A method for wind-following hovering control of tethered unmanned aerial vehicles based on admittance control

Abstract

This invention proposes a wind-following hovering control method for tethered unmanned aerial vehicles (UAVs) based on admittance control. The steps are as follows: Step 1: Establish a complete dynamic model of the tethered UAV and its reel; Step 2: Solve for the desired hovering position of the tethered UAV using an admittance model and design the controller for the tethered UAV; Step 3: Achieve hovering of the tethered UAV at the new desired hovering position through low-level control. In this invention, the attitude angle of the tethered UAV is used as the input to the admittance controller to represent the influence of external forces, and the required hovering position can be obtained through a preset admittance model. Theoretical simulation verification shows that the above control strategy can effectively prevent the multi-rotor from tilting too much when hovering in the wind. In this process, the tension of the tether rope is used to assist in counteracting the wind force, thereby preventing the tethered UAV from hovering too low in the wind.

Description

Technical Field

[0001] This invention proposes a wind-following hovering control method for tethered unmanned aerial vehicles (UAVs) based on admittance control. Upon encountering an external wind field, the tethered UAV using this method actively hovers with the wind, utilizing the tension of the tethering rope to assist in counteracting wind drag, thereby preventing excessive tilting of the tethered UAV due to wind drag. This method uses the attitude angle of the tethered UAV as input, obtains the desired hovering position through an admittance controller, and finally achieves hovering of the tethered UAV at the new desired hovering position through low-level control, belonging to the field of aircraft control. Background Technology

[0002] In recent years, drones have been used more and more widely in fields such as agriculture, logistics, transportation, and inspection. However, in some scenarios that require drones to hover for extended periods, such as emergency lighting and emergency communications, conventional multi-rotor drones are unable to meet the requirements due to their limited flight time. Therefore, tethered drones have come into use.

[0003] Tethered drones are connected to the ground via a tethering cable, and their structure is as follows: Figure 1 As shown, this tether can be used to supply power from the ground or to transmit information between the drone and the ground, allowing the tethered drone to fly for extended periods. This structural feature demonstrates that the tethered drone is an aerial robot that works in conjunction with a cable. Typically, tethered drones are required to hover at a fixed point; however, in strong winds, their attitude may become excessively tilted to counteract wind resistance. On one hand, the working angle of the payload on the tethered drone, such as the camera and communication antenna, is obstructed; on the other hand, a more tilted drone is more prone to crashing if subjected to external interference. However, if the requirement for fixed-point hovering is abandoned, allowing the drone to hover with the wind, the tension of the tether can help counteract wind resistance, thus preventing the multirotor from tilting excessively. How to adaptively allow multirotors to actively adjust their hovering position in the wind and autonomously return to their initial hovering point when the wind disappears is a problem worthy of research, and the development of flexible control for aerial robots has provided a solution to this problem.

[0004] This invention presents a position control strategy based on an admittance controller. The attitude angle of the tethered UAV is used as the input to the admittance controller to represent the influence of external forces, and the desired hovering position can be obtained through a preset admittance model. Theoretical simulations verify that the above control strategy can effectively prevent the multirotor from tilting excessively when hovering in wind. Summary of the Invention

[0005] This invention proposes a wind-following hovering control method for tethered unmanned aerial vehicles (UAVs) based on admittance control. First, a dynamic model of the tethered UAV is established; then, the desired hovering position of the tethered UAV is solved using an admittance model; finally, the tethered UAV is hovered at the new desired hovering position through low-level control. The overall steps are as follows: Figure 2 As shown, the detailed steps are as follows:

[0006] Step 1: Establish a complete dynamic model of the tethered drone and its reel.

[0007] Step 1.1: Establish the dynamic model of the tethered drone. The forces acting on the tethered drone in the air include gravity, aerodynamic forces, tether tension, and rotor thrust, etc. Therefore, the dynamic model can be obtained as follows:

[0008]

[0009] Where, p = [p x p y p z ] T and v = [v x v y v z ] T Represents the position and velocity of the tethered drone, respectively; m is the mass of the tethered drone; T r T t and F a Let g represent the rotor thrust, the three-dimensional tension of the tethered drone at the drone's end, and the wind force, respectively, in the geodetic coordinate system; g = [00g] T [ω] is the acceleration due to gravity; ω is the rotational angular velocity of the tethered drone; × It is the cross product matrix of vector ω; J is the moment of inertia; τ r τ g τ t and τ a These represent the torque generated by rotor thrust, the torque generated by gravity, the tethering rope tension, and the wind torque, respectively. The rotation matrix from the UAV's rotor coordinate system to the geodetic coordinate system can be represented as:

[0010]

[0011] Where φ, θ and ψ represent the roll angle, pitch angle and yaw angle of the UAV, respectively.

[0012] Step 1.2: Establish the dynamic model of the cable reel. The dynamic model of the cable reel can be expressed by important parameters such as the length of the tethering rope, the cable retrieval speed, and the rotational angular velocity of the cable reel as follows:

[0013]

[0014] Where L is the length of the mooring rope, v c J is the linear velocity of the tether rope as it is retrieved by the reel. w It is the moment of inertia of the winding reel, ω w τ is the rotational angular velocity of the cable reel during operation, r is the radius of the cable reel, and τ is the angular velocity of rotation. w It is the torsional torque of the cable reel, F c The tension of the tethering rope at the end of the reel.

[0015] Step Two: Design the controller for the tethered UAV. The controller consists of a top-level admittance controller and bottom-level position and attitude control. The overall control scheme is as follows: Figure 3 As shown, it is mainly divided into two parts: the control of the UAV and the control of the cable reel. For the UAV controller, the desired hovering position p of the tethered UAV under the wind field is first solved by using the UAV's attitude R as the input of the admittance controller. d Then the desired hovering position p d The actual position p is transmitted to the position controller to obtain the desired rotor thrust T. rd Desired UAV attitude R d The next step will be to determine the desired drone attitude R. d The desired three-dimensional torque τ provided by the rotor is obtained by using the actual attitude R as the input to the attitude controller. rd For the cable reel, the length of the tethering rope is used as the control input, and this value can be obtained in the simulation by the position p of the tethered drone. The control output, the control torque τ of the cable reel, can be obtained through the cable reel's controller. wd The control torque of the reel is then transmitted into the reel's dynamic model to obtain the tension T of the mooring rope. t Ultimately, the desired drone rotor thrust T will be achieved. rd The expected three-dimensional torque τ provided by the rotor rd And the tension T of the mooring rope t The values ​​are fed into the dynamic model of the tethered UAV to obtain the UAV's attitude R and position p, which are then fed back to the admittance controller, position controller, attitude controller, and reel controller mentioned above.

[0016] Step 2.1: Design the top-level admittance controller. Using a pre-defined admittance controller, the attitude angle of the tethered UAV is used as the control input to obtain the desired hovering position of the tethered UAV. In this step, only the horizontal position is considered to change with the wind through the admittance model, while the altitude remains constant. It can be noted that the direction of rotor thrust is closely related to the attitude angle and can intuitively reflect the magnitude of external wind drag. Therefore, we consider using the direction of rotor thrust, expressed in terms of attitude angle, as the input to the controller to obtain the desired hovering position. The controller is designed as follows:

[0017]

[0018] in, Let Γ be the desired horizontal hovering position of the tethered UAV; D is the damping coefficient, a positive definite matrix; K is the stiffness coefficient, a positive definite matrix; Γ is the controller input, which can be expressed as rotor thrust and attitude angle respectively:

[0019]

[0020] In the controller described above, the desired horizontal hovering position of the tethered UAV when it is stable is determined by matrices D and K. The smaller the parameter value in matrix D, the faster the adjustment speed of the desired hovering position; the smaller the parameter value in matrix K, the farther the desired hovering position.

[0021] Furthermore, combined with the expected flight altitude of the tethered drone The desired hovering position of the tethered drone can be obtained as follows:

[0022]

[0023] Step 2.2: Set the desired hovering position p d The data is passed to the underlying controllers, including the position controller and attitude controller, to achieve the low-level control of the tethered UAV. These low-level controllers include various types capable of low-level control and tracking the desired hovering position, including but not limited to proportional-integral-derivative (PID) controllers and model predictive control (MPC).

[0024] Step 3: Design of the cable reel controller. In this invention, the tension of the tethering rope is controlled to have an effect similar to that of a spring; therefore, the controller of the cable reel is designed as follows:

[0025]

[0026] The controller input is the angular velocity ω of the winding reel. w Cable length L and initial cable length when entering the hovering state The output is the controlled torque τ of the winding reel. w ;kc It is the control feedback coefficient of the cable reel.

[0027] After completing the above steps, simulations can be performed on the MATLAB / SIMULINK platform. Further, the PX4PSP toolbox and RflySim toolchain can be used to automatically generate code to deploy the tethered drone controller on the flight controller and the reel controller on the Ubuntu18 platform, setting weight coefficients and various feedback coefficients, etc., and conducting experiments.

[0028] The advantages and beneficial effects of this invention are as follows: A wind-following hovering controller based on admittance control is designed for tethered drones. By using the attitude angle of the tethered drone as the control input, the desired hovering position of the tethered drone is actively controlled to drift with the wind. During this process, the tension of the tethering rope is used to assist in counteracting the wind force, thereby preventing the tethered drone from hovering too low in the wind. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of a tethered drone.

[0030] Figure 2 This is the entire implementation process of the present invention.

[0031] Figure 3 This invention relates to the overall control structure of a tethered unmanned aerial vehicle (UAV) system that adopts admittance control for wind-following hovering.

[0032] Figure 4a This is a schematic diagram of the horizontal wind field set up in the simulation.

[0033] Figure 4b This is a comparison chart of the 3D position curves and 3D attitude curves of a tethered UAV hovering with the wind using the method of this invention and hovering at a fixed point without the method of this invention, under a simulated wind field setting. Detailed Implementation

[0034] This invention designs a hovering control algorithm for tethered unmanned aerial vehicles (UAVs). Upon encountering an external wind field, the tethered UAV actively adjusts to its desired hovering position and hovers, utilizing rotor thrust and tether tension to counteract wind resistance, thereby preventing excessively low attitude angles when hovering against headwinds. Simulations were conducted on a computer running Windows 10 with a 3.20GHz CPU and 32.00GB of RAM, using MATLAB R2022b with the PX4PSP toolbox and RflySim toolchain. After successful simulation, code can be automatically generated using the RflySim toolchain for real-world flight testing.

[0035] Step 1: Measure the parameters of the tethered drone system and establish a mechanical model of the tethered drone and the cable reel. The mass of the tethered drone is m = 1.02 kg; the moment of inertia is: J = diag(0.0126-0.0055-0.0158) kg·m 2 The acceleration due to gravity is g = 9.8 m / s². 2 The radius of the cable reel is r = 0.05 m, and the moment of inertia of the cable reel is J. w =1.25·10 -3 kg·m 2 The dynamic model of the tethered UAV is constructed by referring to equations (1)-(3).

[0036] Step 2: Design the top-level admittance controller for the tethered UAV. Set the feedback coefficients D = diag(0.0125 - 0.0125) and K = diag(0.0625 - 0.0625). Obtain the attitude angle of the UAV, calculate the control input Γ with reference to equation (5), and obtain the desired position control quantity of the tethered UAV with reference to equation (4), which is then passed to the bottom-level controller to achieve bottom-level control.

[0037] Step 3: Select a cascaded PID controller as the underlying control method to build the drone-side underlying controller for the tethered drone.

[0038] Step 4: Refer to formula (6) to design the controller for the reel, and select k. c =6, the initial length of the tether rope is L0 = 8.8m.

[0039] Step 5: Conduct a simulation control experiment based on steps 1 through 4. For example... Figure 4a As shown, a wind speed of v is applied 40 seconds after the simulation begins. w =[-8-50] T The wind speed changes to v 90 seconds after the simulation begins. w =[-5-30] T The wind speed changes to v 140 seconds after the simulation begins. w =0. The tethered UAV using the hovering control method of this invention is compared with a tethered UAV not using the wind-flying hovering control method of this invention; that is, the desired horizontal hovering position of the control group's tethered UAV in step two of the control process is 0. The simulated 3D position curves and attitude curves obtained from the two groups are as follows: Figure 4b As shown.

[0040] Simulation results show that, under the same conditions, the tethered drone using the control method of this invention, with the assistance of the tether rope, exhibits a flatter attitude angle compared to the control group. Furthermore, the greater the external wind speed, the more the tethered drone changes its hovering position with the wind; when the wind speed decreases to 0, the tethered drone using the method of this invention can autonomously return to the airspace above the reel.

Claims

1. A method for controlling the wind-following hovering of a tethered unmanned aerial vehicle (UAV) based on admittance control, characterized in that, The steps are as follows: Step 1: Establish a complete dynamic model of the tethered drone and its reel; Step 2: Solve for the desired hovering position of the tethered drone using an admittance model, and design the controller for the tethered drone; Step 3: Achieve hovering of the tethered drone at the new desired hovering location through low-level control; In step two, the controller as a whole comprises a top-level admittance controller and a bottom-level position and attitude control, and is divided into two parts: the control of the UAV and the control of the reel. For the UAV controller, the desired hovering position p of the tethered UAV under the wind field is first solved by using the UAV's attitude R as the input to the admittance controller. d Then the desired hovering position p d The actual position p is transmitted to the position controller to obtain the desired rotor thrust T. rd Desired UAV attitude R d The next step will be to determine the desired drone attitude R. d The desired three-dimensional torque τ provided by the rotor is obtained by using the actual attitude R as the input to the attitude controller. rd For the cable reel, the length of the tethering rope is used as the control input. This value is obtained in the simulation by the position p of the tethered UAV. The control output, τ, is obtained through the cable reel's controller. wd The control torque of the reel is then transmitted into the dynamic model of the reel to obtain the tension T of the mooring rope. t Ultimately, the desired drone rotor thrust T will be achieved. rd The expected three-dimensional torque τ provided by the rotor rd And the tension T of the mooring rope t The dynamic model of the tethered drone is fed into the drone's attitude R and position p, and then fed back to the admittance controller, position controller, attitude controller, and reel controller mentioned above. Design a top-level admittance controller; using a preset admittance controller, the attitude angle of the tethered UAV is used as the control input to obtain the desired hovering position of the tethered UAV; here, only the horizontal position is considered to change with the wind through the admittance model, but the altitude is kept constant; the direction of the rotor thrust, expressed in terms of attitude angle, is used as the input to the controller to obtain the desired hovering position. The controller is designed as follows: in, Let Γ be the desired horizontal hovering position of the tethered UAV; D is the damping coefficient, a positive definite matrix; K is the stiffness coefficient, a positive definite matrix; Γ is the controller input, expressed in terms of rotor thrust and attitude angle respectively: In the controller described above, the desired horizontal hovering position of the tethered UAV when it is stable is determined by matrices D and K. The smaller the parameter value in matrix D, the faster the adjustment speed of the desired hovering position; the smaller the parameter value in matrix K, the farther the desired hovering position.

2. The wind-following hovering control method for a tethered unmanned aerial vehicle based on admittance control according to claim 1, characterized in that: In step one, a dynamic model of the tethered drone is established. The forces acting on the tethered drone in the air include gravity, aerodynamic forces, tether tension, and rotor thrust. The resulting dynamic model is as follows: Where, p = [p x p y p z ] T and v = [v x v y v z ] T Represents the position and velocity of the tethered drone, respectively; m is the mass of the tethered drone; T r T t and F a Let g represent the rotor thrust, the three-dimensional tension of the tethered drone at the drone's end, and the wind force, respectively, in the geodetic coordinate system; g = [00g] T [ω] is the acceleration due to gravity; ω is the rotational angular velocity of the tethered drone; × It is the cross product matrix of vector ω; J is the moment of inertia; τ r τ g τ t and τ a These represent the torque generated by the rotor thrust, the torque generated by gravity, the tethering rope tension, and the wind torque, respectively.

3. The wind-following hovering control method for a tethered unmanned aerial vehicle based on admittance control according to claim 2, characterized in that: The rotation matrix from the UAV's rotor coordinate system to the geodetic coordinate system is expressed as: Where φ, θ and ψ represent the roll angle, pitch angle and yaw angle of the UAV, respectively.

4. The wind-following hovering control method for a tethered unmanned aerial vehicle based on admittance control according to claim 1, characterized in that: In step one, a dynamic model of the cable reel is established; the dynamic model of the cable reel is expressed by the length of the tethering rope, the cable retrieval speed, and the rotational angular velocity of the cable reel as follows: Where L is the length of the mooring rope, v c J is the linear velocity of the tether rope as it is retrieved by the reel. w It is the moment of inertia of the winding reel, ω w τ is the rotational angular velocity of the cable reel during operation, r is the radius of the cable reel, and τ is the angular velocity of rotation. w It is the torsional torque of the cable reel, F c The tension of the tethering rope at the end of the reel.

5. The wind-following hovering control method for a tethered unmanned aerial vehicle based on admittance control according to claim 1, characterized in that: Combined with the expected flight altitude of the tethered drone The desired hovering position of the tethered drone is:

6. The wind-following hovering control method for a tethered unmanned aerial vehicle based on admittance control according to claim 5, characterized in that: Determine the desired hovering position p d The data is passed to the underlying controllers, including the position controller and attitude controller, to achieve the underlying control of the tethered UAV. The underlying controllers include various controllers that can achieve underlying control and track the desired hovering position, including but not limited to proportional-integral-derivative controllers (PID) and model predictive control (MPC).

7. The wind-following hovering control method for a tethered unmanned aerial vehicle based on admittance control according to claim 1, characterized in that: In step three, the controller for the cable reel is designed as follows: The controller input is the angular velocity ω of the winding reel. w Cable length L and initial cable length when entering hovering state The output is the controlled torque τ of the winding reel. w ;k c It is the control feedback coefficient of the cable reel.

Images