in the optimization algorithm, which can be expressed as:

Among them, B(p _i ) represents a fixed-size cube trust region constraint; Denotes waypoint p _i with cylindrical threat /> half-space; /> represents the half-space of waypoint p _i and prismatic threat p _m .

According to the above method, a path point p _i can generate a convex field SFC _i , and a set of convex fields can be obtained by traversing the path points planned by A*, which can be expressed as formula (9):

Γ＝{SFC ₁ ∪SFC ₂ ∪...∪SFC _N } (9)

Based on the obstacle avoidance constraints of the safety corridor, after the UAV motion equations are discretized and linearized, such as formula (10) and formula (11), the flight time between trajectory points is used as the optimization variable, and the sequential convex optimization method is used to solve successively. Obtain a safe flight trajectory with optimal time.

Among them, p＝[p _x , p _y , p _z ] is the three-dimensional space position, v＝[v _x , v _y , v _z ] is the velocity, a＝[a _x , a _y , a _z ] is the acceleration, Δt is the time interval from the kth point to the k+1th point.

After discretization and linearization, the UAV time-optimal trajectory planning is modeled as the following optimization problem:

The present invention provides a time-optimal trajectory sequence convex optimization method for UAVs based on the constraints of safe flight corridors. The path planning method includes: obtaining UAV parameter information, path constraint information, task environment information and algorithm parameter information; obtaining actual Obstacle information in the environment, and build a real environment grid map, and evenly divide the map into grids of equal proportions; obtain UAV trajectory optimization configuration space according to the real environment grid map; use A*-based path planning algorithm Search in the UAV trajectory optimization configuration space to obtain a feasible UAV flight path; based on the SFC-SCP algorithm, a time-optimal safe flight trajectory is obtained.

Embodiment 1: The specific implementation steps of a time-optimal trajectory sequence convex optimization method for UAVs based on the constraints of the safe flight corridor provided by this embodiment are as follows:

Step 1: Input the problem information, including the initial state and terminal state of the UAV, state and control constraint boundary, threat location information, path discrete points, safe flight corridor boundary constraints, initial trust region radius, and convergence threshold.

Step 2: According to the parameter input of the above specific example, establish an environment map for the UAV time optimal trajectory optimization problem. The scale of the environment map is 100*100*20, the resolution is 0.1m, and it contains 50 obstacles, including 30 cylindrical obstacles (with a diameter between 3-6m), 20 polygonal obstacles, and the obstacle height is infinitely high. In order to reflect the planning efficiency of this algorithm in complex environments, the obstacles are randomly generated.

Step 3: Use the A* algorithm to obtain a shortest path from the start point to the end point, and obtain the initial reference trajectory according to the following formula.

In the formula, X ₀ represents the obstacle avoidance path using the A* algorithm as the reference trajectory, and the reference trajectory X ^q in subsequent iterations is the solution result of the previous iteration; q represents the qth iteration in sequential convex optimization, and X ^q represents the qth iteration The solution result of the subsequence iteration.

Step 4: Combining the surrounding obstacles, obtain the semi-plane constraint set corresponding to each path point and assign it to each trajectory point. The semi-plane constraint set is shown in formula (4); the planning path generated by the A* algorithm The set of half-plane constraints of all path points constitutes a set of convex domains, which can be expressed as formula (10).

Figure 2 is a schematic diagram of a half-plane of a cylindrical threat, using formula (5) to describe the half-plane space of a cylindrical obstacle;

Figure 3 is a schematic diagram of the half-plane of the prismatic threat. Firstly, the horizontal center point of the prismatic threat is obtained. According to the formation of multiple rays between this point and each vertex, it is judged which two rays the path point is located in, as shown in formula (6). , use the edge where the two vertices are located as the half-plane, as shown in formula (7);

Combining cube trust region constraints and half-plane constraints can construct a convex polygonal area, which is the constraint area of a trajectory point in the optimization algorithm, as shown in Figure 4, where the green area is the cube boundary constraint, and the red area is the constructed The safe flight corridor can be expressed as formula (9), and the red point is the way point;

A path point p _i can generate a convex field SFC _i , and a convex field set can be obtained by traversing the path points planned by A*, which can be expressed as formula (10), as shown in the red area in Figure 5.

Step 5: Construct a convex optimization model for the time-optimal trajectory optimization problem based on the benchmark trajectory.

Step 6: Solve using convex optimization algorithm to obtain the optimal control sequence of the current iteration.

According to the properties of convex sets, if any two points in a convex set are connected by a line, all points on the line are in the convex set, that is, two path points are in a convex set, and the planned results all meet the path constraint requirements. However, when the two path points are located in different convex sets, the two path points meet the constraint conditions, but the points on the line between the two points do not necessarily meet the requirements of the convex set, as shown by the black straight line in Figure 6. Therefore, l difference points are used to describe the trajectory between two path points, as shown in the green point in the figure, and the green diamond point is constrained at the intersection of two convex sets to ensure that the entire trajectory meets the obstacle avoidance constraint. The red triangle point in the figure is the path point, the blue filled polygon is the safe flight corridor, and the green point is the difference trajectory point. After adding the difference point, the red trajectory can satisfy the corridor constraint.

Step 6: Determine whether the convergence condition is satisfied, if so, output the current trajectory planning result; if not, update the reference trajectory, and return to step 4. Convergence condition: All constraints are satisfied, and the trajectory results of two consecutive iterations are the same within a certain error range, which can be expressed as:

Among them, q represents the qth iteration in sequential convex optimization, and X ^q represents the solution result of the qth iteration.

For the specific examples above, the path results obtained by using the UAV time-optimal trajectory sequence convex optimization method based on the safe flight corridor constraints described in this embodiment are compared with other algorithms as shown in Figures 7-10.

The path results obtained by using the A* algorithm or the SCP algorithm are shown in the paths in Figures 7-10. Although the path result obtained by the traditional method can avoid obstacles and guide the robot to reach the target position, the planned path needs to go around or stick to obstacles to avoid obstacles, and the path planning method proposed in this patent can be more accurate through trajectory sequence convex optimization. Achieve obstacle avoidance well, and be able to reach the target point with optimal time.

As shown in Figure 11 and Figure 12, the trajectory time and solution time of 100 random scene algorithms can be seen. The comparison results of the three algorithms show that the construction of SFC through the front-end alleviates the sensitivity to the initial value and the non-convergence of the result when SCP solves the trajectory. On the basis of losing 10.48% of flight time, the success rate is increased by 10%, and the algorithm solution time is reduced by 56%, which shows that the proposed SFC-SCP method has better robustness and computational efficiency.

According to the above simulation results and analysis of UAV time-optimal trajectory optimization, it can be seen that a UAV time-optimal trajectory sequence convex optimization method based on safe flight corridor constraints described in this embodiment can provide UAVs that meet the actual A feasible path with complex constraints, and compared with the traditional method, the path result has the advantages of optimal time and strong obstacle avoidance ability. Therefore, the present invention has engineering practicability and can achieve the expected purpose of the invention.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

In the description of the present invention, it should be understood that the terms "first" and "second" are used for description purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Thus, a feature defined as "first" and "second" may explicitly or implicitly include one or more of these features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims (10)

1. a sequential convex optimization method for unmanned aerial vehicle time optimal trajectory, is characterized in that, comprises the following steps: Obtaining the obstacle information of the UAV in the actual environment, constructing a real environment grid map, performing path planning through the A* algorithm, and generating the path planning configuration space of the UAV; A trajectory sequence convex optimization algorithm based on SFC-SCP is used, and the flight time between trajectory points is used as an optimization variable to search in the path planning configuration space to obtain a time-optimal safe flight trajectory. 2. according to claim 1, a kind of sequential convex optimization method for unmanned aerial vehicle time optimal track, it is characterized in that: In response to the construction process of the real environment grid map, by obtaining the initial state and terminal state of the UAV, state and control constraint boundaries, threat location information, path discrete points, safe flight corridor boundary constraints, initial trust region radius, and convergence threshold, Establish an environment map for the UAV time optimal trajectory optimization problem, and randomly generate obstacles in the environment map; Based on the environment map and according to the obstacle information, the real environment grid map is constructed, and the map is evenly divided into grids of equal proportions. 3. according to claim 2, a kind of sequential convex optimization method for unmanned aerial vehicle time optimal track, it is characterized in that: In response to the process of path planning through the A* algorithm, based on the environment map, a shortest path from the starting point to the destination is obtained through the A* algorithm, and an initial reference trajectory of the UAV is generated. Wherein, the initial reference trajectory is expressed as: Among them, X ₀ represents the obstacle avoidance path using the A* algorithm as the reference trajectory, and the reference trajectory X ^q in subsequent iterations is the solution result of the previous iteration; q represents the qth iteration in sequential convex optimization, and X ^q represents the qth The solution result of sequence iteration; Constructing the path planning configuration space according to the initial reference trajectory and according to the obstacle information. 4. according to claim 3, a kind of sequential convex optimization method for UAV time optimal trajectory, is characterized in that: In response to the construction process of the path planning configuration space, based on the initial reference trajectory, combined with surrounding obstacles, obtain a half-plane constraint set corresponding to each path point, and assign it to each trajectory point, wherein the half-plane A set of constraints is expressed as: l(i,m)=-H(i,m)×J(i,m), i=1,2,...,N; m=1,2,...,M In the formula, H(i,m) is the semi-plane unit normal vector generated by the i-th path point and the m-th cylindrical obstacle, and J(i,m) is the nearest path point on the m-th circular obstacle to the i-th path point A point coordinate of and /> Indicates the position and radius of the center point and radius of the horizontal plane of the mth cylinder; The semi-plane constraint set of all path points on the planned path generated by the A* algorithm forms a convex domain set, and the path planning configuration space is constructed. 5. according to claim 4, a kind of sequential convex optimization method for unmanned aerial vehicle time optimal track, it is characterized in that: In response to the obstacle acquisition process, the cylindrical threat and the prismatic threat are used as the obstacle, and the height is set to be infinitely high, and the obstacle is expressed as: Among them, ||·|| ₂ is the 2-norm, and /> Indicates a cylindrical threat and a diamond-shaped threat, p _xy indicates the current position of the drone, /> and /> Indicates the position and radius of the center point and radius of the horizontal plane of the mth cylinder, /> and b _m,i represent the half-space coefficient of the m-th rhombus threatening the i-th side, /> Indicates the number of sides of the mth rhombus threat. 6. according to claim 5, a kind of sequential convex optimization method for unmanned aerial vehicle time optimal track, it is characterized in that: In response to the construction process of the convex domain set, a convex polygonal area is constructed by combining the cube trust region constraint and the half-plane constraint, as a constraint area of a trajectory point, and the convex domain set is constructed according to the constraint area, wherein the The above constraint area is expressed as: B(p _i ) represents a fixed-size cube trust region constraint; Denotes waypoint p _i with cylindrical threat /> half-space; /> represents the half-space of waypoint p _i and prismatic threat p _m ; The set of convex domains is expressed as: r={SFC ₁ ∪SFC ₂ ∪...∪SFC _N }. 7. according to claim 6, a kind of sequential convex optimization method for UAV time optimal trajectory, is characterized in that: In response to the acquisition process of the safe flight trajectory, using the trapezoidal integration method, the kinematics model of the UAV is discretized, and the right-hand term of the kinematics model of the UAV after discretization is linearized according to the reference state trajectory , respectively obtain the linear constraints of position and velocity, and obtain the safe flight trajectory through the trajectory sequence convex optimization algorithm based on SFC-SCP, wherein the linear constraints are expressed as: Among them, p＝[p _x , p _y , p _z ] is the three-dimensional space position, v＝[v _x , v _y , v _z ] is the velocity, a＝[a _x , a _y , a _z ] is the acceleration, Δt is the time interval from the kth point to the k+1th point. 8. according to claim 7, a kind of sequential convex optimization method for unmanned aerial vehicle time optimal track, it is characterized in that: In response to the construction of a trajectory sequence convex optimization algorithm based on SFC-SCP, the front-end construction of SFC alleviates the problem of initial value sensitivity and result non-convergence when solving the trajectory of SCP, and generates the trajectory sequence convex optimization algorithm based on SFC-SCP. 9. according to claim 8 a kind of sequential convex optimization method for unmanned aerial vehicle time optimal track, it is characterized in that: In response to the process of obtaining the safe flight trajectory through the trajectory sequence convex optimization algorithm based on SFC-SCP, judge whether the acquisition process satisfies the convergence condition, if so, output the current trajectory planning result; if not, update the reference trajectory, and continue to iteratively solve , where the convergence condition: all constraints are satisfied, and the trajectory results of two consecutive iterations are the same within a certain error range, expressed as: Among them, q represents the qth iteration in sequential convex optimization, and X ^q represents the solution result of the qth iteration. 10. A sequential convex optimization system for the optimal trajectory of unmanned aerial vehicle, it is characterized in that, comprising: The data collection module is used to obtain the obstacle information of the UAV in the actual environment; The data processing module is used to construct a real environment grid map according to the obstacle information, and perform path planning through the A* algorithm to generate the path planning configuration space of the UAV; The flight trajectory planning module is used to adopt the trajectory sequence convex optimization algorithm based on SFC-SCP, and use the flight time between trajectory points as an optimization variable to search in the path planning configuration space to obtain a time-optimal safe flight trajectory.