An obstacle avoidance correction method for a UAV control system based on a control barrier function

Abstract

The application discloses an obstacle avoidance correction method of a UAV control system based on a control barrier function, and belongs to the field of obstacle avoidance correction of a UAV control system. The method is implemented as follows: a UAV motion model on a two-dimensional horizontal plane is constructed; constraint conditions include a no-fly area constraint and a saturation constraint of control input; a UAV obstacle avoidance correction problem model is constructed; the UAV obstacle avoidance correction problem model is converted into a standard quadratic programming problem of UAV obstacle avoidance correction, repeated iterative optimization solving is avoided to ensure real-time performance, a corrected control amount is obtained by solving the quadratic programming problem, a basic control amount output by a basic control algorithm is corrected in real time through the corrected control amount, a gap between the corrected control amount and the basic control amount is reduced on the basis of realizing UAV no-fly area avoidance, that is, the obstacle avoidance correction based on the control barrier function, UAV no-fly area avoidance is realized, and flight safety of the UAV is ensured. The application has the advantages of high real-time performance and good safety.

Description

Technical Field

[0001] This invention belongs to the field of obstacle avoidance correction in unmanned aerial vehicle (UAV) control systems, and relates to an obstacle avoidance correction method for UAV control systems based on a control obstacle function. Background Technology

[0002] Unmanned aerial vehicle (UAV) technology has become a focus of international strategic research and development in recent years, with countries actively investing resources in exploration and innovation. UAVs have attracted considerable attention due to their numerous advantages, including reusability, ease of operation, long-duration flight time, small size and portability, and low cost. This makes UAVs widely applicable in military fields such as target reconnaissance, electronic warfare, equipment delivery, and ground support, as well as in civilian fields such as disaster relief, emergency communications, atmospheric environmental monitoring, and land and mineral exploration. Especially in dangerous, remote, or harsh environments, UAVs are gradually replacing manned aircraft systems. As the application scope of UAVs expands and deepens, the complexity of missions and the challenges of flight environments continue to increase, further driving the continuous improvement of UAV performance and technical specifications.

[0003] Precise control of unmanned aerial vehicles (UAVs) during flight is a crucial technology, ensuring the successful execution of flight missions and significantly enhancing their combat effectiveness. Controlling a UAV involves not only achieving the control objective of reaching the target point but also considering external environmental constraints, such as no-fly zones, as well as the UAV's own performance constraints, such as acceleration or velocity limitations. The method provided in this invention employs corrective control to perform real-time corrections to the basic control, enabling the UAV to avoid no-fly zones within the capabilities of its actuators. Furthermore, the generated corrective control inputs exhibit minimal deviation from the basic control inputs, preserving the characteristics of the basic controller as much as possible to accomplish obstacle avoidance tasks.

[0004] Domestic and international researchers have largely employed trajectory planning methods for obstacle avoidance in unmanned aerial vehicles (UAVs), achieving some success. Algorithms such as genetic algorithms, A* algorithms, fast random search tree algorithms, and convex optimization-based algorithms have seen practical applications. However, while most trajectory planning methods can ensure UAVs do not violate no-fly zone constraints, they are computationally intensive, demanding high computational capabilities from the UAV, making real-time trajectory planning difficult and incompatible with existing methods, thus limiting their application scope. In addition, some researchers have used algorithms based on artificial potential fields or navigation functions to solve UAV obstacle avoidance problems, but these algorithms largely fail to consider the UAV's own performance constraints, further limiting their practical application. Summary of the Invention

[0005] To address the shortcomings of existing algorithms that cannot simultaneously satisfy UAV obstacle avoidance, UAV performance constraints, and ensure real-time performance, this invention aims to provide an obstacle avoidance correction method for UAV control systems based on a control obstacle function. This method constructs a UAV motion model on a two-dimensional horizontal plane, considering constraints including no-fly zone constraints and control input saturation constraints. Based on the UAV motion model and constraints, a UAV obstacle avoidance correction problem model is constructed. The obstacle avoidance execution capability of the control surfaces is analyzed, and the no-fly zone constraints are transformed, converting the UAV obstacle avoidance correction problem model into a standard quadratic programming problem for UAV obstacle avoidance correction. This avoids iterative optimization and ensures real-time performance. By solving this quadratic programming problem, the corrected control quantity is obtained. This corrected control quantity is then used to correct the basic control quantity output by the basic control algorithm in real time. This reduces the gap between the corrected control quantity and the basic control quantity while achieving no-fly zone avoidance for the UAV, thus achieving no-fly zone avoidance and ensuring UAV flight safety. The control quantity is the control surface input that changes the UAV's heading. This invention offers advantages of high real-time performance and good safety.

[0006] The objective of this invention is achieved through the following technical solution.

[0007] This invention discloses an obstacle avoidance correction method for an unmanned aerial vehicle (UAV) control system based on a control obstacle function, comprising the following steps:

[0008] S1: Construct a motion model of the drone on a two-dimensional horizontal plane;

[0009] The UAV motion model established in step S1 is as follows:

[0010] (1)

[0011] in, This represents the horizontal coordinate of the UAV on a two-dimensional plane. Represents the ordinate of the drone on the two-dimensional horizontal plane. For the drone's flight speed, The heading angle is the angle between the drone's flight direction and the horizontal coordinate axis. For control inputs to change the course of the UAV.

[0012] S2: Based on the UAV motion model obtained in step S1, considering the saturation constraints of the UAV control surface input and the no-fly zone constraints during the UAV's flight, construct a UAV obstacle avoidance correction problem model.

[0013] Considering the physical limitations of UAVs during flight, their control surfaces have maximum operating angles in both positive and negative directions. Therefore, their corresponding control inputs are subject to saturation constraints, specifically expressed as follows:

[0014] (2)

[0015] in, and This represents the control input value corresponding to the maximum operating angle of the control surface in both positive and negative directions.

[0016] During the flight of a drone, it is affected by weather conditions and airspace control factors, and there are airspaces that it is prohibited from entering. These are defined as no-fly zones. The no-fly zone is described as the interior of the smallest circumcircle containing the no-entry airspace. The area described by the circumcircle is specifically represented as shown in equation (3):

[0017] (3)

[0018] in, and Let x and y represent the x and y coordinates of the center of the no-fly zone, respectively. Let be the radius of the no-fly zone circle. The no-fly zone constraint, as shown in equation (4), keeps the UAV's trajectory outside the circular area:

[0019] (4)

[0020] Among them, variables This represents the squared difference between the distance from the drone's position to the center of the no-fly zone and the radius of the no-fly zone circle. When it is positive, it indicates that the drone's current position complies with the no-fly zone constraints.

[0021] Provide basic control Correction control input satisfy This ensures that the drone's position remains constant during flight. The model for the UAV obstacle avoidance correction problem is constructed by combining formulas (1), (2), and (4).

[0022] S3: Analyze the obstacle avoidance performance of the control surfaces and transform the no-fly zone constraints. Transform the UAV obstacle avoidance correction problem model into a standard quadratic programming problem for UAV obstacle avoidance correction. Solve this quadratic programming problem to obtain the corrected control quantity.

[0023] Since the UAV motion model established in step S1 is a second-order system, the control input does not directly act on the UAV's horizontal and vertical coordinate variables. Therefore, the constraints in equation (4) need to be transformed and analyzed as follows:

[0024] Define the obstacle avoidance capability variables of the control surfaces. This variable is related to the maximum actuation capability of the control surfaces, and is expressed as the total actuation capability under limited control input throughout the entire process. The lower bound of the maximum value:

[0025] (5)

[0026] Define a new constraint variable Its specific definition is:

[0027] (6)

[0028] in, derivative Through calculation We obtain the results. This is analyzed using the two cases defined in equation (6). If the current... As long as the current moment This ensures that the subsequent trend will remain positive, meaning the constraint will not be violated. If the current... ,like and satisfy:

[0029] (7)

[0030] In this case, it is equivalent to the ability to control input execution. Before reducing to 0, make The sign changes, meaning the no-fly zone constraint will not be violated. In summary, when the constraint shown in equation (8) is satisfied, the constraint described in equation (4) can be guaranteed to hold.

[0031] (8)

[0032] Using the invariant set handling method in the design of control barrier functions, the following inequality constraints are designed:

[0033] (9)

[0034] in, It is a constant greater than zero. When inequality (9) is satisfied, the constraint described in equation (8) is satisfied.

[0035] Further simplifying the analysis of the constraints in equation (9), since in In the case of constraint and correction control input in equation (9) It is irrelevant, meaning that the effect of the correction control input at this time does not affect the constraints. Therefore, the corresponding information regarding the correction control input... The inequality constraints are transformed as follows:

[0036] (10)

[0037] To simplify the representation, variables are defined. and for:

[0038] (11)

[0039] Based on the transformation of the above formula (11), the quadratic programming problem for UAV obstacle avoidance correction is constructed as shown in formula (12):

[0040] (12)

[0041] Equation (12) is a standard and easily solvable quadratic programming problem. Its objective is to find a corrective control input that satisfies the constraints. and with basic control The deviation is minimized, that is, the corrected control quantity is obtained by solving the standard quadratic programming problem of UAV obstacle avoidance correction as shown in equation (12).

[0042] S4: The control quantity output by the basic control algorithm is corrected in real time by correcting the control quantity after correction. While achieving drone no-fly zone avoidance, the difference between the corrected control quantity and the basic control quantity is reduced. That is, obstacle avoidance correction based on the control obstacle function, so as to maintain the characteristics of the basic controller as much as possible to achieve drone no-fly zone avoidance and ensure the flight safety of drone.

[0043] Beneficial effects:

[0044] 1. This invention discloses an obstacle avoidance correction method for UAV control systems based on control obstacle functions. For the obstacle avoidance correction problem of UAV control systems considering no-fly zone constraints and UAV control surface saturation constraints, the method analyzes the obstacle avoidance execution capability of the control surfaces and transforms the no-fly zone constraints, converting the obstacle avoidance correction problem into a standard quadratic programming problem. This avoids repeated iterative optimization and improves the solution efficiency of the quadratic programming problem, ensuring the real-time performance of obstacle avoidance correction. By solving this quadratic programming problem, a correction control quantity can be obtained. Using this correction control quantity to correct the basic control quantity in real time can achieve UAV obstacle avoidance while reducing corrections, ensuring the safety of UAV flight.

[0045] 2. The present invention discloses a UAV constraint avoidance control correction method based on a control obstacle function. The no-fly zone avoidance method is based on the original basic control quantity. By analyzing the obstacle avoidance execution capability of the control surface and transforming the no-fly zone constraint, the obstacle avoidance correction problem is transformed into a standard quadratic programming problem. The correction control quantity is obtained by solving the problem, avoiding the need to design a complete set of control algorithms separately. Moreover, the present invention can be combined with other control algorithms such as sliding mode control algorithm and PID control algorithm, and has a wide range of applications. Attached Figure Description

[0046] Figure 1 This is a flowchart of an obstacle avoidance correction method for an unmanned aerial vehicle (UAV) control system based on a control obstacle function, according to an embodiment of the present invention.

[0047] Figure 2 This is a complete block diagram of the obstacle avoidance correction method for a UAV control system based on a control obstacle function, according to an embodiment of the present invention.

[0048] Figure 3 This is a trajectory change diagram of the basic control algorithm used in an embodiment of the present invention;

[0049] Figure 4 This is a trajectory change diagram of the correction control algorithm used in an embodiment of the present invention;

[0050] Figure 5 This is a graph showing the changes in the basic control input and correction control input in an embodiment of the present invention. Detailed Implementation

[0051] To better illustrate the purpose and advantages of the present invention, the invention will be further described below in conjunction with the accompanying drawings and examples.

[0052] like Figure 1 As shown in this embodiment, the obstacle avoidance correction method for a UAV control system with a control obstacle function is implemented in the following specific steps:

[0053] Step 1: Construct a motion model of the drone on a two-dimensional horizontal plane;

[0054] (1)

[0055] in, This represents the horizontal coordinate of the UAV on a two-dimensional plane. Represents the ordinate of the drone on the two-dimensional horizontal plane. For the flight speed of the drone, The heading angle is the angle between the drone's flight direction and the horizontal coordinate axis. The control input for changing the UAV's heading. In this embodiment, the flight starting point of the UAV flight mission is defined as... The destination is The drone's flight speed is set to The initial heading angle is set to .

[0056] Step 2: Based on the UAV motion model obtained in Step 1, consider the saturation constraints of the UAV control surface input and the no-fly zone constraints during flight, and construct a UAV obstacle avoidance correction problem model.

[0057] Considering the physical limitations of UAVs during flight, their control surfaces have maximum operating angles in both positive and negative directions. Therefore, their corresponding control inputs are subject to saturation constraints, specifically expressed as follows:

[0058] (2)

[0059] in, and This represents the control input value corresponding to the maximum operating angle of the control surface in both positive and negative directions. In this embodiment, it is set... , .

[0060] During the flight of a drone, it is affected by weather conditions and airspace control factors, and there are airspaces that it is prohibited from entering. These are referred to as no-fly zones. The no-fly zone is described as the interior of the smallest circumcircle containing the no-entry airspace. The area described by the circumcircle is specifically represented as shown in equation (3):

[0061] (3)

[0062] in, and Let x and y represent the x and y coordinates of the center of the no-fly zone, respectively. Let be the radius of the no-fly zone circle. The no-fly zone constraint, as shown in equation (4), keeps the UAV's trajectory outside the circular area:

[0063] (4)

[0064] Among them, variables This represents the squared difference between the distance from the drone's position to the center of the no-fly zone and the radius of the no-fly zone circle. A positive value indicates that the drone's current position complies with the no-fly zone constraints. In this embodiment, the drone needs to avoid two no-fly zones. The relevant data for the no-fly zones are as follows: the center of the first no-fly zone is... The radius of the no-fly zone is The center of the second no-fly zone is The radius of the no-fly zone is .

[0065] Provide basic control Correction control input satisfy This ensures that the drone's position remains constant during flight. In this embodiment, the basic control uses standard proportional control. The basic control quantity consists of the deviation between the actual heading angle and the target heading angle. The actual heading angle is the angle between the direction of the UAV's flight speed and the direction of the lateral coordinate axis. The target heading angle is the angle between the line connecting the start and end points of the flight mission and the direction of the lateral coordinate axis. In other words, the control objective is to make the UAV fly as close to the destination as possible. In this embodiment, the target heading angle is... The proportional control coefficient is selected as The obstacle avoidance correction problem of UAVs can be described as a problem model consisting of formulas (1), (2), and (4).

[0066] Step 3: Analyze the obstacle avoidance performance of the control surfaces and transform the no-fly zone constraints. Transform the UAV obstacle avoidance correction problem model into a standard quadratic programming problem for UAV obstacle avoidance correction. Solve the quadratic programming problem to obtain the corrected control quantity.

[0067] Since the mathematical model of the UAV motion established in step 1 is a second-order system, the control input does not directly act on the UAV's horizontal and vertical coordinate variables. Therefore, the constraints in equation (4) need to be transformed and analyzed as follows:

[0068] First, define the obstacle avoidance capability variable of the control surface. This variable is related to the maximum actuation capability of the control surfaces, and is expressed as the total actuation capability under limited control input throughout the entire process. The lower bound of the maximum value:

[0069] (5)

[0070] In practical systems, when the system's physical characteristics and control capabilities are determined, It is a fixed value, but in practice, we often find it difficult to obtain this fixed value. Therefore, we can choose a value smaller than this fixed value based on experience or known trajectories, i.e., its lower limit. When selecting this variable, if there is a lack of established trajectories or other relevant data, an trial-and-error method can be used. When the selected value is too small, the trajectory will be significantly farther from the set no-fly zone; when the selected value is too large, the trajectory will be unable to avoid the no-fly zone. Therefore, careful selection is required. In this embodiment, .

[0071] Next, we define a new constraint variable. Its specific definition is:

[0072] (6)

[0073] in, derivative Through calculation We obtain the results. This is analyzed using the two cases defined in equation (6). If the current... As long as the current moment This means that if the drone does not violate the no-fly zone restrictions, its subsequent trend can be guaranteed to remain positive, indicating that the restrictions will not be violated in the next moment. If the current... ,like and satisfy:

[0074] (7)

[0075] In this case, the ability to control input execution is equivalent to... Before reducing to 0, make The sign changes, meaning the no-fly zone constraint will not be violated. In summary, when the constraint shown in equation (8) is satisfied, the constraint described in equation (4) can be guaranteed to hold.

[0076] (8)

[0077] Using the invariant set handling method in the design of control barrier functions, the following inequality constraints are designed:

[0078] (9)

[0079] in, It is a constant greater than zero, which can be arbitrarily chosen in practice. In this embodiment, it is selected as... Once inequality (9) is satisfied, the constraint described in equation (8) is satisfied.

[0080] Further simplifying the analysis of the constraints in equation (9), since in In the case of constraint and correction control input in equation (9) It is irrelevant, meaning the effect of the correction control input at this time does not affect the constraints. In this case, the basic control does not need correction; that is, the correction control is the same as the basic control. Therefore, the corresponding information regarding the correction control input... The inequality constraints are transformed as follows:

[0081] (10)

[0082] To simplify the representation, variables are defined. and for:

[0083] (11)

[0084] Based on the above transformation, the quadratic programming problem for UAV obstacle avoidance correction is constructed as shown in equation (12):

[0085] (12)

[0086] Equation (12) is a standard and easily solvable quadratic programming problem. Its objective is to find a corrective control input that satisfies the constraints. and with basic control The deviation is minimized, that is, the corrected control quantity is obtained by solving the quadratic programming problem of UAV obstacle avoidance correction as shown in equation (12).

[0087] Step 4: The control quantity output by the basic control algorithm is corrected in real time using the corrected control quantity. While achieving no-fly zone avoidance for the UAV, the difference between the corrected control quantity and the basic control quantity is reduced. That is, obstacle avoidance correction based on the obstacle function is performed to maintain the characteristics of the basic controller as much as possible to achieve no-fly zone avoidance for the UAV and ensure the flight safety of the UAV.

[0088] Figure 2 This diagram illustrates the complete control system block diagram of the obstacle avoidance correction method for the UAV control system based on the obstacle function in this embodiment. The flight trajectory obtained from the basic control is as follows: Figure 3 As shown, the drone violated the no-fly zone restrictions, and the flight trajectory obtained by corrective control is as follows. Figure 4 As shown, the drone avoided the no-fly zone while adhering as closely as possible to the basic control settings, demonstrating that the algorithm can correct the drone's behavior to avoid the no-fly zone. The changes in the basic control input and the corrective control input are as follows: Figure 5 As shown, the control input saturation constraint, i.e., formula (2), is satisfied. Note that this method cannot solve the case where the obstacle is very close to the endpoint. If the obstacle is very close to the endpoint, it may lead to the inability to reach the endpoint or to completely avoid the obstacle. This is limited by the obstacle avoidance execution capability of the control surface. The more accurate the value, the better the correction control effect.

[0089] The above detailed description further illustrates the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An obstacle avoidance correction method for an unmanned aerial vehicle (UAV) control system based on an obstacle control function, characterized in that: Includes the following steps, S1: Construct a motion model of the drone on a two-dimensional horizontal plane; S2: Based on the UAV motion model obtained in step S1, considering the saturation constraints of the UAV control surface input and the no-fly zone constraints during the UAV's flight, construct a UAV obstacle avoidance correction problem model. S3: Analyze the obstacle avoidance performance of the control surfaces and transform the no-fly zone constraints. Transform the UAV obstacle avoidance correction problem model into a standard quadratic programming problem for UAV obstacle avoidance correction. Solve the quadratic programming problem to obtain the corrected control quantity. The implementation method for step S3 is as follows: Since the UAV motion model established in step S1 is a second-order system, the control input does not directly act on the UAV's horizontal and vertical coordinate variables. Therefore, the constraints in equation (4) need to be transformed and analyzed as follows: Define the obstacle avoidance capability variables of the control surfaces. This variable is related to the maximum actuation capability of the control surfaces, and is expressed as the total actuation capability under limited control input throughout the entire process. The lower bound of the maximum value: （5） Define a new constraint variable Its specific definition is: （6） in, derivative Through calculation get, This represents the horizontal coordinate of the UAV on a two-dimensional plane. Represents the ordinate of the drone on the two-dimensional horizontal plane. and Let x and y represent the x and y coordinates of the center of the no-fly zone, respectively; analyze the two cases defined by equation (6); if the current As long as the current moment The subsequent trend will ensure that it remains positive, meaning the constraint will not be violated; if the current ,like and satisfy: （7） variable This represents the squared difference between the distance from the drone's position to the center of the no-fly zone and the radius of the no-fly zone circle. A positive value indicates that the drone's current position complies with the no-fly zone constraints; in this case, it is equivalent to the control input's execution capability being... Before reducing to 0, make The symbol is changed, meaning that the no-fly zone constraint will not be violated; in summary, when the constraint shown in equation (8) is satisfied, the constraint described in equation (4) can be guaranteed to hold. （8） Using the invariant set handling method in the design of control barrier functions, the following inequality constraints are designed: （9） in, It is a constant greater than zero; when inequality (9) is satisfied, the constraint described in equation (8) is satisfied; Further simplifying the analysis of the constraints in equation (9), since in In the case of constraint and correction control input in equation (9) It is irrelevant, meaning that the effect of the correction control input at this time does not affect the constraints. Therefore, the corresponding information regarding the correction control input... The inequality constraints are transformed as follows: （10） Among them, defining variables and for: （11） in, Let the angle between the UAV's flight direction and the horizontal coordinate axis be the heading angle. Based on the above formula (11), the quadratic programming problem for UAV obstacle avoidance correction is constructed as shown in formula (12): （12） in, and Let represent the control input value corresponding to the maximum operating angle of the control surface in both positive and negative directions; Equation (12) is a standard and easily solvable quadratic programming problem, the goal of which is to find a corrective control input that satisfies the constraints. and with basic control The deviation is minimized, that is, the corrected control quantity is obtained by solving the standard quadratic programming problem of UAV obstacle avoidance correction as shown in equation (12); S4: The control quantity output by the basic control algorithm is corrected in real time by the corrected control quantity. While achieving drone no-fly zone avoidance, the difference between the corrected control quantity and the basic control quantity is reduced. That is, obstacle avoidance correction based on the control obstacle function, so as to maintain the characteristics of the basic controller as much as possible to achieve drone no-fly zone avoidance.

2. The obstacle avoidance correction method for an unmanned aerial vehicle (UAV) control system based on a control obstacle function as described in claim 1, characterized in that: The method for implementing step S1 is as follows: The UAV motion model established in step S1 is as follows: （1） in, This represents the horizontal coordinate of the UAV on a two-dimensional plane. Represents the ordinate of the drone on the two-dimensional horizontal plane. For the drone's flight speed, The heading angle is the angle between the drone's flight direction and the horizontal coordinate axis. For control inputs to change the course of the UAV.

3. The obstacle avoidance correction method for an unmanned aerial vehicle (UAV) control system based on a control obstacle function as described in claim 2, characterized in that: The implementation method for step S2 is as follows: Considering the physical limitations of UAVs during flight, their control surfaces have maximum operating angles in both positive and negative directions. Therefore, their corresponding control inputs are subject to saturation constraints, specifically expressed as follows: （2） in, and This represents the control input value corresponding to the maximum operating angle of the control surface in both positive and negative directions; During the flight of a drone, it is affected by weather conditions and airspace control factors, and there are airspaces that it is prohibited from entering. These are defined as no-fly zones. The no-fly zone is described as the interior of the smallest circumcircle containing the no-entry airspace. The area described by the circumcircle is specifically represented as shown in equation (3): （3） in, and Let x and y represent the x and y coordinates of the center of the no-fly zone, respectively. Let be the radius of the no-fly zone circle; the trajectory of the UAV is kept outside the circular area by the no-fly zone constraint as shown in equation (4): （4） Among them, variables This represents the squared difference between the distance from the drone's position to the center of the no-fly zone and the radius of the no-fly zone circle. When it is positive, it indicates that the drone's current position complies with the no-fly zone constraints. Provide basic control Correction control input satisfy This ensures that the drone's position remains constant during flight. The model for the UAV obstacle avoidance correction problem is constructed by combining formulas (1), (2), and (4).

Images