A rotor-based self-balancing unmanned bicycle

Abstract

This invention discloses a rotor-based self-balancing autonomous bicycle, comprising a sensor module, a handlebar balance control module, a rear balance control module, a rear motion control module, a classification and processing module, and a power supply device. These modules can be functionally combined into two subsystems: a self-balancing subsystem and an autonomous driving subsystem. The self-balancing subsystem uses sensors to collect bicycle posture data and adjusts the rotor's direction and speed to balance the original tipping torque, thereby achieving self-balancing of the bicycle in various states, especially at low speeds. The autonomous driving subsystem uses machine learning algorithms to perform automatic path planning and adjusts the handlebar rotor speed and the rear rotor blade angle to achieve steering and speed control. The rotor device used in this invention operates based on the structure of a common bicycle, featuring convenient installation and disassembly, a lightweight structure, and strong functional expandability, providing a new approach to solving the problem of self-balancing autonomous driving of bicycles.

Description

Technical Field

[0001] This invention relates to the field of transportation, and more particularly to a rotor-based self-balancing unmanned bicycle. Background Technology

[0002] Bicycles are a common mode of transportation, boasting advantages such as simple structure, convenience, flexibility, low noise, low price, and energy efficiency. They hold an irreplaceable and important position in today's world where energy conservation, emission reduction, and low-carbon travel have become a national consensus. With increasing public attention to intelligent transportation tools and autonomous driving technology, autonomous bicycles have seen initial development based on this concept. Autonomous bicycles hold promise for applications in disaster relief, forestry operations, and specialized logistics transportation, showing great potential.

[0003] Currently, research on autonomous bicycles mainly focuses on two aspects: dynamic modeling and proposing new control algorithms. Most of this research remains at the stage of theoretical discussion and preliminary testing. Existing balancing systems used in motorcycles or electric bicycles are essentially a superposition of a unicycle balancing system (i.e., an inverted pendulum balancing system) and a bipedal balancing system. Applying this method limits the handlebar dexterity and reduces one degree of freedom. Furthermore, most existing attempts at bicycle self-balancing require complex mechanical modifications to the bicycle, resulting in low practicality and flexibility.

[0004] The complex dynamic characteristics of bicycles make multi-state self-balancing a challenging problem, among which the self-balancing problem of bicycles in a stationary state or at low speed has always been a key factor limiting the development of autonomous bicycles. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing a rotor-based self-balancing unmanned bicycle.

[0006] The objective of this invention is achieved through the following technical solution: a rotor-based self-balancing unmanned bicycle, comprising a sensor module, a handlebar balance control module, a rear balance control module, a rear motion control module, a classification and processing module, and a power supply device.

[0007] The sensor module is used to measure bicycle balance variables and motion path variables. The bicycle balance variables include bicycle handlebar deflection angle α, bicycle body deflection angle β, and rear wheel speed n. The motion path variables include target direction and obstacle orientation.

[0008] The handlebar balance control module is installed on the bicycle handlebars and adjusts the handlebar deflection angle α by adjusting the speed of the left and right rotors and the steering.

[0009] The rear balance control module is installed at the rear of the bicycle frame. By adjusting the speed and steering of the rear rotor, it works together with the handlebar balance control module to adjust the bicycle frame deflection angle β.

[0010] The rear motion control module is installed at the rear of the bicycle. By adjusting the tilt angle of the rear rotor blades, the thrust generated by the rear rotor is adjusted without changing the speed of the rear rotor, thereby adjusting the speed n of the rear wheel and controlling the bicycle to move forward at the expected speed.

[0011] The classification processing module is used to process bicycle balance variables and motion path variables, complete real-time path planning and rotor working parameter calculation, and send instructions to the handlebar balance control module, the rear balance control module, and the rear motion control module.

[0012] The power supply module is installed in the middle of the bicycle and is connected to the handlebar rotor, rear suspension, and microcontroller to provide power for operation.

[0013] Furthermore, the handlebar balance control module comprises rotors and their drive mechanisms mounted on both ends of the handlebars; the main body consists of two horizontally rotating rotors, and the combination of different rotation states of the two rotors can generate an upward lift force Fz and a counter-torque Tz; in the unbalanced state, the change in counter-torque Tz is used to balance the torque generated by the tilting angular velocity in the yaw direction of the bicycle, and in the balanced state, the change in Tz is used for steering during movement; simultaneously, the lift torque Mxy generated by the rotor rotation participates in the balancing process of the bicycle's lateral tilting; the equations are as follows:

[0014] Fz = c T1 (w1 2 +w2 2 )

[0015] Tz = c M1 (w1 2 ±w2 2 )

[0016] Mxy=dc T1 (w1 2 -w2 2 )

[0017] Where d is the distance from the rotor shaft at the left or right end of the handlebars to the handlebar centerline, w1 represents the angular velocity of the left rotor, w2 represents the angular velocity of the right rotor, and c T1 c M1 It is a constant.

[0018] Furthermore, the rear balance control module is a rotor and its drive mechanism mounted above the rear wheel of the bicycle; the main body is a rotating rotor perpendicular to the direction of the bicycle body, and the rotation of the rotor generates a forward thrust Fy and a counter-torque Tx; in the unbalanced state, the change in counter-torque Tx is used to balance the torque generated by the tilting angular velocity in the direction of the bicycle's roll angle Roll, and in the balanced state, the thrust Fy is used for the overall drive of the bicycle; the equation relationship is as follows:

[0019] Fy = c T2 w3 2

[0020] Tx = c M2 w3 2

[0021] Where w3 represents the angular velocity of the tail rotor, c T2 c M2 It is a constant.

[0022] Furthermore, the rear motion control module is a rotor blade tilt angle adjustment device. Since the rotational speed of the rear rotor blades affects both the forward speed and the bicycle's balance, the impact of rotor blade speed changes on the bicycle's balance should be reduced after the bicycle is balanced. When stationary and balanced, the forward thrust should be reduced. The rear motion control module can change the magnitude of the forward thrust by changing the rotor blade tilt angle. In a balanced state, it adjusts the bicycle speed, and in an unbalanced state, it assists in balancing the bicycle. When the blade tilt angle is negative, a reverse pull is generated, and the bicycle's front wheel is regarded as a driven wheel, restricting the front wheel's degree of freedom relative to the handlebars, which is beneficial to the overall balance of the bicycle.

[0023] Furthermore, the classification and processing module is a microcontroller. On one hand, based on the bicycle handlebar deflection angle α and bicycle body deflection angle β input from the pose sensor in the sensor module, it uses a PID control algorithm to output the angular velocities w1 (left rotor), w2 (right rotor), and w3 (rear rotor), and sends corresponding commands to the handlebar balance control module or the rear balance control module, which are manifested as the input PWM wave and direction control voltage of the rotor control motor. Based on the rear wheel speed n input from the speed sensor in the sensor module, it uses a PID control algorithm to output the rotor blade tilt angle θ and sends corresponding commands to the rear motion control module. On the other hand, it uses the image information collected by the vision sensor in the sensor module to obtain the target direction and obstacle position, and performs path planning in real time.

[0024] Furthermore, the self-balancing autonomous bicycle includes a self-balancing control unit, which comprises:

[0025] (1) Simulation modeling: Construct the mapping relationship between the variables measured by the sensor module and the control variables of the handlebar balance control module, the rear balance control module and the rear motion control module, so as to realize the simulated self-balancing of the bicycle; the bicycle variables include the handlebar deflection angle α, the bicycle body deflection angle β and their first and second derivatives; the control variables include the handlebar left rotor angular velocity w1, the handlebar left rotor angular velocity w2, the rear rotor angular velocity w3 and their rotation direction;

[0026] (2) Automated modeling: By combining the simulation controller obtained from the simulation model with the correlation between the variables measured by the real sensor module of the bicycle and the control variables of the real control module, the automated balance control of the bicycle is realized.

[0027] (3) Intelligent Modeling: By combining one or more of evolutionary algorithms, machine learning algorithms, or deep learning algorithms with the real controller of the bicycle obtained through automated modeling, the control parameters c of the bicycle in unfamiliar or complex environments are learned. T1 c M1 c T2 c M2 J enables adaptive balancing in unfamiliar or complex environments.

[0028] Furthermore, the self-balancing autonomous bicycle includes an autonomous driving control unit, which comprises: utilizing machine learning algorithms to continuously sense the surrounding environment and predict possible changes, completing three sub-tasks: environmental perception and localization, decision-making and planning, and execution of decisions.

[0029] (1) Environmental perception and positioning: A monocular camera is used to monitor road obstacles ahead, and lidar and ultrasonic radar are used to collect distance data and draw point cloud maps, thereby realizing the depiction of the shape, size and distance of obstacles; satellite positioning is used to obtain the specific coordinates of the vehicle, providing a basis for vehicle decision-making and planning.

[0030] (2) Decision planning: First, the path planning is completed on the map using the Diikstra or RRT algorithm, and the behavior prediction of other road participants and the decision-making of one's own behavior are made using the deep learning LSTM recurrent neural network.

[0031] (3) Execution decision: Select the desired bicycle variables to achieve unmanned driving control of the bicycle.

[0032] Furthermore, the theoretical modeling process for the self-balancing bicycle in the self-balancing control section includes:

[0033] Establish the rotation matrix R between the inertial coordinate system {E} and the bicycle coordinate system {B}. B→EAnd convert the attitude angles to angular velocities p, q, r in the bicycle coordinate system:

[0034]

[0035]

[0036] In the bicycle frame of reference {B}, the three rotors generate a force F0, a torque T on the fuselage axis, and a gyroscopic torque G:

[0037]

[0038]

[0039]

[0040] The bicycle coordinate system is transformed to the inertial coordinate system {E}, and the equilibrium equations are established as follows:

[0041]

[0042]

[0043] Where w1 represents the angular velocity of the left rotor on the handlebars, w2 represents the angular velocity of the right rotor, w3 represents the angular velocity of the rear rotor, d is the distance from the rotor shaft at the left or right end of the handlebars to the center line of the handlebars, and c T1 c T2 c M1 c M2 Let J1 and J2 be constants, representing the total moment of inertia of the handlebar and tail rotor motor rotors and rotors around the vehicle body, respectively, and J represent the moment of inertia matrix of the entire vehicle around the X, Y, and Z axes. This indicates taking the first derivative with respect to β and α. This indicates taking the second derivative with respect to β and α.

[0044] The beneficial effects of this invention are:

[0045] 1. The driverless bicycle of the present invention has a self-balancing function when stationary;

[0046] 2. This invention enables the bicycle to achieve self-balancing function under various motion states, especially at low speeds, while retaining the high degree of freedom of the bicycle handlebars;

[0047] 3. This invention is based entirely on the structural foundation of a common bicycle, with parts added as needed. The rotor device is more flexible and lightweight than traditional flywheels and other methods, and has advantages such as easy installation and detachability.

[0048] 4. Based on this invention, many in-depth and extended studies can be carried out, such as the optimization of autonomous navigation function, which has strong functional expandability. Attached Figure Description

[0049] Figure 1 This is a system module logic diagram of the present invention;

[0050] Figure 2 This is an overall structural diagram of the self-balancing subsystem of the present invention;

[0051] Figure 3 This is a top view of the bicycle handlebar portion of the present invention;

[0052] Figure 4 This is a flowchart of the handlebar balance control process of the present invention;

[0053] Figure 5 This is a schematic diagram of the bicycle rear view of the present invention;

[0054] Figure 6 This is a flowchart of the rear-end balance and motion control process of the present invention;

[0055] Figure 7 This is a schematic diagram of the monocular camera positioning of the present invention;

[0056] Figure 8 This is a schematic diagram illustrating a practical example of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments. The illustrative embodiments and descriptions of this invention are only used to explain this invention and are not intended to limit this invention.

[0058] like Figure 1 As shown, the present invention provides a rotor-based self-balancing unmanned bicycle, including a sensor module, a handlebar balance control module, a rear balance control module, a rear motion control module, a classification processing module, and a power supply device.

[0059] The sensor module is used to measure bicycle balance variables and motion path variables. The bicycle balance variables include bicycle handlebar deflection angle α, bicycle body deflection angle γ, and rear wheel speed n. The motion path variables include target direction and obstacle orientation.

[0060] The handlebar balance control module is installed on the bicycle handlebars and adjusts the handlebar yaw angle α by adjusting the speed and steering of the left and right rotors. The handlebar balance control module consists of rotors and their drive mechanism installed on both ends of the handlebars. The main body consists of two horizontally rotating rotors. The combination of different rotation states of the two rotors can generate an upward lift force Fz and a counter-torque Tz. In the unbalanced state, the change in counter-torque Tz is used to balance the torque generated by the tilting angular velocity in the yaw direction of the bicycle. In the balanced state, the change in Tz is used for steering during movement. At the same time, the lift torque Mxy generated by the rotor rotation participates in the balancing process of the bicycle body tilting laterally. The equation relationship is as follows:

[0061] Fz = c T1 (w1 2 +w2 2 )

[0062] Tz = c M1 (w1 2 ±w2 2 )

[0063] Mxy=dc T1 (w1 2 -w2 2 )

[0064] Where d is the distance from the rotor shaft at the left or right end of the handlebars to the handlebar centerline, w1 represents the angular velocity of the left rotor, w2 represents the angular velocity of the right rotor, and c T1 c M1 It is a constant.

[0065] The rear balance control module is installed at the rear of the bicycle frame. By adjusting the speed and steering of the rear rotor, it works in conjunction with the handlebar balance control module to adjust the bicycle frame deflection angle β. The rear balance control module consists of a rotor and its drive mechanism installed above the rear wheel of the bicycle. The main body is a rotating rotor perpendicular to the bicycle frame direction. The rotor rotation generates a forward thrust Fy and a counter-torque Tx. In the unbalanced state, the change in counter-torque Tx is used to balance the torque generated by the tilt angular velocity in the roll direction of the bicycle. In the balanced state, the thrust Fy is used for the entire vehicle's drive. The equations are as follows:

[0066] Fy = c T2 w3 2

[0067] Tx = c M2 w3 2

[0068] Where w3 represents the angular velocity of the tail rotor, c T2 c M2 It is a constant.

[0069] The rear motion control module is installed at the rear of the bicycle. By adjusting the tilt angle of the rear rotor blades, it regulates the thrust generated by the rear rotor without changing its rotational speed, thereby adjusting the rear wheel speed n and controlling the bicycle to move forward at the desired speed. The rear motion control module is a rotor blade tilt angle adjustment device. Since the rotational speed of the rear rotor blades affects both the forward speed and the bicycle's balance, the impact of changes in rotor blade speed on the bicycle's balance should be reduced after the bicycle is balanced. When the bicycle is in static balance, the forward thrust should be reduced. The rear motion control module can change the magnitude of the forward thrust by changing the rotor blade tilt angle. In a balanced state, it adjusts the bicycle speed, and in an unbalanced state, it assists in balancing the bicycle. When the blade tilt angle is negative, it generates a reverse pull, and the bicycle's front wheel is regarded as a driven wheel, restricting the front wheel's degree of freedom relative to the handlebars, which is beneficial to the overall balance of the bicycle.

[0070] The classification processing module processes bicycle balance and motion path variables, performs real-time path planning and rotor operating parameter calculation, and sends commands to the handlebar balance control module, rear balance control module, and rear motion control module. The classification processing module is a high-performance microcontroller. On one hand, based on the handlebar deflection angle α and bicycle body deflection angle β input from the pose sensor in the sensor module, it uses a PID control algorithm to output the left rotor angular velocity w1, right rotor angular velocity w2, and rear rotor angular velocity w3, sending corresponding commands to the handlebar balance control module or rear balance control module, manifested as the input PWM wave and direction control voltage of the rotor control motor. On the other hand, based on the rear wheel speed n input from the speed sensor in the sensor module, it uses a PID control algorithm to output the rotor blade tilt angle θ, sending corresponding commands to the rear motion control module. Furthermore, it uses image information collected by the vision sensors (camera and laser rangefinder) in the sensor module to obtain the target direction and obstacle location, performing real-time path planning.

[0071] The power supply module is installed in the middle of the bicycle and is connected to the handlebar rotor, rear suspension, and microcontroller to provide power for operation.

[0072] The self-balancing autonomous bicycle further includes a self-balancing control unit, which includes:

[0073] (1) Simulation modeling: Construct the mapping relationship between the variables measured by the sensor module and the control variables of the handlebar balance control module, the rear balance control module and the rear motion control module, so as to realize the simulated self-balancing of the bicycle; the bicycle variables include the handlebar deflection angle α, the bicycle body deflection angle β and their first and second derivatives; the control variables include the handlebar left rotor angular velocity w1, the handlebar left rotor angular velocity w2, the rear rotor angular velocity w3 and their rotation direction;

[0074] (2) Automated modeling: By combining the simulation controller obtained from the simulation model with the correlation between the variables measured by the real sensor module of the bicycle and the control variables of the real control module, the automated balance control of the bicycle is realized.

[0075] (3) Intelligent Modeling: By combining one or more of evolutionary algorithms, machine learning algorithms, or deep learning algorithms with the real controller of the bicycle obtained through automated modeling, the control parameters c of the bicycle in unfamiliar or complex environments are learned. T1 c M1 c T2 c M2 J enables adaptive balancing in unfamiliar or complex environments.

[0076] The theoretical modeling process for the self-balancing bicycle in the self-balancing control section includes:

[0077] Establish the rotation matrix R between the inertial coordinate system {E} and the bicycle coordinate system {B}. B→e And convert the attitude angles to angular velocities p, q, r in the bicycle coordinate system:

[0078]

[0079]

[0080] In the bicycle frame of reference {B}, the three rotors generate a force F0, a torque T on the fuselage axis, and a gyroscopic torque G:

[0081]

[0082]

[0083]

[0084] The bicycle coordinate system is transformed to the inertial coordinate system {E}, and the equilibrium equations are established as follows:

[0085]

[0086]

[0087] Where w1 represents the angular velocity of the left rotor on the handlebars, w2 represents the angular velocity of the right rotor, w3 represents the angular velocity of the rear rotor, d is the distance from the rotor shaft at the left or right end of the handlebars to the center line of the handlebars, and c T1 c T2 c M1 c M2 Let J1 and J2 be constants, representing the total moment of inertia of the handlebar and tail rotor motor rotors and rotors around the vehicle body, respectively, and J represent the moment of inertia matrix of the entire vehicle around the X, Y, and Z axes. This indicates taking the first derivative with respect to β and α. This indicates taking the second derivative with respect to β and α.

[0088] The self-balancing autonomous bicycle also includes an autonomous driving control unit, which comprises: using machine learning algorithms to continuously sense the surrounding environment and predict possible changes, and completing three sub-tasks: environmental perception and localization, decision planning, and decision execution.

[0089] (1) Environmental perception and positioning: A monocular camera is used to monitor road obstacles ahead, and lidar and ultrasonic radar are used to collect distance data and draw point cloud maps, thereby realizing the depiction of the shape, size and distance of obstacles; satellite positioning is used to obtain the specific coordinates of the vehicle, providing a basis for vehicle decision-making and planning.

[0090] (2) Decision planning: First, path planning is completed on a high-precision map using Dijkstra's or RRT algorithm. Then, deep learning LSTM recurrent neural network is used to predict the behavior of other road participants and make decisions on its own behavior.

[0091] (3) Execution decision: Select the desired bicycle variables to achieve unmanned driving control of the bicycle.

[0092] Example 1:

[0093] When the bicycle in this system is stationary and unattended, the following steps are performed in sequence:

[0094] Step 1: As Figure 2 As shown, under the influence of gravity, the bicycle exhibits a rotational tendency in two directions: roll angle (Roll) and yaw angle (Yaw). The sensor detects an increase in the values ​​of α and β.

[0095] Step 2: Under the action of the motion control device, the tilt angle of the rear rotor is reduced to close to 0°, and it hardly generates any forward thrust.

[0096] Step 3: The tail rotor accelerates and rotates in the opposite direction to β, generating a force to straighten the bicycle. PID feedback control of the tail rotor speed is used to ensure control accuracy and stability.

[0097] Step 4: If the handlebars and front wheel are tilted to the right, and α is greater than 0, the left handlebar rotor accelerates counterclockwise, and the right handlebar rotor decelerates clockwise or moves at a constant speed; if α is less than 0, the left handlebar rotor accelerates clockwise, and the right handlebar rotor decelerates counterclockwise or moves at a constant speed. PID feedback control of the handlebar rotor speed is used to ensure control accuracy and stability.

[0098] Step 5: The sensors detect that the absolute values ​​of α and β have decreased to the specified range, the tri-rotor angular acceleration continues to decrease, and the whole vehicle maintains an upright dynamic balance.

[0099] Example 2:

[0100] like Figure 3 -As shown in 6 Figure 3 The meanings of each parameter are shown in Table 1 below. When the bicycle of this invention needs to avoid obstacles while in a balanced forward state, the following steps are performed in sequence:

[0101] Step 1: The sensor detects the direction and distance of the obstacle, such as... Figure 7 As shown, the required parameters are obtained through image processing:

[0102] Read the number of pixels x+w / 2 between the center point of the object rectangle and the vertical center line of the camera;

[0103] Let m be the number of pixels in the X direction of the camera, f be the focal length, d be the distance measured by a single eye, and Δd be the actual offset distance of the human body relative to the center line of the camera.

[0104]

[0105]

[0106] Step 2: If the obstacle is located to the left front, the right handlebar rotor accelerates clockwise and the left handlebar rotor decelerates counterclockwise, causing the handlebars and front wheel to turn to the right; if the obstacle is located to the right front, the left handlebar rotor accelerates counterclockwise and the right handlebar rotor decelerates clockwise, causing the handlebars and front wheel to turn to the left. PID feedback control of the handlebar rotor is used to ensure control accuracy and stability.

[0107] Step 3: The sensor reads the values ​​of α and β in real time. If β increases clockwise as α increases, the rear rotor accelerates counterclockwise; otherwise, the rear rotor decelerates clockwise. PID feedback control is used to control the speed of the three rotors to ensure the accuracy and stability of the control.

[0108] Step 4: When the sensor reports that the obstacle in front has been cleared, the handlebar return command is executed. If α is greater than 0, the left and right handlebar rotors decelerate clockwise (or accelerate counterclockwise). If α is less than 0, the left and right handlebar rotors accelerate clockwise (or decelerate counterclockwise). PID feedback is used to control the handlebar rotor speed to ensure control accuracy and stability.

[0109] Step 5: The sensor detects that the absolute values ​​of α and β have both decreased to the specified range, completing the turning obstacle avoidance process.

[0110] Table 1

[0111]

[0112] Example 3:

[0113] When the bicycle of this invention brakes, decelerates, and stops, the following steps are performed in sequence:

[0114] Step 1: The rear motion control device applies force to the rear rotor blades, causing the rotor blade tilt angle to decrease continuously, the forward thrust to decrease, and the vehicle speed to decrease.

[0115] Step 2: The sensor reads the α and β values ​​in real time, such as Figures 3-6 The control steps shown employ PID feedback control of the tri-rotor speed to ensure control accuracy and stability.

[0116] Step 3: The sensors detect that the absolute values ​​of α, β and n have all decreased to the specified range, completing the parking operation and keeping the vehicle in an upright dynamic balance.

[0117] Example 4:

[0118] like Figure 8 As shown, when the bicycle of this invention starts from point A, passes through points B and D, and arrives at point E, the following steps are performed in sequence:

[0119] Step 1: The user places a trip order, and the main control chip completes the route planning based on the stored map data;

[0120] Step 2: When starting from point A, the rear rotor rotates at a positive tilt angle, generating forward thrust. The left rotor of the handlebars rotates counterclockwise at a constant speed, and the right rotor rotates clockwise at a constant speed. The handlebar tilt angle α is close to 0°.

[0121] Step 3: When reaching point B', the sensor detects that the distance between the turning intersection and the vehicle body is less than the set value, and the expected deflection angle is θ. The right rotor accelerates clockwise, the left rotor decelerates counterclockwise, the handlebars rotate clockwise, and the vehicle body gradually turns to the right.

[0122] Step 4: Read the α value in real time and use PID feedback to control the rotor speed of the handlebars to ensure the accuracy and stability of the control;

[0123] Step 5: When the rear of the vehicle reaches point B, complete the turning operation. The right rotor decelerates clockwise, the left rotor accelerates counterclockwise, the handlebars quickly return to center, and the left and right rotors maintain a constant speed. Continue straight to point D.

[0124] Step 6: When reaching point D, the sensor detects that the distance between the turning intersection and the vehicle body is less than the set value, and the expected deflection angle δ is greater than 90°. The right rotor decelerates clockwise with a large angular acceleration, the left rotor accelerates counterclockwise with a large angular acceleration, the handlebars rotate counterclockwise, and the vehicle body gradually turns to the left.

[0125] Step 7: When the rear of the vehicle passes point D, the right rotor accelerates clockwise, the left rotor decelerates counterclockwise, and the handlebars return to center.

[0126] Step 8: Keep the left and right rotors at a constant speed, and gradually reduce the tilt angle of the rear rotor. Read the n value in real time, and use PID feedback to control the speed of the rear rotor to ensure the accuracy and stability of the control.

[0127] Step 9: The sensor detects that the distance between the target position and the vehicle body is less than the set value, and the rear rotor tilt angle is reversed, so the vehicle brakes quickly.

[0128] Step 10: The vehicle is parked in the designated position, and the left and right rotors maintain a constant speed of rotation. The β value is read in real time, and the speed of the rear rotor is controlled by PID feedback to ensure the accuracy and stability of the control and keep the vehicle upright and balanced.

[0129] Example 5:

[0130] When the bicycle of this invention is on a downhill section of road, the following steps are performed in sequence:

[0131] Step 1: The sensor detects that the rear wheel speed n is too high and exceeds the safe speed. The main control chip sends a command to the rear motion control device.

[0132] Step 2: The rear motion control device reverses the tilt angle of the rear rotor blades, causing the rear rotor to rotate and generate reverse thrust, resulting in a rapid decrease in vehicle speed.

[0133] Step 3: The sensor reads the α and β values ​​in real time, such as Figures 3-6 The control steps shown employ PID feedback control of the three-rotor speed to ensure control accuracy and stability. During straight downhill motion, the absolute values ​​of α and β remain within a small range, and the vehicle maintains dynamic balance.

[0134] Step 4: As the absolute value of the rotational speed n decreases, the anti-tilt angle of the tail rotor blades decreases. PID feedback control is used to control the rotor tilt angle and rotational speed to ensure control accuracy and stability. When the specified range is reached, the blade tilt angle approaches 0°.

[0135] Step 5: After completing the downhill section, the speed n decreases rapidly, and the rear rotor blades return to a positive tilt angle. PID feedback control is used to control the tilt angle and speed of the rear rotor to ensure control accuracy and stability.

[0136] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rotor-based self-balancing unmanned bicycle, characterized in that, It includes a sensor module, a handlebar balance control module, a rear balance control module, a rear motion control module, a classification and processing module, and a power supply device; The sensor module is used to measure bicycle balance variables and motion path variables, the bicycle balance variables including the handlebar deflection angle. Bicycle body deflection angle And the rear wheel speed n, the motion path variables include the target direction and the obstacle orientation; The handlebar balance control module is installed on the bicycle handlebars and adjusts the handlebar deflection angle by adjusting the speed of the left and right rotors and the steering. The adjustment; the handlebar balance control module consists of rotors and their drive mechanism installed on both ends of the handlebars; the main body consists of two horizontally rotating rotors, and the combination of different rotation states of the two rotors can generate upward lift. and counter torque ; The rear balance control module is installed at the rear of the bicycle frame. By adjusting the speed and steering of the rear rotor, it works in conjunction with the handlebar balance control module to control the bicycle's yaw angle. The adjustment; the rear balance control module is a rotor and its drive mechanism installed above the rear wheel of the bicycle; the main body is a rotating rotor perpendicular to the direction of the bicycle body, and the rotation of the rotor generates forward thrust. and counter torque ; The rear motion control module is installed at the rear of the bicycle. By adjusting the tilt angle of the rear rotor blades, the thrust generated by the rear rotor is adjusted without changing the speed of the rear rotor, thereby adjusting the speed n of the rear wheel and controlling the bicycle to move forward at the expected speed. The classification processing module is used to process bicycle balance variables and motion path variables, complete real-time path planning and rotor working parameter calculation, and send instructions to the handlebar balance control module, the rear balance control module, and the rear motion control module. The power supply device is installed in the middle of the bicycle and connected to the handlebar rotor, rear suspension, and microcontroller to provide power for driving.

2. The self-balancing unmanned bicycle based on a rotor according to claim 1, characterized in that, In an unbalanced state, the counter torque Variations are used to balance the bicycle's yaw angle. The torque generated by the tilting angular velocity in the direction of the tilt, in equilibrium state The changes are used for steering during motion; simultaneously, the lift torque generated by the rotor rotation... Participating in the balancing process when the vehicle body tilts to the side; the equations are as follows: Where d is the distance from the rotor shaft at the left or right end of the handlebars to the center line of the handlebars. Indicates the angular velocity of the left rotor on the handlebars. Indicates the angular velocity of the right rotor. , It is a constant.

3. The self-balancing unmanned bicycle based on a rotor according to claim 2, characterized in that, In an unbalanced state, the counter torque Variations are used to balance the roll angle of a bicycle. The torque generated by the tilting angular velocity in the direction, and the thrust in equilibrium. Used for all vehicle drives; the equation is as follows: in, Indicates the angular velocity of the tail rotor. , It is a constant.

4. The self-balancing unmanned bicycle based on a rotor according to claim 1, characterized in that, The rear motion control module is a rotor blade tilt angle adjustment device. Since the rotor blade speed at the rear affects both the forward speed and the bicycle's balance, the impact of rotor blade speed changes on the bicycle's balance should be reduced after the bicycle is balanced. When in static balance, the forward thrust should be reduced. The rear motion control module can change the forward thrust by changing the rotor blade tilt angle. In a balanced state, it adjusts the bicycle speed, and in an unbalanced state, it assists in balancing the bicycle. When the blade tilt angle is negative, it generates a reverse pull force, and the front wheel of the bicycle is regarded as the driven wheel. This restricts the degree of freedom of the front wheel relative to the handlebars, which is beneficial to the balance of the whole vehicle.

5. A rotor-based self-balancing unmanned bicycle according to claim 1, characterized in that, The classification and processing module is a microcontroller. On one hand, it uses the bicycle handlebar deflection angle input from the pose sensor in the sensor module. Bicycle body deflection angle The PID control algorithm is used to output the angular velocity of the left rotor on the handlebars. Angular velocity of the right rotor on the handlebars angular velocity of the tail rotor The system sends corresponding commands to the handlebar balance control module or the rear balance control module, which are manifested as the input PWM wave and direction control voltage of the rotor control motor; based on the rear wheel speed n input by the speed sensor in the sensor module, the system uses a PID control algorithm to output the rotor blade tilt angle. It sends corresponding commands to the rear motion control module; on the other hand, it obtains the target direction and obstacle location based on the image information collected by the vision sensor in the sensor module, and performs path planning in real time.

6. A rotor-based self-balancing unmanned bicycle according to claim 3, characterized in that, The self-balancing autonomous bicycle includes a self-balancing control unit, which comprises: (1) Simulation modeling: Construct the mapping relationship between the variables measured by the sensor module and the control variables of the handlebar balance control module, the rear balance control module, and the rear motion control module, so as to realize the simulated self-balancing of the bicycle; the bicycle balance variables include the handlebar deflection angle. Bicycle body deflection angle and their first and second derivatives; the control variables include the angular velocity of the left rotor of the handlebars. Angular velocity of the left rotor on the handlebars angular velocity of the tail rotor And their directions of rotation; (2) Automated modeling: By combining the simulation controller obtained from the simulation model with the correlation between the variables measured by the real sensor module of the bicycle and the control variables of the real control module, the automated balance control of the bicycle is realized; (3) Intelligent modeling: By combining one or more of evolutionary algorithms, machine learning algorithms, or deep learning algorithms with the real controller of the bicycle obtained by automated modeling, the control parameters of the bicycle in unfamiliar or complex environments are learned. , , , J enables adaptive balancing in unfamiliar or complex environments.

7. A rotor-based self-balancing unmanned bicycle according to claim 1, characterized in that, The self-balancing autonomous bicycle includes an autonomous driving control unit, which comprises: using machine learning algorithms to continuously sense the surrounding environment and predict possible changes, and completing three sub-tasks: environmental perception and localization, decision planning, and decision execution. (1) Environmental perception and positioning: A monocular camera is used to monitor road obstacles ahead, and lidar and ultrasonic radar are used to collect distance data and draw point cloud maps, thereby realizing the depiction of the shape, size and distance of obstacles; satellite positioning is used to obtain the specific coordinates of the vehicle, providing a basis for vehicle decision-making and planning; (2) Decision planning: First, the path planning is completed on the map using Dijkstra or RRT algorithm, and the behavior prediction of other road participants and the decision of its own behavior are made using deep learning LSTM recurrent neural network; (3) Execution decision: Select the desired bicycle variables to achieve unmanned driving control of the bicycle.

8. A rotor-based self-balancing unmanned bicycle according to claim 6, characterized in that, The theoretical modeling process for the self-balancing bicycle in the self-balancing control section includes: Establish the rotation matrix between the inertial coordinate system {E} and the bicycle coordinate system {B}. And convert the attitude angles to angular velocities p, q, r in the bicycle coordinate system: In the bicycle frame of reference {B}, the three rotors generate force. The torque generated on the machine body axis and gyro torque : The bicycle coordinate system is transformed to the inertial coordinate system {E}, and the equilibrium equations are established as follows: in Indicates the angular velocity of the left rotor on the handlebars. Indicates the angular velocity of the right rotor. Indicates the angular velocity of the tail rotor. The distance from the rotor shaft at the left or right end of the handlebars to the handlebar centerline. , , , It is a constant. , These represent the total moment of inertia of the handlebar and tail rotor motor rotors and rotors around the vehicle body, respectively. This represents the moment of inertia matrix of the entire vehicle about its X, Y, and Z axes. , Indicates to , Find the first derivative. , Indicates to , Find the second derivative.

Images