Front wheel turning and anti-swing electric actuating device of unmanned aerial vehicle

Abstract

The invention discloses an unmanned aerial vehicle front wheel turning and anti-swing electric actuating device, and relates to the technical field of unmanned aerial vehicles, the device comprises a motor, the motor serves as a power source to provide power for front wheel turning, a planetary reducer is arranged below the motor, and the output end of the motor is connected with the input end of the planetary reducer. By introducing intelligent control based on an improved deep reinforcement learning algorithm, the device can automatically adjust turning and shimmy damping parameters according to multi-dimensional information such as flight height, speed, attitude, ground friction force, gradient and the like acquired in real time, so that the device adapts to different flight and ground conditions, the universality and adaptability of the device are improved, and the device is suitable for popularization and application. The application range of the unmanned aerial vehicle is widened, manual intervention is reduced by means of the characteristic of automatically adjusting parameters by means of an intelligent algorithm, hysteresis and errors of manual operation are avoided, the automation degree and reliability of operation of the unmanned aerial vehicle are remarkably improved, and classified storage, encryption management and backup of various types of data are achieved by means of a data storage module.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and in particular to an electric actuating device for turning and reducing sway of the front wheels of UAVs. Background Art

[0002] UAVs, as unmanned aerial vehicles controlled by radio remote control equipment or their own program control devices, have been widely used in many fields such as military and civilian in recent years. With its advantages of flexibility, efficiency and low cost, it completes various tasks such as reconnaissance, mapping, logistics and distribution. The UAV's front wheel turning and anti-sway electric actuator is an important part of the UAV. During the UAV's take-off, landing and ground taxiing stages, the device is responsible for controlling the turning action of the front wheel and suppressing the swing of the front wheel to ensure that the UAV can move smoothly and accurately on the ground, which plays a key role in the safe operation of the UAV. The aircraft's front wheel turning drive device is an important component to ensure the aircraft's normal take-off and landing and taxiing maneuvers. It improves the aircraft's landing reliability and anti-deflection capability while improving the aircraft's ground maneuverability.

[0003] However, existing UAV front wheel steering and sway reduction electric actuators still have certain drawbacks. Most existing UAV front wheel steering and sway reduction electric actuators use fixed parameter control methods, which cannot adapt to changes in flight status (such as altitude, speed, and attitude) and ground conditions (such as runway material, slope, and flatness). They rely on manual intervention to adjust parameters, which has a delayed response and cannot promptly respond to rapid changes in the UAV's status, reducing the safety and reliability of UAV operation. Therefore, it is necessary to propose a UAV front wheel steering and sway reduction electric actuator to address the problems of the existing technology. Summary of the Invention

[0004] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide an electric actuation device for turning and reducing the sway of the front wheels of a UAV. The device can automatically and accurately obtain the flight status and ground condition information of the UAV, adjust the turning and reducing sway parameters in real time, improve the versatility and adaptability of the device, and reduce manual intervention.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solutions: an electric actuation device for turning and sway reduction of the front wheels of a drone, the device comprising a motor, which serves as a power source to provide power for turning the front wheels; a planetary reducer disposed below the motor, the output end of the motor being connected to the input end of the planetary reducer; an output shaft disposed below the planetary reducer; an electromagnetic damper mounted between the planetary reducer and the output shaft, the electromagnetic damper being used to suppress sway of the front wheels; a sensor module mounted on the outer surface of the planetary reducer for collecting information on the flight status and ground conditions of the drone; an intelligent controller mounted on the outer surface of the motor, the intelligent controller using an improved deep reinforcement learning algorithm to receive information collected by the sensor module and calculate control parameters of the motor and electromagnetic damper; a communication interface disposed on the outer surface of the intelligent controller for exchanging data with other systems of the drone; and a data storage module mounted on the outer surface of the planetary reducer for storing flight status and ground condition data collected by the sensor module, control parameters of the motor and electromagnetic damper calculated by the intelligent controller, and device operating status information.

[0006] The sensor module is used to collect the flight status and ground condition information of the UAV. The flight status information collected by the sensor module includes the flight altitude h, flight speed v, pitch angle θ, roll angle φ and yaw angle ψ, and the ground condition information includes the ground friction F f , runway slope α and ground roughness parameter S;

[0007] The intelligent controller uses an improved deep reinforcement learning algorithm to receive the information collected by the sensor module and calculate the control parameters of the motor and electromagnetic damper. The improved deep reinforcement learning algorithm is based on the state space S, the action space A and the reward function R for operation, wherein the state space S is composed of the flight height h, the flight speed v, the pitch angle θ, the roll angle φ, the yaw angle ψ, the ground friction F f , runway slope α and ground flatness parameter S, that is, S={h,v,θ,φ,ψ,F f ,α,S}, the action space A is determined by the motor drive parameters P m and the damping force control parameter D of the electromagnetic damper, that is, A={P m ,D}, driving parameter P m Including voltage U, current I, speed n, the reward function R is designed based on the stability, controllability and energy consumption factors of the drone, through the formula R = w1 × S stability +w2×S maneuverability -w3×E consumption Calculate, where S stability is the stability index S maneuverability is the controllability index, E consumptionis the energy consumption index, w1, w2, w3 are weight coefficients, and w1+w2+w3=1.

[0008] Furthermore, the altitude sensor uses the high-precision laser ranging principle to measure the flight altitude h. The laser ranging sensor emits a laser beam to the ground. The laser beam is reflected by the ground and received by the sensor. According to the laser propagation time t and the speed of light c, the laser is measured by the formula The flight altitude is calculated, and the measurement accuracy of the laser ranging sensor is at the centimeter level.

[0009] Furthermore, the speed sensor uses the Doppler speed measurement principle to monitor the flight speed v. The speed sensor transmits electromagnetic waves of a specific frequency. When the electromagnetic waves encounter the moving drone, the Doppler effect occurs and the frequency of the reflected wave changes. By detecting the frequency difference Δf between the reflected wave and the transmitted wave, according to the Doppler effect formula Calculate the flight speed of the drone, where λ is the wavelength of the emitted electromagnetic wave.

[0010] Furthermore, when the acceleration sensor and gyroscope are combined to obtain attitude information, a complementary filtering algorithm is used for data fusion. The acceleration sensor measures the acceleration a of the drone in three axes. x 、a y 、a z , the gyroscope measures the angular velocity ω of the drone x 、ω y 、ω z ,The complementary filtering algorithm obtains attitude information by weighted fusion of the data of the acceleration sensor and the gyroscope, including the pitch angle θ, the roll angle φ and the yaw angle ψ. The specific fusion formula is as follows: θ filtered =(1-K)×θ gyro +K×θ accel 、φ filtered =(1-K)×φ gyro +K×φ accel , ψ filtered =(1-K)×ψ gyro +K×ψ mag Among them, θ gyro 、φ gyro , ψ gyro is the angle measured by the gyroscope, θ accel 、φ accel is the angle calculated by the acceleration sensor, ψ mag is the yaw angle measured by the magnetometer, and K is the filter coefficient.

[0011] Furthermore, the ground friction sensor uses a strain gauge measurement principle to detect the friction force F between the ground and the front wheel. fThe strain gauge sensor is installed on the support structure of the front wheel. When the front wheel contacts the ground and generates friction, the support structure will deform slightly. The strain gauge will change its resistance value as the structure deforms. By measuring the change in the resistance value of the strain gauge ΔR, the strain value is calculated according to the characteristic formula of the strain gauge ΔR=k×∈×R0, where k is the sensitivity coefficient of the strain gauge, ∈ is the strain, and R0 is the initial resistance of the strain gauge. Then, the friction force F between the ground and the front wheel is calculated based on the mechanical relationship. f .

[0012] Furthermore, the slope sensor uses the principle of a dual-axis accelerometer to measure the runway slope α. The dual-axis accelerometer measures the acceleration components in two axes. By analyzing the ratio of the two axial acceleration components, the runway slope is calculated. The calculation formula is: where a x and a y The acceleration components measured by the dual-axis accelerometer in the x- and y-axis directions.

[0013] Furthermore, the ground flatness parameter S is measured by an optical sensor. The optical sensor emits light to the ground and obtains ground flatness information by analyzing the distribution and intensity changes of the reflected light. The optical sensor converts the collected light information into an electrical signal, and obtains the ground flatness parameter S through signal processing and algorithm analysis.

[0014] Furthermore, when the intelligent controller uses the improved deep reinforcement learning algorithm, it adopts the experience replay mechanism to optimize the algorithm performance. The experience replay mechanism replays the sample data (s) of the state, action, reward and next state generated during the interaction between the intelligent controller and the environment. t ,a t ,r t ,s t+1 ) is stored in the experience replay buffer, the capacity of the experience replay buffer is N, and M sample data (M <N)。

[0015] Furthermore, the data storage module uses solid-state hard drive technology for data storage. The data storage module classifies and stores the flight status and ground condition data collected by the sensor module, the control parameters of the motor and electromagnetic damper calculated by the intelligent controller, and the device working status information. The flight altitude and speed data are stored in a specific storage area according to time series, the ground condition data is stored in another area, and the control parameters of the motor and electromagnetic damper and the device working status information are respectively stored in corresponding areas. The data storage module uses advanced encryption standards to encrypt and protect the stored data.

[0016] Compared with the existing technology, the UAV front wheel turning and sway reduction electric actuation device has the following beneficial effects:

[0017] By introducing intelligent control based on an improved deep reinforcement learning algorithm, the present invention can enable the device to automatically adjust turning and swing reduction parameters according to multi-dimensional information such as flight altitude, speed, posture, ground friction, slope, etc. collected in real time, thereby adapting to different flight and ground conditions, improving the versatility and adaptability of the device, and broadening the application scope of drones. By relying on the characteristics of automatic parameter adjustment of intelligent algorithms, it can reduce human intervention, avoid the lag and errors of manual operation, and significantly improve the degree of automation and reliability of drone operation. The data storage module is used to classify, store, encrypt and manage various types of data and back up them.

[0018] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0020] Figure 1 This is a schematic diagram of the three-dimensional structure of the left view of a UAV front wheel steering and sway reduction electric actuation device;

[0021] Figure 2 This is a schematic diagram of the three-dimensional structure of the right side view of a UAV front wheel steering and sway reduction electric actuation device;

[0022] Figure 3 This is a system control diagram of a UAV front wheel steering and sway reduction electric actuation device.

[0023] In the figure: 1. Motor; 2. Planetary reducer; 3. Electromagnetic damper; 4. Output shaft; 5. Sensor module; 6. Intelligent controller; 7. Communication interface; 8. Data storage module. DETAILED DESCRIPTION

[0024] The following is a clear and complete description of the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] Example 1

[0026] In a large logistics park, hundreds of drones shuttle back and forth every day, carrying out heavy cargo transportation tasks. The flight routes of these drones cover the entire park and the surrounding distribution areas. The flight environment is complex and changeable. In different seasons and weather conditions, drones need to cope with severe weather such as strong winds and rain. The runway conditions at multiple take-off and landing points in the park are also uneven. Some runways have suffered varying degrees of wear and tear due to long-term use, resulting in poor flatness, while others have a certain slope due to terrain reasons. The runway surface materials include cement, asphalt and other types.

[0027] In such an environment, the front wheel turning and anti-sway electric actuators face severe tests every time the drone takes off and lands. Any slight control error may cause the drone to deviate from its route, excessive swing of the front wheels, or even fall while gliding on the ground, thereby affecting the on-time delivery of goods. In serious cases, it may also cause damage to the drone and loss of goods. Therefore, ensuring that the device can operate stably and accurately is crucial to ensuring the efficient operation of the logistics park.

[0028] Before the UAV is ready to take off to perform its transport mission, the sensor module 5 installed on the outer surface of the planetary reducer 2 enters a fully operational state. The altitude sensor uses a high-precision laser ranging principle to emit a laser beam to the ground at a very high frequency. Each laser beam is quickly reflected after encountering the ground. The altitude sensor accurately records the propagation time t of the laser. According to the constant characteristic of the speed of light c, the altitude sensor can be used to calculate the altitude. Real-time and accurate measurement of the drone's current flight altitude h is crucial for attitude adjustment and path planning during different flight phases. For example, during takeoff and landing, accurate altitude data can help the drone reasonably control its speed and angle.

[0029] The speed sensor uses the Doppler speed measurement principle to continuously emit electromagnetic waves of a specific frequency. When these electromagnetic waves encounter a flying drone, the Doppler effect causes the reflected wave frequency to change. The speed sensor detects the frequency difference Δf between the reflected wave and the transmitted wave, and calculates the frequency difference according to the formula Where λ is the wavelength of the emitted electromagnetic wave, and the flight speed v of the drone is accurately calculated. Accurate acquisition of the flight speed enables the intelligent controller 6 to reasonably adjust the output power of the motor 1 according to different flight phases and mission requirements to ensure stable flight of the drone.

[0030] The accelerometer and gyroscope work closely together and use complementary filtering algorithms to obtain accurate attitude information. The accelerometer monitors the acceleration of the drone in three axes in real time. x 、a y 、a z, while the gyroscope synchronously measures the angular velocity ω of the drone x 、ω y 、ω z , through a complex and precise complementary filtering algorithm, namely formula θ filtered =(1-K)×θ gyro +K×θ accel 、φ filtered =(1-K)×φ gyro +K×φ accel , ψ filtered =(1-K)×ψ gyro +K×ψ mag , the pitch angle θ, roll angle φ and yaw angle ψ of the UAV are calculated. These attitude information are crucial for the UAV to maintain a stable flight attitude and avoid rolling and yaw, especially in complex weather conditions and changeable flight environments.

[0031] When the front wheel contacts the ground and generates friction, the supporting structure will undergo extremely small deformation. This deformation will cause the resistance value of the strain gauge to change. The ground friction sensor accurately measures the change in the strain gauge resistance value ΔR and calculates the strain value according to the formula ΔR=k×∈×R0, where k is the sensitivity coefficient of the strain gauge, ∈ is the strain, and R0 is the initial resistance of the strain gauge. The friction force F between the ground and the front wheel is then obtained based on complex mechanical relationships. f , knowing the magnitude of the ground friction force helps the intelligent controller 6 to reasonably adjust the driving force of the motor 1 and the damping force of the electromagnetic damper 3 when the drone is gliding, ensuring the accuracy and stability of the front wheel turning.

[0032] The slope sensor uses the principle of dual-axis accelerometer to continuously monitor the acceleration components a in two axes. x and a y , through the precise analysis of these two acceleration components, using the formula The runway slope α is measured. The runway slope information is crucial for the attitude adjustment and power distribution of the UAV during takeoff, landing and taxiing. The intelligent controller 6 can adjust the output torque of the motor 1 and the damping force of the electromagnetic damper 3 in advance according to the size and direction of the slope to ensure that the UAV can travel smoothly on runways with different slopes.

[0033] The optical sensor emits a specific pattern of light to the ground, then carefully analyzes the distribution and intensity changes of the reflected light. After complex signal processing and advanced algorithm analysis, it obtains the ground flatness parameter S. The ground flatness parameter can help the intelligent controller 6 understand the condition of the runway surface. For drones flying on uneven runways, the intelligent controller 6 can adjust the damping force of the electromagnetic damper 3 according to the value of S to better suppress abnormal swinging of the front wheel caused by uneven ground.

[0034] The sensor module 5 collects the flight height h, flight speed v, pitch angle θ, roll angle φ, yaw angle ψ, ground friction F f , runway slope α and ground flatness parameter S and other data are transmitted to the intelligent controller 6 and the data storage module 8 in a high-speed and stable manner in real time.

[0035] The intelligent controller 6 quickly receives the rich data from the sensor module 5 and uses it as the current state s t Based on the powerful improved deep reinforcement learning algorithm, the intelligent controller 6 clearly defines the state space S={h,v,θ,φ,ψ,F f ,α,S} and action space A={P m ,D}, where P m It covers key driving parameters of the motor 1 such as voltage U, current I, and speed n, and D represents the damping force control parameter of the electromagnetic damper 3.

[0036] Intelligent controller 6 is based on a carefully designed reward function R = w1 × S stability +w2×S maneuverability -w3×E consumption , where S stability It is a stability index used to measure the attitude stability of the UAV during flight; S maneuverability E is a controllability index that reflects the UAV's ability to respond to various operating instructions; consumption is the energy consumption index, which reflects the energy consumption of the drone during operation. w1, w2, and w3 are weight coefficients, and w1+w2+w3=1. Carefully select an optimal action a from the action space A. t , that is, accurately calculating the optimal control parameters of the motor 1 and the electromagnetic damper 3 in the current state. The calculation of these parameters fully considers multiple factors such as the stability, controllability and energy consumption of the UAV to ensure that the UAV can achieve efficient and safe operation in various complex environments. At the same time, the intelligent controller 6 promptly feeds back the relevant parameters and detailed calculation process information to the data storage module 8 for storage, so as to facilitate subsequent data analysis and system optimization.

[0037] Motor 1, as the power source for turning the drone's front wheels, quickly and accurately outputs the corresponding torque based on the precise control signal output by the intelligent controller 6. This torque is transmitted to the front wheels through a carefully designed transmission system to achieve precise turning of the front wheels. Throughout the entire process, the output torque of motor 1 is dynamically adjusted according to the real-time flight status and ground conditions, ensuring that the drone can achieve stable and accurate turning under different glide speeds and turning radius requirements.

[0038] The electromagnetic damper 3 quickly and accurately adjusts its own damping force according to the instructions of the intelligent controller 6. During the gliding process of the UAV, especially when passing through an uneven runway or being disturbed by external factors, the electromagnetic damper 3 can respond in time and effectively suppress the swing of the front wheel by adjusting the damping force, so that the front wheel maintains a stable running trajectory. This precise damping force control greatly improves the stability and controllability of the UAV when gliding on the ground, and reduces the risk of accidents caused by the swing of the front wheel.

[0039] The intelligent controller 6 exchanges data with the UAV's flight control system, navigation system, and other key systems at high speed and in a stable manner through a powerful communication interface 7. On the one hand, the intelligent controller 6 obtains more important flight-related information from these systems, such as navigation data and flight plan adjustment instructions, in order to further optimize its own control strategy. On the other hand, the intelligent controller 6 feeds back key information such as the device's real-time working status, current turning and sway reduction parameters, etc. to other systems, thereby achieving collaborative work and information sharing among the various systems. This efficient communication interaction mechanism ensures close coordination among the various systems of the UAV throughout the entire flight process, thereby improving the overall operational efficiency and safety of the UAV.

[0040] The data storage module 8 uses advanced solid-state hard drive technology to classify and orderly store the large amount of data received. Flight status data such as flight altitude and speed are accurately stored in time series to form a detailed flight record; ground condition data, including ground friction, runway slope, ground flatness and other information, are stored in a dedicated area to facilitate analysis of the impact of different ground conditions on the operation of the drone. The control parameters of the motor 1 and the electromagnetic damper 3 and the working status information of the device are also stored in corresponding areas, providing rich data support for subsequent fault diagnosis and system optimization. At the same time, the data storage module 8 uses advanced encryption standards to strictly encrypt and protect the stored data to ensure the security and integrity of the data and prevent the data from being illegally obtained or tampered with during storage and transmission.

[0041] Effects brought about by this embodiment: In this embodiment, through the above-mentioned complete and detailed process, the UAV demonstrates excellent operating performance in the complex and changeable logistics park environment. The application of intelligent control algorithms enables the device to adapt to various runway conditions and flight mission requirements in a highly automated manner without the need for frequent and complex manual intervention. When facing strong winds, the UAV can automatically adjust the output power of the motor 1 and the damping force of the electromagnetic damper 3 according to the real-time wind speed and direction and its own flight status, maintain a stable flight attitude and accurate gliding trajectory, and when passing through uneven runways, it can quickly respond to changes in ground conditions, accurately control the front wheel turning and swing reduction, and ensure safe passage. The real-time collection, efficient storage and in-depth analysis of data provide a solid foundation for subsequent flight optimization and system improvement.

[0042] Example 2

[0043] In a certain military exercise area, the terrain is complex and diverse, including mountains, jungles, and various simulated battlefield environments. UAVs in this area undertake important tasks such as reconnaissance, target positioning, and intelligence transmission. During the execution of the mission, the UAV needs to fly at different altitudes, cross narrow valleys and dense jungles, and take off and land on temporary runways. These runways may have irregular slopes and uneven surfaces due to terrain restrictions. At the same time, the electromagnetic interference in the military environment is also relatively strong, posing a high challenge to the UAV's electronic equipment and control systems. In addition, the UAV's flight mission requires a high degree of confidentiality and real-time performance, which requires the UAV's front wheel steering and anti-sway electric actuators to operate stably and accurately in complex electromagnetic environments and changeable terrain conditions to ensure that the UAV completes various military missions.

[0044] When the UAV receives a reconnaissance mission and prepares to take off, the sensor module 5 installed on the outer surface of the planetary reducer 2 immediately starts working. The altitude sensor continuously emits a laser beam to the environment below based on the principle of high-precision laser ranging. In mountainous and jungle environments, the reflection of the laser beam is more complicated, but the altitude sensor, with its high sensitivity and fast data processing capabilities, accurately records the propagation time t of the laser and calculates the value of the laser beam through the formula Obtaining accurate flight altitude h in real time is crucial for drones to maintain a safe flight altitude when crossing valleys and avoiding jungle obstacles.

[0045] The speed sensor can still stably monitor the flight speed v using the Doppler speed measurement principle in a complex electromagnetic interference environment. Although electromagnetic interference may have a certain impact on the propagation and reception of electromagnetic waves, the speed sensor uses advanced anti-interference technology and signal processing algorithms to accurately detect the frequency difference Δf between the reflected wave and the transmitted wave, and calculates the speed v according to the formula Calculating the flight speed of the drone ensures that the drone can adjust its flight attitude and power output according to the actual speed during different flight phases, such as acceleration, deceleration and cruising.

[0046] The accelerometer and gyroscope work closely together under complex flight attitude changes to obtain accurate attitude information through complementary filtering algorithms. When crossing valleys and avoiding obstacles, the attitude of the drone changes frequently and drastically. The accelerometer measures the acceleration of the three axes in real time. x 、a y 、a z , the gyroscope synchronously measures the angular velocity ω x 、ω y 、ω z , using the formula θ filtered =(1-K)×θ gyro +K×θ accel 、φ filtered =(1-K)×φ gyro +K×φ accel , ψ filtered =(1-K)×ψ gyro +K×ψ mag , accurately calculate the pitch angle θ, roll angle φ and yaw angle ψ, providing accurate attitude data support for the stable flight of the UAV.

[0047] The ground friction sensor uses the strain gauge measurement principle to accurately detect the friction force F between the ground and the front wheel under the special ground conditions of the simple runway. f The surface material of a simple runway may be soil, sand, etc., which is very different from a conventional runway. This requires the sensor to be able to accurately adapt to different ground conditions. When the front wheel contacts the ground, the sensor measures the change in the resistance value of the strain gauge ΔR, and calculates the strain value according to the formula ΔR=k×∈×R0, thereby deriving the friction force F between the ground and the front wheel. f This data is crucial for the intelligent controller 6 to adjust the driving force of the motor 1 and the damping force of the electromagnetic damper 3 during the take-off and landing phases, ensuring that the drone can take off and land smoothly on surfaces with different friction forces.

[0048] The slope sensor uses the principle of dual-axis accelerometer to accurately measure the runway slope α on a simple runway with irregular slope. Since the slope of the simple runway may change at any time, the slope sensor needs to quickly and accurately monitor the two axial acceleration components a. x and a y , through the formula The runway slope is calculated in real time, which enables the intelligent controller 6 to adjust the attitude and power of the drone in advance to adapt to different slope conditions and ensure safe take-off and landing.

[0049] In a complex battlefield environment, the optical sensor obtains the ground flatness parameter S by emitting and analyzing reflected light. On a temporary runway, there may be potholes, bumps and other unevenness. The optical sensor uses advanced image processing algorithms to conduct a detailed analysis of the distribution and intensity changes of the reflected light to obtain an accurate ground flatness parameter S. This parameter helps the intelligent controller 6 better control the electromagnetic damper 3 during the sliding process, suppressing the swing of the front wheel caused by the uneven ground.

[0050] The sensor module 5 collects all the data, including flight altitude h, flight speed v, pitch angle θ, roll angle φ, yaw angle ψ, ground friction F f , runway slope α and ground flatness parameter S are transmitted to the intelligent controller 6 and data storage module 8 in real time and accurately through the anti-interference data transmission line.

[0051] The intelligent controller 6 receives the data from the sensor module 5 quickly and accurately in a complex electromagnetic interference environment and uses it as the current state s t , based on the improved deep reinforcement learning algorithm, the state space S={h,v,θ,φ,ψ,F f ,α,S} and action space A={P m ,D}, under the special requirements of military tasks, the reward function R=w1×S stability +w2×S maneuverability -w3×E consumption The weight coefficients w1, w2, and w3 in the task are dynamically adjusted according to the priority and characteristics of the task. For example, when performing a reconnaissance mission, in order to ensure the concealment and safety of the UAV, the stability index S may be appropriately increased. stability The weight w1.

[0052] The intelligent controller 6 carefully selects an optimal action a from the action space A based on the adjusted reward function. t That is, the optimal control parameters of the motor 1 and the electromagnetic damper 3 in the current state are accurately calculated. During the calculation process, the intelligent controller 6 fully considers various factors in the military environment, such as the impact of electromagnetic interference on the motor and sensors, the requirements of complex terrain on flight attitude, etc. At the same time, the relevant parameters and calculation process information are promptly fed back to the data storage module 8 for storage, so as to facilitate subsequent detailed analysis and summary of the mission execution.

[0053] In a complex military environment, motor 1 strictly follows the control signal output by intelligent controller 6 and accurately outputs the corresponding torque. When crossing a valley, in order to maintain a stable flight speed and attitude, motor 1 needs to dynamically adjust the output torque according to the real-time flight status and terrain information to ensure that the drone can smoothly pass through narrow spaces. During the take-off and landing phase, motor 1 accurately controls the driving force of the front wheel according to the slope, friction and other conditions of the runway, so that the drone can take off and land smoothly on uneven simple runways.

[0054] The electromagnetic damper 3 quickly adjusts the damping force in a complex battlefield environment according to the instructions of the intelligent controller 6. During the gliding process of the UAV, facing the unevenness of the ground and external interference, the electromagnetic damper 3 can respond quickly and effectively suppress the swing of the front wheel by accurately adjusting the damping force, ensuring that the UAV can maintain a stable driving trajectory under various complex ground conditions. This is especially important for UAVs taking off and landing on simple runways, and can greatly improve the controllability and safety of the UAV.

[0055] The intelligent controller 6 exchanges data securely and stably with the UAV's flight control system, navigation system, and other military mission-related systems through a communication interface 7 with high anti-interference capabilities. In a complex electromagnetic interference environment, the communication interface 7 uses advanced encryption technology and anti-interference communication protocols to ensure accurate data transmission. The intelligent controller 6 obtains important information such as mission instructions and navigation data from other systems, and at the same time feeds back the device's working status, current turning and sway reduction parameters, etc. to other systems, achieving collaborative work between systems and ensuring the smooth execution of military missions.

[0056] In response to the special needs of military missions, the data storage module 8 uses highly secure solid-state hard drive technology to strictly classify and store the collected and processed data. In order to ensure the confidentiality of the data, the data storage module 8 uses multiple encryption algorithms to encrypt and protect the data. Flight status data, ground condition data, control parameters, and device working status information are all stored in an orderly manner so that detailed review and analysis can be carried out after the mission is completed. These data are of great significance for evaluating the performance of drones in complex military environments, optimizing control algorithms, and improving device design.

[0057] Effects brought about by this embodiment: In this military application scenario embodiment, through the above-mentioned complete and targeted process, the UAV demonstrates excellent performance and reliability in a complex military environment. The intelligent control algorithm and advanced sensor technology enable the device to be highly adaptable to complex and changeable terrain and strong electromagnetic interference environment, and precise control can be achieved without excessive human intervention. When crossing narrow valleys and dense jungles, the UAV can automatically adjust the flight attitude and power output according to real-time environmental information and mission requirements to ensure flight safety and smooth execution of the mission. During takeoff and landing on a simple runway, the device can accurately cope with irregular slopes and uneven ground, and achieve smooth takeoff and landing by precisely controlling the motor 1 and the electromagnetic damper 3. The secure storage and efficient interaction of data provide strong support for the command and decision-making of military missions, and also accumulate valuable data experience for subsequent equipment improvements and tactical optimization.

[0058] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and scope of the same elements of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An electric actuation device for turning and reducing the sway of the front wheels of a UAV, characterized in that: The device comprises a motor (1), the motor (1) serving as a power source to provide power for the front wheel to turn, a planetary reducer (2) being provided below the motor (1), the output end of the motor (1) being connected to the input end of the planetary reducer (2), an output shaft (4) being provided below the planetary reducer (2), an electromagnetic damper (3) being installed between the planetary reducer (2) and the output shaft (4), the electromagnetic damper (3) being used to suppress the swing of the front wheel, a sensor module (5) being installed on the outer surface of the planetary reducer (2) for collecting information on the flight status and ground conditions of the UAV, and the outer surface of the motor (1) An intelligent controller (6) is installed, and the intelligent controller (6) uses an improved deep reinforcement learning algorithm to receive information collected by the sensor module (5) and calculate the control parameters of the motor (1) and the electromagnetic damper (3). The outer surface of the intelligent controller (6) is provided with a communication interface (7) for data interaction with other systems of the unmanned aerial vehicle. The outer surface of the planetary reducer (2) is provided with a data storage module (8) for storing the flight status and ground condition data collected by the sensor module (5), the control parameters of the motor (1) and the electromagnetic damper (3) calculated by the intelligent controller (6), and the device working status information; The sensor module (5) is used to collect the flight status and ground condition information of the UAV. The flight status information collected by the sensor module (5) includes the flight altitude h, flight speed v, pitch angle θ, roll angle and yaw angle ψ, ground condition information includes ground friction F f , runway slope α and ground roughness parameter S; The intelligent controller (6) uses an improved deep reinforcement learning algorithm to receive information collected by the sensor module (5) and calculate the control parameters of the motor (1) and the electromagnetic damper (3). The improved deep reinforcement learning algorithm is based on the state space S, the action space A and the reward function R. The state space S is composed of the flight height h, the flight speed v, the pitch angle θ, the roll angle Yaw angle ψ, ground friction F f , runway slope α and ground flatness parameter S, that is The action space A is determined by the drive parameters P of the motor (1) m and the damping force control parameter D of the electromagnetic damper (3), namely A={P m ,D}, driving parameter P m Including voltage U, current I, speed n, the reward function R is designed based on the stability, controllability and energy consumption factors of the drone, through the formula R = w1 × S stability +w2×S maneuverability -w3×E consumption Calculate, where S stability is the stability index S maneuverability is the controllability index, E consumption is the energy consumption index, w1, w2, w3 are weight coefficients, and w1+w2+w3=1.

2. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The height sensor uses the high-precision laser ranging principle to measure the flight height h. The laser ranging sensor emits a laser beam to the ground. The laser beam is reflected by the ground and received by the sensor. According to the laser propagation time t and the speed of light c, the flight height h is measured by the formula The flight altitude is calculated, and the measurement accuracy of the laser ranging sensor is at the centimeter level.

3. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The speed sensor uses the Doppler speed measurement principle to monitor the flight speed v. The speed sensor transmits electromagnetic waves of a specific frequency. When the electromagnetic waves encounter a moving drone, the Doppler effect occurs and the frequency of the reflected wave changes. By detecting the frequency difference Δf between the reflected wave and the transmitted wave, according to the Doppler effect formula Calculate the flight speed of the drone, where λ is the wavelength of the emitted electromagnetic wave.

4. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: When the acceleration sensor and gyroscope are combined to obtain attitude information, a complementary filtering algorithm is used for data fusion. The acceleration sensor measures the acceleration a of the drone in three axes. x 、a y 、a z , the gyroscope measures the angular velocity ω of the drone x 、ω y 、ω z The complementary filtering algorithm obtains attitude information by weighted fusion of the data of the acceleration sensor and the gyroscope, including the pitch angle θ and the roll angle and yaw angle ψ, the specific fusion formula is as follows: θ filtered =(1-K)×θ gyro +K×θ accel 、 ψ filtered =(1-K)×ψ gyro +K×ψ mag Among them, θ gyro 、 ψ gyro is the angle measured by the gyroscope, θ accel 、 is the angle calculated by the acceleration sensor, ψ mag is the yaw angle measured by the magnetometer, and K is the filter coefficient.

5. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The ground friction sensor uses a strain gauge measurement principle to detect the friction force F between the ground and the front wheel. f The strain gauge sensor is installed on the support structure of the front wheel. When the front wheel contacts the ground and generates friction, the support structure will deform slightly. The strain gauge will change its resistance value as the structure deforms. By measuring the change in the resistance value of the strain gauge ΔR, the strain value is calculated according to the characteristic formula of the strain gauge ΔR=k×∈×R0, where k is the sensitivity coefficient of the strain gauge, ∈ is the strain, and R0 is the initial resistance of the strain gauge. Then, the friction force F between the ground and the front wheel is calculated based on the mechanical relationship. f .

6. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The slope sensor uses the principle of a dual-axis accelerometer to measure the runway slope α. The dual-axis accelerometer measures the acceleration components in two axes. By analyzing the ratio of the two axial acceleration components, the runway slope is calculated. The calculation formula is: where a x and a y The acceleration components measured by the dual-axis accelerometer in the x- and y-axis directions.

7. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The ground flatness parameter S is measured by an optical sensor. The optical sensor emits light to the ground and obtains ground flatness information by analyzing the distribution and intensity changes of the reflected light. The optical sensor converts the collected light information into an electrical signal. After signal processing and algorithm analysis, the ground flatness parameter S is obtained.

8. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The intelligent controller (6) uses an experience replay mechanism to optimize the algorithm performance when using the improved deep reinforcement learning algorithm. The experience replay mechanism replays the sample data (s) of the state, action, reward and next state generated during the interaction between the intelligent controller (6) and the environment. t ,a t ,r t ,s t+1 ) is stored in the experience replay buffer, the capacity of the experience replay buffer is N, and M sample data (M <N)。 9. The UAV front wheel turning and sway reduction electric actuation device according to claim 1, characterized in that: The data storage module (8) uses solid-state hard disk technology to store data. The data storage module (8) classifies and stores the flight status and ground condition data collected by the sensor module (5), the control parameters of the motor (1) and the electromagnetic damper (3) calculated by the intelligent controller (6), and the device working status information. The flight altitude and speed data are stored in a specific storage area according to time series, the ground condition data is stored in another area, and the control parameters of the motor (1) and the electromagnetic damper (3) and the device working status information are respectively stored in corresponding areas. The data storage module (8) uses the advanced encryption standard to encrypt and protect the stored data.

Images