RC model car intelligent empennage control system and method based on multi-information fusion

Abstract

The invention relates to the related technical field of RC model cars, and discloses an RC model car intelligent empennage control system based on multi-information fusion, which comprises a signal sensing unit, a core processing unit, a servo drive unit, a rotating shaft execution unit, an empennage body and a power management unit. The signal sensing unit collects and outputs vehicle speed, posture and driving intention information; the core processing unit carries out preprocessing, fusion and decision making on the signals and generates PWM control signals; the servo driving unit drives a servo motor to generate an accurate rotating angle; the rotating shaft executing mechanism converts the rotating angle into empennage angle adjustment; the empennage body generates pneumatic downward pressure adaptive to working conditions; the power management unit provides stable power supply and has a protection function. Through multi-information fusion perception, the arc-shaped groove rail rotating shaft executing mechanism and the closed-loop feedback unit, accurate smooth angle control and multi-scene adaptation of the empennage are achieved, the control stability and the transmission precision are improved, and the problems that in the prior art, angle control is rough, and adaptability is poor are solved.

Description

Technical Field

[0001] This invention relates to the technical field of RC model cars, specifically to an intelligent rear wing control system and method for RC model cars based on multi-information fusion. Background Technology

[0002] RC model cars are widely used in racing and entertainment. As a core aerodynamic component, the tail wing's angle directly determines the aerodynamic performance and handling stability of the model car. At high speeds, a large-angle tail wing is needed to generate downforce and improve grip, while at low speeds, a small-angle tail wing is needed to reduce drag. Therefore, precise control of the tail wing angle is crucial to the performance of RC model cars. Existing RC model car rear wing control schemes have significant technical flaws and are difficult to meet the requirements of high-performance applications: First, fixed-angle rear wings cannot simultaneously meet the dual requirements of low-speed low drag and high-speed high stability, resulting in extremely poor adaptability; Second, manual remote control adjustment is cumbersome, requiring the driver to be distracted by operating the rear wing, and cannot respond in real time to dynamic conditions such as vehicle acceleration, braking, and cornering, which can easily lead to loss of vehicle attitude control; Third, a few schemes that use a single speed signal for control only achieve angle mapping through simple logic, without integrating key information such as vehicle attitude and driving intention, resulting in low control accuracy and a tendency for rear wing oscillation to occur near the speed threshold. Meanwhile, the existing tail wing actuators have unreasonable structural designs, mostly using simple shaft drives, resulting in poor installation adaptability. They cannot flexibly adapt to the installation space of the RC model car's roof, and the transmission accuracy is insufficient, making it difficult to accurately convert the servo motor's rotation angle into tail wing angle changes, thus affecting the control effect. The signal sensing modules mostly do not support multi-protocol parsing, making them incompatible with mainstream remote control receivers, and their attitude detection accuracy is insufficient and their anti-interference ability is weak. The filtering algorithm parameters of the core processing unit are unreasonable, resulting in poor signal fusion and a lack of flexibility in intelligent decision-making. Furthermore, existing power management units lack robust protection mechanisms, making them susceptible to damage from overcurrent, overvoltage, and reverse connection issues, resulting in insufficient power supply stability. In summary, current technology lacks an intelligent tail wing control system that integrates multi-source information, boasts strong structural adaptability, precise control, and stable reliability, making it difficult to meet the demands of high-performance RC model car racing. Developing an intelligent tail wing control system that addresses these shortcomings has become a pressing technical problem in this field. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an intelligent rear wing control system for RC model cars based on multi-information fusion, which solves the problems mentioned in the background. This invention mainly sets up a multi-information fusion sensing, an arc-shaped grooved track rotating shaft actuator, and a closed-loop feedback unit to achieve precise and smooth rear wing angle control and multi-scenario adaptation, improve handling stability and transmission accuracy, and solve the problems of rough angle control, poor adaptability, lack of feedback correction, and inaccurate transmission in existing technologies.

[0004] The intelligent rear wing control system for RC model cars based on multi-information fusion includes a signal sensing unit, a core processing unit, a servo drive unit, a rotating shaft actuator, a rear wing body, and a power management unit. The signal sensing unit is used to collect the vehicle speed information, dynamic posture information and driving intention information of the RC model car, and output the corresponding detection signal to the core processing unit. The core processing unit is used to receive the detection signal output by the signal sensing unit, preprocess, fuse and make intelligent decisions on the signal, generate PWM control signal and output it to the servo drive unit; The servo drive unit is used to receive the PWM control signal output by the core processing unit and drive the servo motor to generate a precise mechanical rotation angle. The rotating shaft actuator is used to convert the mechanical rotation angle of the servo motor into an angle change of the tail fin body, so as to achieve precise adjustment of the tail fin angle; The tail fin body is used to generate aerodynamic downforce adapted to the current working conditions of the RC model car according to the adjusted angle; The power management unit is used to provide stable and isolated power supply to all units of the entire control system, and has power protection functions.

[0005] Preferably, the signal sensing unit includes a speed detection submodule, an attitude detection submodule, and a driving intention detection submodule; The speed detection submodule consists of a Hall sensor, a magnet, and a signal conditioning circuit. The magnet is set in the stable force area of ​​the rear wheel hub, and the Hall sensor is fixed to the corresponding frame. The distance between the Hall sensor and the magnet is 2-5mm. The attitude detection submodule is a six-axis IMU module with a three-axis acceleration measurement range of ±16g, a three-axis angular velocity measurement range of ±2000° / s, and a data update rate of ≥200Hz. The IMU module is installed at the center of gravity of the RC model car and fixed by a shock-absorbing pad. The driving intention detection submodule includes a signal parsing circuit that supports standard PWM, SBUS2, and FPort protocols. It communicates with the RC model car remote control receiver via a UART interface to capture signals from the accelerator, steering, and braking channels. The braking signal is achieved by parsing the reverse accelerator command or the independent braking channel PWM signal.

[0006] Preferably, the core of the core processing unit is a microcontroller, and the core processing unit further includes a signal preprocessing module, an intelligent decision-making algorithm module, a control parameter memory, and a PWM signal generator; The signal preprocessing module incorporates a Kalman filter algorithm.

[0007] The intelligent decision-making algorithm module stores and runs an adaptive fuzzy control algorithm. The input variables of this algorithm include vehicle speed V, longitudinal acceleration A_x, pitch rate ω_y, and throttle opening T. The input variables are divided into seven linguistic variables using a triangular membership function: negative large (NB), negative medium (NM), negative small (NS), zero (Z), positive small (PS), positive medium (PM), and positive large (PB). The output variable is the tail wing angle increment Δα, which is divided into five linguistic variables. The centroid method is used for defuzzification. The control parameter memory stores a speed-angle reference mapping table, 35 fuzzy rule bases and user-defined mode parameters. The speed-angle reference mapping table is as follows: V1=60km / h corresponds to α1=15°, V2=90km / h corresponds to α2=30°, and V3=120km / h corresponds to α3=45°. The PWM signal generator is used to output PWM control signals in the range of 500-2500μs.

[0008] Preferably, the servo drive unit includes a dedicated servo drive chip and an external filtering circuit. The dedicated servo drive chip is used to drive an RC servo motor. The output shaft rotation angle range of the RC servo motor is 0°-180°, and the output torque is ≥13kg·cm. The servo drive unit receives the PWM control signal output by the core processing unit, outputs a large current to the servo motor through the drive chip, and converts the digital command into the precise mechanical rotation angle of the servo motor.

[0009] Preferably, the rotating shaft actuator includes a servo arm, a rotating shaft, a sliding block, and a support base. One end of the servo arm is fixed to the output end of an RC servo, the RC servo is fixedly mounted on a servo mounting base, and the servo mounting base is fixed to the inner bottom surface of the upper hood corresponding to the rear of the vehicle. The other end of the servo arm is fixedly connected to the middle of the rotating shaft. Both ends of the rotating shaft are inserted into the sliding block. One end of the sliding block is slidably inserted into the support base, and the other end of the sliding block is fixedly connected to the tail wing body. The lower end of the support base is fixed to the upper surface of the upper hood corresponding to the rear of the vehicle. In this invention, the RC servo drives the servo arm to rotate, and the servo arm drives the rotating shaft to move up and down. The sliding groove in the support base is arc-shaped. When the servo drives the rotating shaft to move up and down, it will in turn drive the sliding block to move up and down within the sliding groove of the support base. Because the sliding groove is arc-shaped, the angle of the tail wing body will also change while moving up and down.

[0010] Preferably, the power management unit includes a lithium battery pack, a DC-DC step-down module, and a power protection circuit; the power protection circuit includes overcurrent protection, overvoltage protection, reverse connection protection, and a filtering circuit; the output terminal of the lithium battery pack is connected to the input terminal of the DC-DC step-down module; the output terminal of the DC-DC module is connected to the power interface of each unit; and the protection circuit is connected in series with the power input terminal.

[0011] Preferably, the closed-loop intelligent control system further includes a feedback unit, which is a miniature conductive plastic potentiometer. The potentiometer is installed at the pivot of the tail fin body and is used to collect the actual angle signal of the tail fin and feed it back to the core processing unit. The core processing unit corrects the tail fin angle deviation according to the feedback signal to achieve full closed-loop control.

[0012] Preferably, the intelligent decision-making of the core processing unit includes a basic mode and a preferred mode; The basic mode adopts a speed-angle mapping method with hysteresis. When the speed of the RC model car exceeds 90km / h, the tail wing body angle increases; when the speed is below 85km / h, the tail wing body angle decreases, and the hysteresis difference ΔV=5km / h. The preferred mode initiates an adaptive fuzzy control algorithm, which takes the filtered vehicle state vector as input, calls the fuzzy rule base for inference calculation, and obtains the tail wing angle increment Δα after defuzzification. This increment is then added to the current tail wing angle to obtain the new target angle. Preferably, the core processing unit also supports a learning mode. In the learning mode, the system records the lap time data for three consecutive laps, removes outliers exceeding ±5% of the average lap time, takes the minimum remaining lap time as the optimal lap time, and simultaneously records the vehicle speed, attitude, and rear wing angle parameters for that lap. Based on the parameters corresponding to the optimal lap time, the speed-angle reference mapping table is fine-tuned in 0.5° increments, with the adjustment range not exceeding ±10% of the original parameters, and parameter drift is avoided through a limiting function.

[0013] The intelligent rear wing control method for RC model cars based on multi-information fusion, applied to any of the above-mentioned intelligent rear wing control systems for RC model cars based on multi-information fusion, includes the following steps: Step 1: Synchronous acquisition of multi-source information. After the system is powered on, the pulse signal of the speed detection submodule, the attitude data of the IMU module, and the throttle / steering / brake PWM signal of the driving intention submodule are acquired synchronously with a control cycle of 10ms. Step 2: Data preprocessing and fusion. The core processing unit filters the raw data through the signal preprocessing module. The speed signal is shaped and filtered, and then converted into real-time vehicle speed through pulse counting. The IMU data and vehicle speed data are fused through Kalman filtering to eliminate errors caused by wheel slippage and vibration, and obtain an accurate vehicle state vector. Step 3: Intelligent decision-making: Select the basic mode or the preferred mode according to the preset mode to generate the tail fin target angle; Step 4: Precise execution and closed-loop feedback. The core processing unit generates a PWM signal corresponding to the target angle and transmits it to the servo drive unit. The servo motor drives the tail fin body to rotate to the target position through the four-axis mechanism. The feedback unit collects the actual angle signal of the tail fin and feeds it back to the core processing unit. The core processing unit corrects the angle deviation. Step 5: Adaptive learning. If learning mode is enabled, the system will determine the best lap time and fine-tune the parameters to achieve personalized adaptation of the bike.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention relates to an intelligent rear wing control system for RC model cars based on multi-information fusion. It incorporates a signal sensing unit that integrates vehicle speed, dynamic attitude, and driving intention information. This is combined with a core processing unit that utilizes Kalman filtering data fusion and adaptive fuzzy control for intelligent decision-making. The signal sensing unit simultaneously collects three types of key information through its three sub-modules. Kalman filtering eliminates interference from wheel slippage and vibration, achieving precise data fusion. Adaptive fuzzy control then performs expert-level inference calculations on multiple variables to generate the target angle. This addresses the problems of existing technologies that rely solely on speed control, lack multi-source information fusion leading to low control accuracy, oscillations near speed thresholds, and inability to respond to complex conditions. It achieves proactive and adaptive adjustment of the rear wing angle. Furthermore, this system not only completely eliminates the risk of loss of control during rapid acceleration, heavy braking, and cornering, but also proactively suppresses "nose-up" and "tail-swing" phenomena, increasing cornering speed by 10-15% and breaking through the handling limits of existing RC model cars. 2. This invention relates to an intelligent rear wing control system for RC model cars based on multi-information fusion. It features a rotating shaft actuator composed of a servo linkage, a rotating shaft, sliding blocks, and a support base. The support base contains an arc-shaped sliding groove. By fixing the RC servo to the inner bottom surface of the upper cover and the support base to the outer surface of the upper cover, the RC servo drives the servo linkage to rotate. The servo linkage then drives the rotating shaft to move up and down. Simultaneously, the sliding blocks at both ends of the rotating shaft move up and down within the arc-shaped sliding groove of the support base. With the guiding effect of the arc-shaped groove, the movement of the sliding blocks simultaneously changes the angle of the rear wing body, thus solving the problem of… Existing tail wing actuators suffer from simple structures, poor installation adaptability, inability to fit the installation space of the RC model car's canopy, and insufficient transmission precision. This new structure achieves precise and smooth transmission of tail wing angle. On the other hand, it perfectly fits the installation space of the RC model car's canopy, and through the guiding effect of the arc-shaped sliding rail, it makes the tail wing angle adjustment more in line with the aerodynamic requirements, with smoother and more seamless movements. The transmission error is controlled within 0.5°. At the same time, it takes into account the compactness and durability of the structure. The arc-shaped rail can also reduce transmission wear and further improve the service life of the mechanism, meeting the high-frequency and high-precision adjustment requirements in competition scenarios. 3. This invention is based on a multi-information fusion-based intelligent rear wing control system for RC model cars. It includes a driving intention detection submodule supporting standard PWM, SBUS2, and FPort protocols, and a high-precision six-axis IMU attitude detection submodule. Through multi-protocol parsing, it adapts to mainstream remote control receivers. The IMU module is fixed at the center of gravity and equipped with shock-absorbing pads to improve detection accuracy. This solves the problems of existing signal sensing modules not supporting multiple protocols, insufficient attitude detection accuracy, weak anti-interference capabilities, and poor versatility, enabling adaptation to multiple brands of RC model cars and accurate signal acquisition. Furthermore, this design significantly improves the system's versatility, allowing it to adapt to over 90% of RC racing models on the market without additional modifications. The attitude detection response speed is faster than existing products, capturing millisecond-level attitude changes, providing accurate data support for intelligent decision-making. 4. This invention relates to an intelligent rear wing control system for RC model cars based on multi-information fusion. It incorporates a signal preprocessing module with a built-in dedicated parameter Kalman filter algorithm. By pre-setting parameters for process noise covariance, observation noise covariance, and initial state covariance adapted to the RC model car scenario, it specifically optimizes the filtering effect. On one hand, this solves the problems of unreasonable filtering algorithm parameters and poor signal fusion effect in existing core processing units, achieving efficient fusion of multi-source data. On the other hand, this dedicated parameter setting reduces the error of the filtered signal by more than 60%, ensuring the stability of data acquisition even in the severe vibration environment of complex tracks, and avoiding incorrect rear wing adjustments caused by signal interference. 5. This invention, based on a multi-information fusion-based intelligent rear wing control system for RC model cars, incorporates a power management unit with overcurrent, overvoltage, reverse connection, and filtering protection. It provides stable power supply at multiple levels via a DC-DC step-down module and achieves comprehensive protection through a protection circuit connected in series at the power input. This addresses the problems of existing power management units lacking robust protection mechanisms, being prone to module damage due to power supply anomalies, and having insufficient power supply stability, thus achieving stable power supply for the entire system. Furthermore, this design reduces the power supply failure rate to below 0.1% under high-frequency, high-load operation. Simultaneously, the filtering circuit suppresses voltage ripple, preventing power supply interference from affecting the core processing unit's algorithm operation and signal acquisition, thereby improving the overall reliability of the system. 6. This invention is based on a multi-information fusion-based intelligent tail wing control system for RC model cars. It features a miniature conductive plastic potentiometer feedback unit installed at the tail wing pivot. This unit collects the actual tail wing angle signal and feeds it back to the core processing unit, forming a closed-loop control and correcting angle deviations. On the one hand, it solves the problems of existing control systems lacking feedback mechanisms, insufficient control precision, and inability to correct angle deviations, thus achieving precise control of the tail wing angle. On the other hand, the accuracy of this feedback unit can reach 0.1°, enabling real-time correction of minor deviations in the transmission and execution process. This ensures that the deviation between the actual tail wing angle and the target angle is always controlled within 0.5°, significantly improving the stability of aerodynamic performance and indirectly shortening the lap time in a race. 7. This invention is based on a multi-information fusion-based intelligent rear wing control system for RC model cars. It sets up two intelligent decision-making modes, a basic mode and an optimal mode, as well as a learning mode. The basic mode avoids speed threshold oscillations, the optimal mode handles complex working conditions, and the learning mode records the best lap time parameters and fine-tunes the control parameters. On the one hand, it solves the problems of existing control systems having a single decision-making mode, being unable to adapt to different driving styles and track characteristics, and lacking personalized adaptation, thus realizing multi-scenario and personalized intelligent control. On the other hand, the learning mode allows the same RC model car to adapt to different tracks and driving preferences, reducing the lap time by ≥0.8 seconds, and the parameter fine-tuning does not require manual intervention, achieving an intelligent breakthrough of "personalized adaptation for a single car", far exceeding the adaptation capabilities of existing control systems. 8. This invention is based on a multi-information fusion-based intelligent tail wing control system for RC model cars. It is equipped with a DC-DC step-down module and a high-torque RC servo that are adapted to the power supply requirements of each unit. The DC-DC module outputs three stable voltages at different levels, which drive the high-torque servo through a dedicated servo drive chip. On the one hand, this solves the problems of power supply mismatch, insufficient servo torque, and inability to achieve precise angle driving in existing servo drives, thus achieving smooth and precise driving of the servo motor. On the other hand, this setting improves the response speed of the servo motor by 20%, and there will be no stuttering even under high-frequency angle adjustments. At the same time, it avoids damage to the servo motor caused by power supply mismatch and extends the service life of the system. Attached Figure Description

[0015] Figure 1 This diagram illustrates the connection relationship between the rotating shaft actuator and the hood in the intelligent rear wing control system for RC model cars based on multi-information fusion, as described in this invention. Figure 2 This is a schematic diagram of the rotating shaft actuator of the intelligent rear wing control system for RC model cars based on multi-information fusion according to the present invention. Figure 1 ; Figure 3 This is a schematic diagram of the rotating shaft actuator of the intelligent rear wing control system for RC model cars based on multi-information fusion according to the present invention. Figure 2 ; Figure 4 This is a schematic diagram of the sliding block and support seat of the intelligent rear wing control system for RC model cars based on multi-information fusion according to the present invention.

[0016] In the diagram: 1. Tail wing body; 2. Servo linkage; 3. Rotary shaft; 4. Sliding block; 5. Support base; 501. Slide rail; 6. RC servo; 7. Servo mounting base; 8. Car cover. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solution of the present invention, the product of the present invention will be further described in detail below with reference to embodiments and accompanying drawings.

[0018] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element; when an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation. Example

[0019] The intelligent rear wing control system for RC model cars based on multi-information fusion includes a signal sensing unit, a core processing unit, a servo drive unit, a rotating shaft actuator, a rear wing body, a power management unit, and a feedback unit. Each unit is connected through a clear electrical interface and cooperates with the mechanical structure to form a complete closed-loop intelligent control system, ensuring that there are no breaks in the entire process of signal acquisition, processing, decision-making, execution, and feedback.

[0020] The signal sensing unit synchronously captures key vehicle operation information through three types of sub-modules, converts vehicle speed, attitude, operation commands, etc. into electrical signals, and outputs them to the core processing unit after standardized processing to provide raw data support for control decisions. The core processing unit receives digital or analog signals from the signal sensing unit through a preset interface, and generates precise PWM control commands after filtering, fusion, and algorithm inference, thereby realizing the core logic conversion of "perception-decision". The servo drive unit is used to receive the weak PWM command from the core processing unit, amplify it into a strong electrical signal through the drive chip, and drive the servo motor to generate a precise mechanical rotation angle that meets the command requirements, thus completing the conversion of "electrical signal - mechanical motion". The rotating shaft actuator converts the rotational motion of the servo motor into the angle adjustment motion of the tail fin body through rigid connection and sliding cooperation, ensuring the accuracy and response speed of mechanical transmission; The tail fin body changes its contact attitude with the airflow according to the adjusted angle, generating aerodynamic downforce adapted to the current operating conditions, balancing driving resistance and grip. The power management unit provides each unit with a matching voltage and stable current, while also preventing circuit damage through multiple protection mechanisms to ensure continuous and reliable system operation. The feedback unit collects the actual angle of the tail fin in real time and feeds it back to the core processing unit, forming a closed-loop control of "command-execution-feedback-correction" to eliminate angle deviation.

[0021] The signal sensing unit includes a speed detection submodule, an attitude detection submodule, and a driving intention detection submodule. The data from these three submodules complement and verify each other, ensuring the comprehensiveness and reliability of the collected information. The speed detection submodule consists of a Hall sensor, a magnet, and a signal conditioning circuit. The Hall sensor is model A3144, with a response time ≤1μs, an operating voltage range of 4.5-24V, and is compatible with a 5V system power supply. It has strong anti-electromagnetic interference capabilities and can effectively resist electromagnetic radiation from the RC model car motor. The magnet is made of neodymium iron boron, with a diameter of 5mm and a thickness of 2mm. This material has a remanence strength ≥1200mT, good stability, and is not easily demagnetized over long-term use, ensuring the consistency of the detected signal. The signal conditioning circuit includes an LM324 operational amplifier, whose low power consumption and wide voltage operating characteristics are suitable for the power supply environment of the model car, and can achieve pulse signal shaping, filtering, and amplitude amplification. The magnet is attached to the stress-stabilizing area of ​​the rear wheel hub of the RC model car using high-temperature resistant strong adhesive, avoiding the wheel hub bolt holes and tire mounting positions. The Hall sensor is fixed to the corresponding frame position, ensuring that the distance between the Hall sensor and the magnet is controlled at 2-5mm. This distance ensures that the magnet can effectively trigger the magnetoelectric conversion of the Hall sensor when it rotates, while avoiding collisions caused by wheel vibration. The input terminal of the signal conditioning circuit is soldered to the signal pin of the Hall sensor, and the output terminal is connected to the digital input pin of the core processing unit, such as the PA0 pin of the STM32G474RET6, via a connecting wire. The power supply terminal is connected to the 5V output interface of the power management unit. In this embodiment, when the wheel rotates, the magnet rotates synchronously with the hub. Each time it passes the Hall sensor, the sensor outputs a pulse signal with alternating high and low levels. The signal conditioning circuit shapes the pulse signal: irregular pulses are corrected into square waves. Filtering: after filtering out noise generated by motor interference through an RC filter network, the signal is transmitted to the core processing unit. The core processing unit calculates the real-time vehicle speed by counting the number of pulses per unit time and combining parameters such as wheel circumference and hub teeth. The calculation formula is: V = (number of pulses × wheel circumference) / (sampling time × number of hub teeth). For example, if the wheel diameter is 60mm, the circumference is 188.4mm, the number of hub teeth is 20, and the sampling time is 10ms, and the number of pulses is 10, then the vehicle speed V = (10 × 0.1884m) / (0.01s × 20) = 9.42m / s = 33.91km / h.

[0022] The attitude detection submodule uses a six-axis IMU module, model MPU6050, which integrates a three-axis accelerometer and a three-axis gyroscope. It is specifically designed for dynamic attitude detection of RC model cars. The MPU6050's three-axis acceleration measurement range is ±16g, covering the maximum acceleration of RC model cars during rapid acceleration and braking. The limit acceleration of ordinary model cars does not exceed 8g. The three-axis angular velocity measurement range is ±2000° / s, which is suitable for rapid attitude changes of model cars when cornering and pitching up. The data update rate is ≥200Hz, and a set of data is output every 5ms, which meets the synchronous acquisition requirements of a 10ms control cycle. The module is small in size and low in power consumption, which is suitable for the limited installation space and power supply budget of model cars. The IMU module is fixed to the center of gravity of the RC model car via a shock-absorbing pad, typically located below the center of the vehicle body, avoiding vibration sources such as motors and servos. The shock-absorbing pad absorbs vibrations during vehicle movement, preventing attitude data distortion caused by vibration. The module's SDA pin is connected to the PB7 pin of the core processing unit via an I2C bus, the SCL pin is connected to the PB6 pin, the VCC pin is connected to the 3.3V output interface of the power management unit, and the GND pin is grounded with the system. In this embodiment, the accelerometer inside the module detects the vehicle's acceleration in real time along the X (longitudinal), Y (lateral), and Z (vertical) axes, and the gyroscope detects the vehicle's angular velocity around these three axes. The core processing unit reads this raw data via the I2C bus and, in conjunction with the module's built-in DMP (Digital Motion Processor), performs attitude calculations to obtain key parameters such as the vehicle's pitch angle, roll angle, pitch rate, and longitudinal acceleration. This allows the unit to determine whether the vehicle is in an unstable state, such as "nose-up" (positive longitudinal acceleration and positive pitch angle) or "tail-up" (excessive lateral acceleration and abnormal roll angle).

[0023] The driving intent detection submodule includes a signal parsing circuit. Its core design goal is to be compatible with mainstream remote control protocols and accurately capture driving operation commands. The signal parsing circuit supports standard PWM, SBUS2, and FPort protocols, making it compatible with mainstream RC remote control receivers from brands such as Futaba, Spektrum, and FrSky, offering strong versatility. It uses a UART interface for communication, supporting baud rates of 9600 / 19200 / 115200bps, enabling rapid transmission of throttle, steering, and braking signals. The signal parsing circuit connects to the signal output of the RC model car remote control receiver via a connecting cable. It connects to the PA2 pin of the core processing unit via the TX pin of the UART interface, and the RX pin to the PA3 pin. The power supply is connected to the 5V output interface of the power management unit. Braking signals are acquired through two methods. The implementation involves two aspects: First, parsing reverse throttle commands: when the throttle channel PWM signal pulse width is ≤1000μs, it is determined to be a braking intention. Second, parsing independent braking channels: if the receiver supports an independent braking channel, when the PWM signal pulse width of that channel is ≥2000μs, it is determined to be a braking intention. In this embodiment, the remote control receiver converts the driver's operations, such as pushing the accelerator, turning the steering wheel, and pressing the brake, into PWM signals of the corresponding channels. The signal parsing circuit decodes PWM signals of different protocols, such as converting the serial signals of SBUS2 and FPort into parallel digital signals that can be recognized by the core processing unit, extracting information such as throttle opening (e.g., 1000-2000μs corresponding to 0-100% throttle), steering angle (e.g., 1000-2000μs corresponding to full left-full-right steering), and braking status, and transmitting this information to the core processing unit as a feedforward input for driving intention.

[0024] The core processing unit is a high-performance microcontroller, model STM32G474RET6, supplemented by a signal preprocessing module, an intelligent decision-making algorithm module, a control parameter memory, and a PWM signal generator, forming the core computing and control hub of the system. The STM32G474RET6 adopts an ARM Cortex-M4 core with a main frequency of up to 170MHz and a computing performance of 213DMIPS, which can meet the complex computing requirements of multi-source data fusion, Kalman filtering, fuzzy control algorithms, etc. It has rich built-in peripherals, including a 12-bit ADC acquisition unit, two I2C buses, three UART interfaces, and eight PWM output units. The 12-bit ADC acquisition unit adapts to the potentiometer signals of the feedback unit, the two I2C buses are used to connect to the IMU module, the three UART interfaces connect the driver intention detection submodule to the host computer, and the eight PWM output units meet the servo drive control requirements, realizing the system functions without the need for additional expansion chips. The on-chip Flash capacity is 512KB, and the RAM capacity is 128KB. In this embodiment, the microcontroller is fixed to the PCB mounting bracket of the vehicle frame with four M3 screws. The PCB board has reserved interface pads for each module. The signal input terminals are respectively connected to the output pins of the signal sensing unit, with the speed detection submodule connected to PA0, the IMU module to PB6 / PB7, the driving intention detection submodule to PA2 / PA3, and the signal output pin of the feedback unit. The signal output terminal is connected to the control input terminal of the servo drive unit through the PB0 pin. The power supply terminal is connected to the 3.3V output interface of the power management unit, and the power supply is further stabilized by the onboard LDO voltage regulator circuit.

[0025] The signal preprocessing module incorporates a Kalman filter algorithm, the core function of which is to remove interference noise from the original data and improve data reliability.

[0026] In this embodiment, the Kalman filter algorithm processes data through a "prediction-update" iterative process: first, it predicts the current state based on the state vector of the previous moment (prediction stage); then, it corrects the predicted value by combining the current sensor observation data (update stage); finally, it outputs the filtered and accurate state vector. This algorithm can effectively eliminate interference factors such as wheel slippage (causing distortion of vehicle speed data) and vibration (causing fluctuations in attitude data). For example, when the model vehicle's wheels slip, the vehicle speed output by the speed detection submodule is too high. The Kalman filter determines that the actual vehicle speed should be lower than the detected value by fusing the longitudinal acceleration data of the IMU module, thereby correcting the vehicle speed data and ensuring the reliability of the state vector.

[0027] The intelligent decision-making algorithm module stores and runs an adaptive fuzzy control algorithm to simulate expert driving experience and achieve intelligent decision-making. The input variables of this algorithm are vehicle speed V, longitudinal acceleration A_x, pitch rate ω_y, and throttle opening T. All variables are divided into seven linguistic variables using triangular membership functions: "Negative Large (NB), Negative Medium (NM), Negative Small (NS), Zero (Z), Positive Small (PS), Positive Medium (PM), and Positive Large (PB)," covering the full range of each variable. For example, NB for vehicle speed V corresponds to 0-30 km / h, and PB corresponds to 100- 130km / h; the output variable is the tail wing angle increment Δα, divided into 5 linguistic variables: "negative large (NB), negative small (NS), zero (Z), positive small (PS), positive large (PB)," corresponding to an angle increment range of -5° to +5°; defuzzification uses the centroid method, obtaining the precise angle increment value by calculating the centroid coordinates of the fuzzy set; in this embodiment, the core processing unit converts the filtered vehicle state vector (V, A_x, ω_y, T) into fuzzy linguistic variables, calls 35 fuzzy rule bases in the control parameter memory, such as "IF V=PB AND Ax=PB AND ω_y=PB THEN Δα=PB" and "IF V=PM AND A_x=NB AND brake signal=PS THEN Δα=NS," and performs inference operations; each rule calculates the activation intensity based on the membership degree of the input variable, and after weighted averaging, obtains the fuzzy output set, then defuzzifies it using the centroid method to obtain the precise angle increment Δα, adds it to the current tail wing angle, and after amplitude limiting processing, obtains the new target angle.

[0028] The control parameter memory uses the microcontroller's built-in on-chip Flash to store various parameters required for control. The control parameter memory includes a speed-angle reference mapping table (V1=60km / h corresponds to α1=15°, V2=90km / h corresponds to α2=30°, V3=120km / h corresponds to α3=45°), a 35-rule fuzzy rule base (stored in array form, each rule occupying 4 bytes, recording the correspondence between the linguistic variable combinations of input variables and the output variables), and user-defined mode parameters (speed thresholds, angle ranges, etc., adjusted by the user via a host computer). In this embodiment, parameter writing adopts a sector erase + byte write method to avoid overwriting critical program code; during reading, it is quickly called through address index, with a reading speed of ≤1μs / byte, which meets the parameter calling requirements of real-time control; users can connect to the host computer via USB interface or Bluetooth module to modify custom parameters, and the modifications are automatically saved to Flash and will not be lost when power is off.

[0029] The PWM signal generator is implemented using the timer peripheral of the STM32G474RET6. Its core function is to output a PWM signal with a precise pulse width. The frequency of the output PWM signal is 50Hz, and the pulse width range is 500-2500μs, which has a linear mapping relationship with the tail fin angle. 500μs corresponds to 0°, and 2500μs corresponds to 48°. The mapping formula is: pulse width = 500 + (target angle / 48) × 2000 μs. In this embodiment, the core processing unit calculates the corresponding pulse width value based on the target angle obtained by intelligent decision-making, through the PWM generation mode of the timer, configures the timer's automatic reload register (ARR) and compare register (CCR), and generates a PWM signal with a specified pulse width. The signal is output to the servo drive unit through the PB0 pin of the microcontroller to provide instructions for the angle control of the servo motor.

[0030] The servo drive unit includes a dedicated servo drive chip and an external filtering circuit. Its core function is to convert the low-voltage commands from the core processing unit into high-voltage drive signals for the servo motor. The dedicated servo drive chip is model TB6612, which supports dual-channel motor drive and can continuously output current up to 1.2A. It is compatible with the RC servo unit (MG996R) and requires a static current ≤500mA and a dynamic current ≤1.5A. The chip has built-in overcurrent and overheat protection functions to prevent circuit damage caused by motor stall or overload. The external filtering circuit consists of an RC filter network composed of a 1kΩ resistor and a 104μF capacitor to filter out high-frequency noise in the PWM signal, ensuring stable drive signal. The RC servo unit is model MG996R, with an output shaft rotation range of 0°-180°, an output torque ≥13kg·cm, and a response speed ≤0.15s / 60°. The PWM output pin (PB0) of the core processing unit is connected to the IN1 pin of the TB6612 chip. The VCC pin of the TB6612 chip is connected to the 6V output interface of the power management unit, and the GND pin is grounded with the system. The OUT1 and OUT2 pins of the chip are connected to the signal input and power input terminals of the MG996R servo, respectively, and the GND pin of the servo is grounded with the system. The servo is fixed to the rear of the frame via a servo mounting bracket, which is connected to the frame with M3 screws to ensure that the servo does not move during operation. In this embodiment, the PWM signal output by the core processing unit is transmitted to the IN1 pin of the TB6612 chip. The chip identifies the target rotation angle command based on the PWM pulse width and controls the direction and magnitude of the output current through the internal H-bridge circuit to drive the DC motor inside the servo motor to rotate. The potentiometer integrated on the servo motor shaft detects the actual rotation angle in real time and feeds the rotation angle signal back to the control circuit inside the servo motor for comparison with the target rotation angle until the motor rotates to the target position, thereby achieving precise rotation angle control of the servo motor and providing stable power input to the shaft actuator.

[0031] like Figures 1 to 4 As shown, the rotating shaft actuator includes a servo arm 2, a rotating shaft 3, a sliding block 4, and a support base 5. One end of the servo arm 2 is fixed to the output end of an RC servo unit 6. The RC servo unit 6 is fixedly mounted on a servo mounting base 7. The servo mounting base 7 is fixed to the inner bottom surface of the upper cover 8 corresponding to the rear of the vehicle. The other end of the servo arm 2 is fixedly connected to the middle of the rotating shaft 3. Both ends of the rotating shaft 3 are inserted into the sliding block 4. One end of the sliding block 4 is slidably inserted into the support base 5. The other end of the sliding block 4 is fixedly connected to the tail wing body 1. The lower end of the support base 5 is fixed to the upper surface of the upper cover 8 corresponding to the rear of the vehicle. In this embodiment, the sliding block 4 is horizontally L-shaped, and there are two sliding blocks 4. There are also two support seats 5. The interior of the support seat 5 is hollow to form a slide rail 501, and its upper end is open. The upper part of the support seat 5 is arc-shaped, so that the slide rail 501 inside it is also arc-shaped. One end of the sliding block 4 can be slidably inserted into the slide rail 501 of the support seat 5. When the RC servo 6 drives one end of the servo arm 2 to rotate, the other end of the servo arm 2 can drive the sliding block 4 to move up and down in the slide rail 501 in the support seat 5 through the rotating shaft 3. Since the upper part of the support seat 5 is arc-shaped, the sliding block 4 will slide along the arc-shaped guide of the slide rail 501 while moving up and down in the slide rail 501, thereby driving the tail fin body 1 to move up and down and also changing different pitch angles.

[0032] The power management unit includes a 7.4V / 3000mAh lithium battery pack, a DC-DC step-down module, and a power protection circuit, providing a stable and safe power supply to the system. The lithium battery pack uses a 2S... The battery is a 7.4V / 3000mAh lithium polymer battery with a built-in XT60 connector for easy insertion and removal. The DC-DC step-down module is model LM2596S, with a conversion efficiency ≥85% and output voltage accuracy ±2%. It can stably output three voltages: 3.3V, 5V, and 6V. At 3.3V, it can output a current ≥1A, compatible with core processing units and IMU modules; at 5V, it can output a current ≥2A, compatible with Hall sensors, driver intention detection submodules, and feedback units; and at 6V, it can output a current ≥3A, compatible with servo drive units and servos. The power protection circuit includes overcurrent protection, overvoltage protection, reverse connection protection, and filtering circuitry. The positive terminal of the lithium battery pack is connected to the input of the power protection circuit via the XT60 connector, and the output of the power protection circuit is connected to the input of the LM2596S step-down module. The 3.3V output of the LM2596S module... The power supply is connected to the core processing unit and IMU module via DuPont wires; the 5V output is connected to the power supply interface of the Hall sensor, driving intention detection submodule, and feedback unit; the 6V output is connected to the power supply interface of the TB6612 driver chip and MG996R servo; the GND pins of all units share a common ground with the GND pin of the lithium battery pack, forming a complete power supply loop; in this embodiment, the 7.4V DC output from the lithium battery pack is processed by the power protection circuit to remove noise and voltage spikes in the circuit, while also providing overcurrent, overvoltage, and reverse connection protection; the LM2596S step-down module uses its internal PWM adjustment circuit to step down the 7.4V voltage to 3.3V, 5V, and 6V respectively, outputting a stable DC voltage; the filter circuit further filters out the voltage ripple generated during the step-down process, ensuring that each unit receives a stable and clean power supply, and avoiding abnormal module operation or damage caused by voltage fluctuations.

[0033] The feedback unit of the closed-loop intelligent control system is a miniature conductive plastic potentiometer, model WDD35D-4, which is the core guarantee for achieving precise angle control. The WDD35D-4 potentiometer has an accuracy of 0.1°, a service life of ≥50,000 cycles, an output resistance range of 10kΩ, and an operating voltage of 5V. The potentiometer is small in size and can be embedded in the tail fin's pivot without affecting the overall structure. The potentiometer is mounted on the pivot of the tail fin body, coaxially fixed to the pivot, and locked with a set screw to ensure that the potentiometer shaft moves and rotates synchronously when the tail fin rotates. The potentiometer's VCC pin is connected to the 5V output interface of the power management unit, the GND pin is grounded with the system, and the signal output terminal is connected to the ADC acquisition pin of the core processing unit. In this embodiment, when the tail fin moves and rotates, the potentiometer shaft rotates synchronously with the rotating shaft, causing a linear change in the output resistance of the potentiometer, which is then converted into an analog voltage signal of 0-5V. The core processing unit acquires this voltage signal through a built-in 12-bit ADC and calculates the actual angle of the tail fin using the following formula: Actual angle = (acquired voltage / 5V) × 48°. The core processing unit compares the actual angle with the target angle. If the absolute value of the deviation is > 0.5°, the PWM signal pulse width is adjusted. If the deviation is positive, the pulse width is increased; if the deviation is negative, the pulse width is decreased. The tail fin angle is corrected through the servo drive unit and the rotating shaft actuator until the absolute value of the deviation is ≤ 0.5°, thus achieving full closed-loop control and ensuring that the tail fin angle accurately tracks the target value.

[0034] The core processing unit's intelligent decision-making supports a basic mode and an optimal mode, which users can switch between via remote control commands to adapt to different usage scenarios. (1) Basic mode The system employs a speed-angle mapping method with hysteresis, with the core objective of avoiding tail wing oscillations near the threshold and improving control stability. The core processing unit acquires filtered vehicle speed data in real time and compares it with a preset threshold. When the vehicle speed exceeds 90 km / h for three consecutive control cycles, it is determined to be a high-speed condition, and the tail wing body angle is raised to the corresponding speed reference angle, such as 35° for 100 km / h. When the vehicle speed is below 85 km / h for three consecutive control cycles, it is determined to be a low-speed condition, and the tail wing body angle is lowered to the corresponding reference angle, such as 25° for 80 km / h. The hysteresis difference ΔV = 5 km / h is set based on the speed response characteristics of RC model cars. Vehicle speed fluctuations are usually ≤3 km / h, and a difference of 5 km / h can effectively avoid oscillations without affecting control timeliness. It is suitable for ordinary recreational scenarios or simple track conditions, mainly straight roads, with simple control logic, fast response, and no need for complex algorithm calculations.

[0035] (2) Preferred mode The adaptive fuzzy control algorithm is initiated, with the core objective of handling complex operating conditions and improving dynamic stability and racing performance. The core processing unit takes the filtered vehicle state vector, vehicle speed V, longitudinal acceleration A_x, pitch rate ω_y, and throttle opening T as input, and calls 35 fuzzy rule bases for inference calculations. For example, when the vehicle speed, longitudinal acceleration, and pitch rate are all positive (PB) (i.e., the vehicle accelerates rapidly at high speed and pitches up), the output angle increment is positive (PB), increasing the rear wing angle in advance to suppress pitching up; when... When the vehicle speed is at a moderate level (PM), the longitudinal acceleration is negatively large (NB), and the braking signal is positively small (PS) (i.e., the vehicle is braking at a moderate speed), the output angle increment is negatively small (NS), reducing the rear wing angle and lowering the risk of rear wheel lock-up. After inference calculation, the angle increment Δα is obtained by defuzzification, which is added to the current rear wing angle and then processed by 0-48° amplitude limiting to obtain the new target angle. It is suitable for high-performance racing scenarios or complex track conditions with many curves and undulating surfaces. It can adjust the rear wing angle in advance according to the vehicle's dynamic attitude and driving intention, break through the cornering limit, and shorten the lap time.

[0036] The core processing unit supports a learning mode, which can adapt personalized parameters according to the user's driving style and track characteristics. The user activates the learning mode by outputting a specified PWM signal through a custom channel on the remote control. The system automatically records lap times for three consecutive laps, including the timestamp of each lap, vehicle speed, attitude data, rear wing angle, and throttle / brake signals for each control cycle. The system first calculates the average lap time for the three laps, eliminating outliers exceeding ±5% of the average. For example, lap times outside the range of 47.5-52.5 seconds (average lap time 50 seconds) are considered abnormal, possibly due to operational errors or track interference. The minimum of the remaining lap times is taken as the "optimal lap time," and the corresponding vehicle speed-rear wing angle matching parameters are extracted simultaneously. Based on these parameters, the system sets the optimal lap time to 0. The 5° step size is used to fine-tune the speed-angle reference mapping table. The adjustment range does not exceed ±10% of the original parameters. For example, if the original V1=60km / h corresponds to α1=15°, and the optimal angle at the best lap speed of 60km / h is 15.5°, then the fine-tuning is 15.5°. If the optimal angle is 13°, then because it exceeds the lower limit of 13.5° (15°×0.9), it is adjusted to 13.5°. The fine-tuned personalized parameters are automatically saved to the control parameter memory. Users can switch to enable personalized parameters or restore default parameters via remote control commands. The system avoids parameter drift through the amplitude limiting function to ensure control stability. It achieves personalized adaptation for individual vehicles and can optimize the rear wing control parameters according to different users' driving styles and track characteristics to further improve racing performance. Example

[0037] The intelligent rear wing control method for RC model cars based on multi-information fusion is applied to the above-mentioned control system. The specific steps are as follows: Step 1: Synchronous Acquisition of Multi-Source Information After the system is powered on, the core processing unit is configured with a 10ms control cycle via the TIM timer, triggering a synchronous acquisition once per cycle: The speed detection submodule collects the pulse signal of the wheel rotation, which is then shaped and filtered by the signal conditioning circuit before being transmitted to the core processing unit. The IMU module collects triaxial acceleration and triaxial angular velocity data and transmits them to the core processing unit via the I2C bus. The driving intention detection submodule collects PWM signals from the accelerator, steering, and brake channels and transmits them to the core processing unit via the UART interface; The acquisition process uses interrupt priority configuration, with IMU data acquisition having the highest priority, followed by speed signals, and then driving intention signals, to ensure data time synchronization and avoid fusion errors caused by timing deviations.

[0038] Step 2: Data Preprocessing and Fusion The core processing unit performs layered processing on the acquired raw data through a signal preprocessing module: Speed ​​signal: First, it is shaped by a Schmitt trigger (to remove glitches), then filtered by a 5-point moving average (to remove high-frequency noise), and finally converted into real-time vehicle speed by pulse counting; Attitude data: Attitude calculation is performed using the DMP built into the IMU module to obtain parameters such as pitch angle, roll angle, and longitudinal acceleration; Multi-source data fusion: By fusing vehicle speed data with longitudinal acceleration and pitch rate data from the IMU through the Kalman filter algorithm, errors caused by wheel slippage and vibration are eliminated, and accurate vehicle state vectors such as vehicle speed V, longitudinal acceleration A_x, pitch rate ω_y, throttle opening T, and braking status are obtained.

[0039] Step 3: Intelligent Decision Making (Pattern Selection and Target Angle Generation) The core processing unit selects the signal based on the mode of the remote control command, switching between the basic mode and the preferred mode: If the basic mode is selected: a speed-angle mapping with hysteresis is used to generate the target angle of the tail wing based on the comparison between the vehicle speed and the threshold. If the preferred mode is selected: the adaptive fuzzy control algorithm is started, the vehicle state vector is used as input, the fuzzy rule base is called for inference and calculation, the fuzzification is defuzzified to obtain the angle increment Δα, and after being added to the current tail wing angle, the target angle is obtained after 0-48° amplitude limiting processing.

[0040] Step 4: Precise Execution and Closed-Loop Feedback The core processing unit calculates the corresponding PWM pulse width based on the target angle and outputs it to the servo drive unit; The servo drive unit converts the PWM signal into a high-voltage drive signal, driving the MG996R servo to rotate to the target angle; The linkage actuator converts the rotational motion of the servo into an angle adjustment of the tail fin body, causing the tail fin to rotate to a position close to the target. The feedback unit (potentiometer) collects the actual angle signal of the tail fin and transmits it to the core processing unit; The core processing unit compares the target angle with the actual angle. If the deviation is greater than 0.5°, it fine-tunes the PWM pulse width and repeats the above process until the deviation is less than or equal to 0.5°.

[0041] Step 5: Adaptive Learning (Personalized Adaptation) If the user activates the learning mode via remote control, the system will perform the following operations: Record lap times and corresponding parameters for three consecutive laps, such as vehicle speed, attitude, rear wing angle, and driving commands. Abnormal lap times exceeding ±5% of the average lap time were removed; The minimum remaining lap time is taken as the optimal lap time, and the vehicle speed-rear wing angle matching parameters for that lap are extracted. Fine-tune the speed-angle reference mapping table in 0.5° increments, with the adjustment range not exceeding ±10% of the original parameters, and avoid parameter drift through a limiting function; Personalized parameters are saved to the control parameter memory and enabled by default upon the next power-on. Users can switch parameter modes via remote control or host computer.

[0042] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Anyone skilled in the art can readily implement the present invention based on the accompanying drawings and the above description. However, any modifications, alterations, and variations made by those skilled in the art without departing from the scope of the present invention using the disclosed technical content are equivalent embodiments of the present invention. Furthermore, any modifications, alterations, and variations made to the above embodiments based on the essential technology of the present invention are still considered part of the technical solution of the present invention.

Claims

1. An intelligent rear wing control system for RC model cars based on multi-information fusion, characterized in that: It includes a signal sensing unit, a core processing unit, a servo drive unit, a rotating shaft actuator, a tail fin body, and a power management unit; The signal sensing unit is used to collect the vehicle speed information, dynamic posture information and driving intention information of the RC model car, and output the corresponding detection signal to the core processing unit. The core processing unit is used to receive the detection signal output by the signal sensing unit, preprocess, fuse and make intelligent decisions on the signal, generate PWM control signal and output it to the servo drive unit; The servo drive unit is used to receive the PWM control signal output by the core processing unit and drive the servo motor to generate a precise mechanical rotation angle. The rotating shaft actuator is used to convert the mechanical rotation angle of the servo motor into an angle change of the tail fin body, so as to achieve precise adjustment of the tail fin angle; The tail fin body is used to generate aerodynamic downforce adapted to the current working conditions of the RC model car according to the adjusted angle; The power management unit is used to provide stable and isolated power supply to all units of the entire control system, and has power protection functions.

2. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The signal sensing unit includes a speed detection submodule, an attitude detection submodule, and a driving intention detection submodule; The speed detection submodule consists of a Hall sensor, a magnet, and a signal conditioning circuit. The magnet is set in the stable force area of ​​the rear wheel hub, and the Hall sensor is fixed to the corresponding frame. The distance between the Hall sensor and the magnet is 2-5mm. The attitude detection submodule is a six-axis IMU module with a three-axis acceleration measurement range of ±16g, a three-axis angular velocity measurement range of ±2000° / s, and a data update rate of ≥200Hz. The IMU module is installed at the center of gravity of the RC model car and fixed by a shock-absorbing pad. The driving intention detection submodule includes a signal parsing circuit that supports standard PWM, SBUS2, and FPort protocols. It communicates with the RC model car remote control receiver via a UART interface to capture signals from the accelerator, steering, and braking channels. The braking signal is achieved by parsing the reverse accelerator command or the independent braking channel PWM signal.

3. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The core of the core processing unit is a microcontroller, and the core processing unit also includes a signal preprocessing module, an intelligent decision-making algorithm module, a control parameter memory, and a PWM signal generator. The signal preprocessing module incorporates a Kalman filter algorithm. ； The intelligent decision-making algorithm module stores and runs an adaptive fuzzy control algorithm. The input variables of this algorithm include vehicle speed V, longitudinal acceleration A_x, pitch rate ω_y, and throttle opening T. The input variables are divided into seven linguistic variables using a triangular membership function: negative large (NB), negative medium (NM), negative small (NS), zero (Z), positive small (PS), positive medium (PM), and positive large (PB). The output variable is the tail wing angle increment Δα, which is divided into five linguistic variables. The centroid method is used for defuzzification. The control parameter memory stores a speed-angle reference mapping table, 35 fuzzy rule bases and user-defined mode parameters. The speed-angle reference mapping table is as follows: V1=60km / h corresponds to α1=15°, V2=90km / h corresponds to α2=30°, and V3=120km / h corresponds to α3=45°. The PWM signal generator is used to output PWM control signals in the range of 500-2500μs.

4. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The servo drive unit includes a dedicated servo drive chip and an external filtering circuit. The dedicated servo drive chip is used to drive an RC servo motor. The output shaft rotation angle range of the RC servo motor is 0°-180°, and the output torque is ≥13kg·cm. The servo drive unit receives the PWM control signal output by the core processing unit, outputs a large current to the servo motor through the drive chip, and converts the digital command into the precise mechanical rotation angle of the servo motor.

5. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The rotating shaft actuator includes a servo arm, a rotating shaft, a sliding block, and a support base. One end of the servo arm is fixed to the output end of an RC servo. The RC servo is fixedly mounted on a servo mounting base, which is fixed to the inner bottom surface of the upper hood corresponding to the rear of the vehicle. The other end of the servo arm is fixedly connected to the middle of the rotating shaft. Both ends of the rotating shaft are inserted into the sliding block. One end of the sliding block is slidably inserted into the support base, and the other end of the sliding block is fixedly connected to the tail wing body. The lower end of the support base is fixed to the upper surface of the upper hood corresponding to the rear of the vehicle.

6. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The power management unit includes a lithium battery pack, a DC-DC step-down module, and a power protection circuit. The power protection circuit includes overcurrent protection, overvoltage protection, reverse connection protection, and a filtering circuit. The output terminal of the lithium battery pack is connected to the input terminal of the DC-DC step-down module, and the output terminal of the DC-DC module is connected to the power interface of each unit. The protection circuit is connected in series with the power input terminal.

7. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The closed-loop intelligent control system also includes a feedback unit, which is a miniature conductive plastic potentiometer. The potentiometer is installed at the pivot of the tail fin body and is used to collect the actual angle signal of the tail fin and feed it back to the core processing unit. The core processing unit corrects the tail fin angle deviation according to the feedback signal to achieve full closed-loop control.

8. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The intelligent decision-making of the core processing unit includes a basic mode and a preferred mode; The basic mode adopts a speed-angle mapping method with hysteresis. When the speed of the RC model car exceeds 90km / h, the tail wing body angle increases; when the speed is below 85km / h, the tail wing body angle decreases, and the hysteresis difference ΔV=5km / h. The preferred mode initiates an adaptive fuzzy control algorithm, which takes the filtered vehicle state vector as input, calls the fuzzy rule base for inference calculation, and obtains the tail wing angle increment Δα after defuzzification. This increment is then added to the current tail wing angle to obtain the new target angle.

9. The intelligent rear wing control system for RC model cars based on multi-information fusion according to claim 1, characterized in that: The core processing unit also supports a learning mode. In the learning mode, the system records the lap time data for three consecutive laps, removes outliers exceeding ±5% of the average lap time, and uses the minimum remaining lap time as the optimal lap time. Simultaneously, it records the vehicle speed, attitude, and rear wing angle parameters for that lap. Based on the parameters corresponding to the optimal lap time, it fine-tunes the speed-angle reference mapping table in 0.5° increments, with the adjustment range not exceeding ±10% of the original parameters. A limiting function is used to prevent parameter drift.

10. A method for intelligent rear wing control of an RC model car based on multi-information fusion, characterized in that: The method applied to the intelligent rear wing control system for an RC model car based on multi-information fusion as described in any one of claims 1-9 includes the following steps: Step 1: Synchronous acquisition of multi-source information. After the system is powered on, the pulse signal of the speed detection submodule, the attitude data of the IMU module, and the throttle / steering / brake PWM signal of the driving intention submodule are acquired synchronously with a control cycle of 10ms. Step 2: Data preprocessing and fusion. The core processing unit filters the raw data through the signal preprocessing module. The speed signal is shaped and filtered, and then converted into real-time vehicle speed through pulse counting. The IMU data and vehicle speed data are fused through Kalman filtering to eliminate errors caused by wheel slippage and vibration, and obtain an accurate vehicle state vector. Step 3: Intelligent decision-making: Select the basic mode or the preferred mode according to the preset mode to generate the tail fin target angle; Step 4: Precise execution and closed-loop feedback. The core processing unit generates a PWM signal corresponding to the target angle and transmits it to the servo drive unit. The servo motor drives the tail fin body to rotate to the target position through the four-axis mechanism. The feedback unit collects the actual angle signal of the tail fin and feeds it back to the core processing unit. The core processing unit corrects the angle deviation. Step 5: Adaptive learning. If learning mode is enabled, the system will determine the best lap time and fine-tune the parameters to achieve personalized adaptation of the bike.

Images