A rigid connection-based left-right splicing type motorcycle control method and system

Abstract

The application discloses a left-right splicing type motorcycle control method and system based on rigid connection. The scheme proposes a rigid-flexible integrated construction method, and a flexible control system is constructed on a rigid truss. The rigid truss guarantees the structural integrity of the vehicle body and the conduction of force, the Ackermann geometry correction model is used to correct the Ackermann geometry relationship, the sensor data is used to correct the steering angle and driving torque in real time, the dynamic weight model is used to predict the controller to coordinate the steering and torque distribution of the two sides of the wheel, and the steering error, the vertical vibration error and the slip rate error are optimized, so that the multi-objective cooperative control problems such as load transfer, vertical disturbance, differential constraint and steering stability are solved. The application takes into account the overall structural rigidity and the control system flexibility, can effectively suppress the slip and vibration caused by the road disturbance, significantly improve the response speed and precision of steering and driving, and ensure the stability and safety under complex road conditions.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of vehicle engineering control, and particularly relates to a control method and system for a left-right spliced motorcycle based on rigid connection. BACKGROUND

[0002] A traditional double-car system realizes coordinated steering through a hinged connection and a mechanical differential, and the structure thereof can expand the equivalent wheelbase to improve stability, but has significant technical defects. The mechanical differential device relies on gear transmission and hydraulic adjustment, and the hinged connection structure has insufficient rigidity, and is prone to structural deformation and steering imbalance under high dynamic conditions, causing structural looseness. Steering coordination relies on a mechanical transmission chain with a fixed transmission ratio, and physical transmission delay makes the system unable to quickly respond to dynamic requirements under complex conditions, and the response time for steering angle adjustment is usually long. Due to the change in the equivalent wheelbase caused by rigid connection, the traditional Ackerman steering model is invalid, and the steering trajectory deviation problem is particularly prominent. Although the existing electronic differential control method partially replaces the mechanical device, it lacks a multi-target coordinated optimization mechanism, and does not fully consider the influence of load transfer and vertical vibration coupling, and cannot simultaneously process the multi-target optimization problem of steering error, vertical vibration error and slip rate error, and the dynamic control precision is insufficient, and it is difficult to meet the real-time requirements of high-speed driving or emergency obstacle avoidance. SUMMARY

[0003] The application aims to provide a control method and system for a left-right spliced motorcycle based on rigid connection, which takes into account the overall structural rigidity and the flexibility of the control system, and can significantly improve the response speed, precision and stability of steering and driving on the basis of effectively suppressing slip and vibration caused by road disturbances.

[0004] In order to achieve the purpose of the application, on the one hand, the application provides a control method for a left-right spliced motorcycle based on rigid connection, comprising the following steps:

[0005] Step 1: collecting basic control parameters through a sensor; and collecting the maximum power of a battery of the motorcycle, and obtaining the wheel radius through the overall vehicle structure;

[0006] Step 2: determining the virtual Ackerman angle and the actual left and right wheel steering angles through an Ackerman geometry correction model for the longitudinal vehicle speed and the steering wheel angular velocity;

[0007] Step 3: determining the left and right wheel slip rates through a slip rate calculation function for the longitudinal vehicle speed and the steering angle;

[0008] Step 4: determining the left and right wheel driving torques through a driving torque distribution function for the left and right wheel slip rates and the left and right wheel vertical forces, and performing normalization adjustment thereon;

[0009] Step 5, the suspension compression speed is determined by a vertical compensation function to determine a vertical vibration compensation term, and a MPC linearization prediction model and a prediction equation thereof are constructed according to the actual left and right wheel steering angles and the left and right wheel torques;

[0010] Step 6, a multi-objective optimization function with physical constraints is constructed according to the MPC linearization prediction model;

[0011] Step 7, a quadratic programming algorithm is used to solve the multi-objective optimization function to obtain the optimal steering angles and torques of the left and right wheels, and the optimal steering angles and torques are sent to the left and right front wheel steering motors and the left and right rear wheel driving motors in the form of control instructions, and a receding horizon optimization strategy is used to optimize and solve the control input of the next prediction sequence.

[0012] On the other hand, the application also provides a left-right splicing type motorcycle control method system based on rigid connection, comprising the following modules:

[0013] A sensor module, wherein the sensors include a steering angle encoder, a wheel speed sensor, an IMU, and a suspension displacement sensor, and are used to collect basic control parameters, and the basic control parameters include steering wheel angle speed, longitudinal vehicle speed, lateral acceleration, left and right side suspension displacement, suspension compression speed, left and right wheel vertical force, vertical force distribution ratio, and road friction coefficient ratio;

[0014] A battery management module, used to collect the maximum power of the battery of the motorcycle;

[0015] An MPC linearization prediction module, used to predict the prediction state sequence in each prediction step range at the sampling point;

[0016] A multi-objective optimization module, used to optimize and solve the output data of the MPC linearization prediction module to generate an optimal solution suitable for the sampling point, and convert the optimal solution into control instructions to send to the left and right front wheel steering motors and the left and right rear wheel driving motors;

[0017] A rigid connection module, comprising a high-rigidity cross-truss structure and a connecting piece connected together by welding, and used to connect the left and right vehicles together to constrain the degrees of freedom of the two vehicles and improve the torsional performance of the spliced vehicle.

[0018] Compared with the prior art, the application has the following advantages: (1) the multi-objective coordinated optimization algorithm of the MPC linearization prediction model solves the multi-objective coordinated optimization problem that cannot be solved by the traditional control method; the MPC linearization prediction model optimizes the tracking performance, coordination and control smoothness simultaneously in the prediction time domain through the receding horizon prediction mechanism, and solves the problems of slow dynamic response and insufficient multi-objective coordination of the traditional control method;

[0019] (2) The application solves the problems of lack of predictive control and insufficient consideration of vertical vibration coupling of the traditional electronic differential scheme by using the dynamic weight adjustment and vertical vibration compensation strategy of the MPC algorithm, adjusting the weight matrix Q, R, S and T in real time, dynamically optimizing the control strategy according to the road conditions and vehicle state, controlling the wheel slip rate within 0.1 while ensuring the steering accuracy, and significantly improving the stability and safety under complex road conditions;

[0020] (3) The application significantly improves the torsional performance of the spliced vehicle by replacing the traditional hinged scheme with a high-rigidity cross-truss structure. The rigid frame formed by welding constrains the degrees of freedom of the two vehicle bodies, solves the problem of steering imbalance caused by the loose structure of the traditional mechanical differential device, and reduces the maintenance cost caused by mechanical wear (such as gear lubrication and wear compensation requirements).

[0021] To more clearly illustrate the functional characteristics and structural parameters of the application, the following further describes the application in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the application, and form a part of the present application. The illustrative embodiments of the application and their descriptions serve to explain the application, and do not constitute an improper limitation on the application. In the drawings:

[0023] Figure 1 is the principle architecture diagram of the application;

[0024] Figure 2 is the control method flowchart of the application;

[0025] Figure 3 is the MPC algorithm trajectory tracking simulation result of the application;

[0026] Figure 4 is the MPC algorithm simulation driver input simulation result of the application;

[0027] Figure 5 is the MPC algorithm control input simulation result of the application;

[0028] Figure 6 is the MPC algorithm vehicle trajectory simulation result of the application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the application will be described clearly and completely below in conjunction with the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the application.

[0030] The control method of the application is specifically designed for the structure of a left-right splicing motorcycle based on rigid connection. The vehicle adopts a rigid truss frame instead of a traditional hinged scheme, and the relative movement freedom of the left and right bodies is constrained by rigid connection, thereby improving the torsional performance of the vehicle. Figure 1 As shown in the structural schematic diagram of the vehicle, L is the wheelbase, and w is the track, is the virtual Ackerman angle, is the actual steering angle of the right front wheel, is the actual steering angle of the left front wheel. On the basis of this structure, the traditional mechanical differential is cancelled, and electronic differential control is adopted to realize flexible coordinated control under the physical constraint of rigid connection by independently controlling the steering angles of the left and right front wheels and the driving torques of the left and right rear wheels. The rigid truss frame ensures the strong coupling characteristics of the movement of the two bodies, provides a clear constraint boundary for the control method, and enables the control method to realize precise force transmission and differential control through an optimization algorithm on the premise of ensuring the stability of the structure.

[0031] The control method of the left-right splicing motorcycle based on rigid connection comprises the following steps: Figure 2 , comprising the following steps:

[0032] Step 1, collecting basic control parameters through sensors, the sensors including a steering angle encoder, a wheel speed sensor, an IMU, and a suspension displacement sensor, the basic control parameters including steering wheel angular velocity, longitudinal vehicle speed, lateral acceleration, left and right side suspension displacement, suspension compression speed, left and right wheel vertical force, vertical force distribution ratio, and road friction coefficient ratio; collecting the maximum power of a battery through a battery management system commonly used in motorcycles, and obtaining wheel radii through the structure of the vehicle;

[0033] Step 1-1, collecting steering wheel angular velocity (driver steering angular velocity, unit: rad / s) through the steering angle encoder to obtain real-time driver steering intention data;

[0034] Step 1-2, collecting longitudinal vehicle speed (unit: m / s) through the wheel speed sensor and the IMU (inertial measurement unit) to obtain a high-precision vehicle speed estimate;

[0035] Step 1-3, collecting lateral acceleration (unit: m / s²) through the IMU to obtain a body roll angle change rate;

[0036] Step 1-4, collecting left and right side suspension displacement , suspension compression speed , and left and right wheel vertical force (unit: N), get real-time vertical force distribution ratio;

[0037] ;

[0038] wherein, is the height of the center of mass (unit: m), is the acceleration of gravity (9.81 m / s²), is the wheelbase (unit: m);

[0039] Step 1-5, get the road friction coefficient ratio As an input parameter for the subsequent control algorithm, assuming that the left and right wheel vertical forces are proportional to the road friction coefficient (i.e. ), the vertical force distribution ratio can be used as an approximation of the road friction coefficient ratio .

[0040] Step 2, determine the virtual Ackerman angle and the actual left and right wheel steering angle through the Ackerman geometry correction model by the longitudinal vehicle speed and steering wheel angular velocity;

[0041] Step 2-1, generate a virtual Ackerman angle through the Ackerman geometry correction model by the longitudinal vehicle speed and the steering wheel angular velocity;

[0042] ;

[0043] wherein, is the wheelbase (unit: m);

[0044] Step 2-2, generate the actual left and right front wheel steering angle through the Ackerman model by the virtual Ackerman angle and the road friction coefficient ratio;

[0045] ;

[0046] wherein, is the speed-sensitive coefficient (default value 0.02), when the adhesion is low , the correction coefficient is reduced to suppress excessive steering.

[0047] Step 3, determine the left and right wheel slip ratio through the slip ratio calculation function by the longitudinal vehicle speed and the steering angle;

[0048] Step 3-1, obtain the actual left and right wheel speed through the wheel speed sensor and IMU data fusion by the longitudinal vehicle speed and the actual left and right front wheel steering angle ;

[0049] Step 3-2, obtain the left and right wheel slip ratio through the slip ratio calculation formula by the actual left and right wheel speed and the longitudinal vehicle speed , for subsequent torque correction;

[0050] ;

[0051] wherein 0.1 m / s is the low-speed working condition threshold.

[0052] Step 4, the left and right wheel slip rates and the left and right wheel vertical forces are determined through a driving torque distribution function to determine left and right wheel driving torques, and the left and right wheel driving torques are normalized and adjusted;

[0053] Step 4-1, the left and right tire vertical forces , the friction coefficient ratio , and the wheel radius are calculated through a torque calculation formula to calculate the maximum friction torque formula :

[0054] ;

[0055] The longitudinal vehicle speed and the maximum battery power are used to calculate the maximum driving torque according to a torque calculation formula:

[0056] ;

[0057] The minimum value of the two is taken as the total driving torque :

[0058] ;

[0059] Step 4-2, the left and right wheel torque distribution is determined according to the vertical force distribution ratio to determine the preliminary left and right wheel driving torques ;

[0060] Step 4-3, the left and right wheel slip rates and the preliminary left and right wheel driving torques are adjusted through a dynamic adjustment algorithm of a slip penalty coefficient (default value 0.5) to correct the torques to generate corrected left and right wheel driving torques , ; as shown in the following formula:

[0061] ;

[0062] Step 4-4, the corrected left and right wheel driving torques are compared with the total driving torque, and if it exceeds, it is normalized and adjusted to ensure that the torque output to the MPC linear prediction model is within the maximum torque range.

[0063] Step 5, the suspension compression velocity is determined by a vertical compensation function to determine a vertical vibration compensation term, and a MPC (Model Predictive Control) linear prediction model and a prediction equation thereof are constructed according to the actual left and right wheel steering angles and the left and right wheel torques;

[0064] Step 5-1, a state vector is defined by using the basic control parameters:

[0065] ;

[0066] wherein, is the position of the system mass center (unit: m), is the heading angle (unit: rad), , is the longitudinal and lateral velocity (unit: m / s), is the yaw rate (unit: rad / s).

[0067] Step 5-2, the suspension compression velocity is determined by a vertical compensation function to generate a vertical vibration compensation term :

[0068] ;

[0069] wherein, is the vertical compensation coefficient (default value 20 Ns / m);

[0070] The compensation term will be included in the MPC optimization process as a disturbance input;

[0071] Step 5-3, the state vector and the vertical vibration compensation term are determined by a linearization function to construct a MPC linear prediction model near the current working point:

[0072] ;

[0073] wherein, is the discretized state transition matrix (6x6 dimension), is the control input matrix (6x4 dimension), is the disturbance input matrix (6x2 dimension), is the control input vector, and are the actual steering angles of the left and right front wheels, and are the corrected left and right wheel driving torques, is the disturbance vector, which includes the vertical vibration compensation term;

[0074] Step 5-4, the MPC linearization prediction model is extended by prediction horizon, and a prediction equation is constructed in the prediction horizon:

[0075] ;

[0076] wherein, is a prediction state sequence, is a control input sequence, wherein is a control horizon length, and the prediction horizon .

[0077] Step 6, a multi-objective optimization function with physical constraints is constructed according to the MPC linearization prediction model;

[0078] Step 6-1, a multi-objective optimization function is constructed based on the prediction equation and the control input vector, and a comprehensive objective function including tracking performance, coordination and control smoothness is established:

[0079] ;

[0080] wherein, is a reference input, is a double-vehicle coordination error, and is a modified left and right wheel driving torque, is a control increment, and a weight matrix , , , is dynamically adjusted according to the control performance requirement, is a prediction horizon, is a prediction step number from the current time, is a current sampling time;

[0081] Step 6-2, physical constraints and safety constraints are added to the multi-objective optimization function by setting physical constraints:

[0082] ;

[0083] wherein, is a maximum steering angle difference, is a maximum total torque, is a maximum slip rate difference, is a maximum side slip angle, is a maximum yaw rate.

[0084] Step 7, the multi-objective optimization function is solved by using a quadratic programming algorithm to obtain the optimal turning angles and torques of the left and right wheels, and the optimal turning angles and torques are sent to the left and right front wheel steering motors and the left and right rear wheel driving motors in the form of control instructions, and a receding horizon optimization strategy is used to optimize and solve the control input of the next prediction sequence;

[0085] Step 7-1, the multi-objective optimization function is solved by using a receding horizon optimization strategy through a sequential quadratic programming (SQP) solver to obtain an optimal control sequence:

[0086] ;

[0087] wherein, is an optimal control sequence containing all control inputs in a prediction horizon, is the multi-objective optimization function, is a prediction state sequence, is a control input sequence, is a constraint function vector, the constraint function vector is the physical constraint and the safety constraint, is a target function gradient vector, and a convergence condition of the solving algorithm is ;

[0088] Step 7-2, the optimal control sequence is extracted as the optimal control input at the current time through a control instruction:

[0089] ;

[0090] wherein, is an optimal control input vector at the current time, is an optimal steering angle of the left front wheel (unit: rad), is an optimal steering angle of the right front wheel (unit: rad), is an optimal driving torque of the left rear wheel (unit: Nm), is an optimal driving torque of the right rear wheel (unit: Nm);

[0091] Step 7-3, the optimal control input at the current time is sent to the left and right front wheel steering motors and the left and right rear wheel driving motors in the form of an instruction at a control frequency of 100 Hz, so as to realize coordinated control of the two vehicles;

[0092] Step 7-4, the basic control parameters are fed back to the MPC linearization prediction model as the actual state at the current time, the working point of the MPC linearization prediction model is updated, initial conditions are provided for optimization and solving of the next control period, and a closed-loop control is formed;

[0093] Step 7-5, the actual state of the current time is adjusted by an adaptive weight adjustment function according to the current driving condition to dynamically adjust the weight matrix in the target function of step 6-1:

[0094] In the emergency obstacle avoidance scene, the tracking weight Q is increased to preferentially ensure the trajectory tracking accuracy;

[0095] In high-speed driving, the coordination weight R is increased to ensure the coordination of the double-vehicle motion;

[0096] On a bumpy road, the control smoothness weight T is increased to reduce the drastic change of the control command.

[0097] The system of the motorcycle control method based on the rigid connection left-right splicing type of the application comprises the following modules:

[0098] The sensor module comprises a steering angle encoder, a wheel speed sensor, an IMU, and a suspension displacement sensor, and is used to collect basic control parameters, wherein the basic control parameters comprise steering wheel rotation speed, longitudinal vehicle speed, lateral acceleration, left and right side suspension displacement, suspension compression speed, left and right wheel vertical force, vertical force distribution ratio, and road friction coefficient ratio;

[0099] The battery management module is used to collect the maximum power of the battery of the motorcycle;

[0100] The MPC linearization prediction module is used to predict the prediction state sequence in each prediction step range at the sampling point;

[0101] The multi-objective optimization module is used to optimize and solve the output data of the MPC linearization prediction module, generate an optimal solution suitable for the sampling point, and convert the optimal solution into a control command to be sent to the left and right front wheel steering motors and the left and right rear wheel driving motors;

[0102] The rigid connection module comprises a high-rigidity cross-truss structure and a connecting piece connected together by welding, and is used to connect the left vehicle and the right vehicle together, constrain the degrees of freedom of the two vehicles, and improve the torsional performance of the spliced vehicle.

[0103] The specific implementation modes of the modules of the system are the same as those of the control method described above, and will not be described here.

[0104] The simulation of the method is performed through the above steps, and the simulation results are as follows:

[0105] As Figure 3As shown, the vehicle trajectory graph has the horizontal axis representing the vehicle's lateral position coordinates and the vertical axis representing its longitudinal position coordinates. This graph details the time history of the trajectory tracking error, including lateral and heading deviations. The red trajectory points represent the desired trajectory drawn based on the driver's input, while the blue trajectory points represent the trajectory of the vehicle model controlled by the MPC control algorithm. The two trajectories almost overlap, verifying the tracking performance and robustness of the MPC control algorithm. The driver input graph has the horizontal axis representing time and the vertical axis representing the normalized throttle opening, brake opening, and steering angle opening, showing the changes in throttle (green line), brake (red line), and steering (blue line) over time. The control input graph has the horizontal axis representing time and the vertical axis representing speed, showing the actual speed (blue line) and steering angle (red line) data input to the control system.

[0106] like Figure 4 As shown in the figure, the horizontal axis represents time, and the vertical axis represents the normalized throttle opening, brake opening, and steering opening, demonstrating the accuracy and response characteristics of the MPC control algorithm in tracking driver inputs (throttle, brake, steering).

[0107] like Figure 5 As shown in the figure, the horizontal axis represents time, the vertical axis represents speed (unit: m / s), and the vertical axis represents steering angle (unit: rad). It shows the changes in control input of the MPC control algorithm under typical operating conditions, including the time history of steering angle and speed. This input is the overall input to the control system after comprehensively considering various inputs from the driver.

[0108] like Figure 6 As shown in the figure, the horizontal axis represents the vehicle's lateral position coordinates, and the vertical axis represents the vehicle's longitudinal position coordinates. This graph illustrates the vehicle's actual driving trajectory under the MPC control algorithm, including the shape and... Figure 3 The expected trajectories almost overlapped, verifying the accuracy and stability of trajectory tracking.

[0109] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0110] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A control method for a right-left split type motorcycle based on rigid connection, characterized by, The method comprises the following steps: Step 1, collecting basic control parameters through a sensor; and collecting maximum power of a battery of the motorcycle, and obtaining wheel radius through a whole vehicle structure; The basic control parameters include steering wheel angular velocity, longitudinal vehicle speed, lateral acceleration, left and right side suspension displacement, suspension compression speed, left and right wheel vertical force, vertical force distribution ratio, and road surface friction coefficient ratio; Step 2, determining virtual Ackerman angle and actual left and right wheel steering angles of the longitudinal vehicle speed and the steering wheel angular velocity through an Ackerman geometry correction model; Step 3, determining left and right wheel slip rates of the longitudinal vehicle speed and the steering angle through a slip rate calculation function; Step 4, determining left and right wheel driving torques through a driving torque distribution function according to the left and right wheel slip rates and the left and right wheel vertical forces, and performing normalization adjustment on the left and right wheel driving torques; Step 4-1, determining total driving torque according to the maximum power of the battery and the road surface friction coefficient ratio; Step 4-2, distributing left and right wheel driving torques according to the vertical force distribution ratio to determine preliminary left and right wheel driving torques; Step 4-3, correcting the torques through a dynamic adjustment algorithm of a slip penalty coefficient according to the left and right wheel slip rates and the preliminary left and right wheel driving torques to generate corrected left and right wheel driving torques; Step 4-4, comparing the corrected left and right wheel driving torques with the total driving torque, and performing normalization adjustment on the corrected left and right wheel driving torques if the corrected left and right wheel driving torques exceed the total driving torque to ensure that the output torque is within the maximum torque range; Step 5, determining a vertical vibration compensation term through a vertical compensation function according to the suspension compression speed, and constructing an MPC linearization prediction model and a prediction equation thereof according to the actual left and right wheel steering angles and the left and right wheel driving torques; Step 6, constructing a multi-objective optimization function with physical constraint conditions according to the MPC linearization prediction model; Step 7, solving the multi-objective optimization function by using a quadratic programming algorithm to obtain optimal angles and torques of the left and right wheels, and issuing the optimal angles and torques to left and right wheel steering motors and left and right rear wheel driving motors in the form of control instructions, and simultaneously optimizing and solving control inputs of a next prediction sequence by using a receding horizon optimization strategy.

2. The control method of a left-right splicing type motorcycle based on rigid connection according to claim 1, characterized in that, The sensor comprises a steering angle encoder, a wheel speed sensor, an IMU, and a suspension displacement sensor; Step 1 comprises the following steps: Step 1-1, collecting steering wheel angular velocity through the steering angle encoder to obtain real-time driver steering intention data; Step 1-2, collecting longitudinal vehicle speed through the wheel speed sensor and the IMU to obtain vehicle speed estimation value; Step 1-3, collecting lateral acceleration through the IMU to obtain body roll angle change rate; Step 1-4, collecting left and right side suspension displacement, suspension compression speed, and left and right wheel vertical force through the suspension displacement sensor and the IMU to obtain real-time vertical force distribution ratio; Step 1-5, obtaining road surface friction coefficient ratio through the real-time vertical force distribution ratio.

3. The control method of a left-right splicing type motorcycle based on rigid connection according to claim 2, characterized in that, Step 2 comprises the following steps: Step 2-1, generating virtual Ackerman angle through the Ackerman geometry correction model according to the longitudinal vehicle speed and the steering wheel angular velocity; Step 2-2, generating actual left and right front wheel steering angles through the Ackerman model according to the virtual Ackerman angle and the road surface friction coefficient ratio.

4. The left-right splicing type motorcycle control method based on rigid connection according to claim 3, characterized in that, The step 3 comprises the following steps: Step 3-1, fusing the longitudinal vehicle speed and the left and right front wheel actual steering angles through wheel speed sensors and IMU data to obtain left and right wheel actual speeds; Step 3-2, calculating the left and right wheel actual speeds and the longitudinal vehicle speed through a slip ratio calculation formula to obtain left and right wheel slip ratios for subsequent torque correction.

5. The left-right splicing type motorcycle control method based on rigid connection according to claim 4, characterized in that, The step 4-1 is specifically: The left and right wheel vertical forces , the road surface friction coefficient ratio , and the wheel radius The maximum friction torque is calculated by a torque calculation formula : ； The longitudinal vehicle speed is used and the maximum battery power The maximum drive torque is calculated according to the torque calculation formula: ； Taking the minimum of the above two as the total drive torque : 。 6. The left-right splicing type motorcycle control method based on rigid connection according to claim 5, characterized in that, The step 5 comprises the following steps: Step 5-1, defining a state vector with the base control parameters : ； wherein, is the system center of mass position, is the heading angle, , is the longitudinal and lateral velocity, is the yaw angular velocity; Step 5-2, generating a vertical vibration compensation term through a vertical compensation function based on the suspension compression speed; Step 5-3, constructing an MPC linearization prediction model near the current working point through a linearization function based on the state vector and the vertical vibration compensation term: The MPC linearization prediction model is specifically as shown in the following formula:

7. The left-right splicing type motorcycle control method based on rigid connection according to claim 6, characterized by, The prediction equation is specifically as shown in the following formula: ； wherein, is a discretized state transition matrix, is a control input matrix, is a disturbance input matrix, is a control input vector, and are the actual steering angles of the left and right front wheels, and are the corrected driving torques of the left and right wheels, is a disturbance vector, which includes the vertical vibration compensation term; The step 6 comprises the following steps: ； wherein, is a sequence of predicted states, is a sequence of control inputs, wherein is a control time horizon length.

8. The left-right splicing type motorcycle control method based on rigid connection according to claim 7, characterized in that, Step 6-1, constructing a multi-objective optimization function based on the prediction equation and the control input vector: Step 6-2, setting physical constraint conditions for the multi-objective optimization function to add physical constraints and safety constraints: ； wherein, is a reference input, is a two-vehicle coordination error, and is a corrected left and right wheel driving torque, is a control increment, a weight matrix , , , is dynamically adjusted according to control performance requirements, is a prediction time domain, is a prediction step number from the current time, is a current sampling time; The step 7 comprises the following steps: ； wherein, is the maximum difference in steering angle, is the maximum total torque, is the maximum difference in slip ratio, is the maximum side slip angle, is the maximum yaw rate.

9. The left-right splicing type motorcycle control method based on rigid connection according to claim 8, characterized in that, Step 7-1, solving an optimal control sequence through a sequential quadratic programming solver based on the multi-objective optimization function by adopting a rolling horizon optimization strategy: Step 7-3, sending the optimal control input at the current time in the form of an instruction to left and right front wheel steering motors and left and right rear wheel driving motors to realize double vehicle coordinated control; ； wherein, is the optimal control sequence containing all control inputs within the prediction horizon, is the multi-objective optimization function, is the predicted state sequence, is the control input sequence, is the constraint function vector, which is the physical constraints and safety constraints, is the objective function gradient vector, and the solution algorithm convergence condition is ; Step 7-2, extracting the optimal control input at the current time instant from the optimal control sequence extracting the optimal control input at the current time instant from the control command ； wherein, is the optimal control input vector at the current time instant, is the optimal steering angle of the left front wheel, is the optimal steering angle of the right front wheel, is the optimal drive torque of the left rear wheel, is the optimal drive torque of the right rear wheel; Step 7-4, feeding back the basic control parameters as the actual state at the current time to the MPC linearization prediction model through the sensor to update the working point of the MPC linearization prediction model, provide initial conditions for optimization solution of the next control period, and form a closed-loop control; Step 7-5, adjusting the weight matrix in the objective function of step 6-1 according to the current driving condition through an adaptive weight adjustment function based on the actual state at the current time. The following modules are included:

10. A system for controlling a right and left splicing type motorcycle based on rigid connection according to any one of claims 1 to 9, characterized in that, A sensor module including the sensor, including a steering angle encoder, a wheel speed sensor, an IMU, and a suspension displacement sensor, for collecting basic control parameters, the basic control parameters including a steering wheel angle speed, a longitudinal vehicle speed, a lateral acceleration, left and right side suspension displacements, a suspension compression speed, left and right wheel vertical forces, a vertical force distribution ratio, and a road surface friction coefficient ratio; A battery management module for collecting the maximum power of the motorcycle battery; An MPC linearization prediction module for predicting a prediction state sequence within each prediction step range at a sampling point; A multi-objective optimization module for optimizing and solving the output data of the MPC linearization prediction module to generate an optimal solution suitable for the sampling point and converting the optimal solution into a control instruction to be sent to left and right front wheel steering motors and left and right rear wheel driving motors; A rigid connection module including a high-rigidity cross-truss structure and a connecting piece connected together through welding, for connecting the left vehicle and the right vehicle together, constraining the degrees of freedom of the two vehicles while improving the torsional performance of the spliced vehicle. ​

Images