Unmanned bicycle course trajectory tracking method based on internal and external double closed-loop control

Abstract

The invention discloses an unmanned bicycle course trajectory tracking method based on internal and external double closed-loop control. In the method, a nonsingular terminal sliding mode controller is used as an inner ring controller, and a PID (Proportion Integration Differentiation) controller is used as an outer ring controller to form an inner ring controller and an outer ring controller. In the inner ring controller, the actual state value and the expected state value of the lateral inclination angle of the unmanned bicycle are used as errors to design a sliding mode surface, then a control law is designed, and lateral balance is mainly achieved; in the outer ring controller, a PID controller is designed by taking an actual state value and an expected state value of a course angle of the unmanned bicycle as an error, and the PID controller is mainly used for compensating inner ring control and realizing course control; and then the output of the inner and outer ring controllers is combined to act in the unmanned bicycle LPV model. According to the method, in an innovative control architecture, an outer ring PID controller and an inner ring nonsingular terminal sliding mode controller act synergistically, and a precise inner and outer ring regulation and control system is constructed. The outer ring PID controller effectively weakens the disturbance caused by the precision limitation of the dynamic model by virtue of the precise compensation capability of the outer ring PID controller on the error of the inner ring model, remarkably improves the precision performance of the inner ring controller, and accelerates the convergence process of the inner ring controller. Meanwhile, the inner-ring nonsingular terminal sliding mode controller greatly simplifies the parameter adjustment process of the controller due to the excellent dynamic response characteristic of the inner-ring nonsingular terminal sliding mode controller, achieves double leap of regulation and control performance and parameter adjustment efficiency, and has very high engineering significance.

Description

Technical Field

[0001] The invention relates to the technical field of unmanned bicycle control, and in particular to a method for tracking the heading trajectory of an unmanned bicycle without a mechanical auxiliary structure. Background Art

[0002] Unmanned bicycles have the advantages of being lightweight, flexible, environmentally friendly and energy-saving. They can play an important role in short-distance lightweight logistics scenarios and are a potential intelligent transportation tool. In the rapid development of unmanned bicycle technology, heading control, as one of its core technologies, has received widespread attention. However, due to the complex dynamic characteristics of bicycles, such as nonlinearity, time-varying, and multiple-input multiple-output (MIMO), how to achieve autonomous balance, path planning, and heading control has become a key issue in research.

[0003] In the current field of unmanned bicycle control, classical control methods such as PID and LQR control are widely used for early balance and heading control. These methods are simple in structure and easy to implement, but have limited performance in nonlinear dynamic environments. Subsequently, nonlinear control methods such as feedback linearization and sliding mode control were introduced to cope with the complex dynamic characteristics of unmanned bicycles and improve the robustness and stability of the system. In addition, advanced control strategies such as multi-sensor fusion and model predictive control (MPC) further enhance the navigation accuracy and stability of the system in complex environments. Nevertheless, these methods still have shortcomings. Classical control methods are difficult to handle complex nonlinear systems, and have slow response time and poor robustness to the system; nonlinear control and adaptive control rely on accurate models and high computing resources, and their practicality is limited. Although sensor fusion technology improves the accuracy of the system, it increases hardware complexity and power consumption, and faces problems such as data noise and sensor failure. Therefore, in the future, it is necessary to further optimize the control algorithm, improve the system control accuracy, and reduce the difficulty of controller parameter adjustment to a certain extent. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention proposes a method for tracking the heading trajectory of an unmanned bicycle based on an inner and outer loop controller. The purpose is to simplify the controller design through the inner and outer loops, make full use of prior knowledge, reduce the difficulty of parameter adjustment, improve the accuracy of the unmanned bicycle heading control system and accelerate the response capability of the system, so that the unmanned bicycle heading control system without auxiliary structure has stronger adaptability and generalization ability.

[0005] The overall technical solution block diagram of the present invention is as follows Figure 1 As shown, the specific steps are as follows:

[0006] Step 1: Refer to the classic LPV model in the field of unmanned bicycle research and rewrite its model formula into a state space equation.

[0007] Step 2: The reference "Arobust two-stage active disturbance rejection control for the stabilization of a riderless bicycle" describes the lateral inclination angle, handlebar angle, and heading angle of the unmanned bicycle. The corresponding relationship between the lateral inclination angle and the heading angle is expressed through simultaneous equations, and then the expected trajectory of the unmanned bicycle is designed.

[0008] Step 3: According to the principle of non-singular terminal sliding mode control, a nonlinear sliding surface s is designed, and then a non-singular terminal is introduced to eliminate the singular point. Then, according to the sliding surface and the non-singular terminal term, a control law is designed as the inner loop controller of the unmanned bicycle to control the lateral balance control of the unmanned bicycle.

[0009] Step 4: Get the improved relationship between the lateral inclination angle and the heading angle in step 2, and design a heading controller as the outer loop controller of the unmanned bicycle control system according to the principle of PID control.

[0010] Step 5: Combining the inner loop controller of the non-singular terminal sliding mode in step 3 and the outer loop controller of the PID control in step 4, a controller of the unmanned bicycle system is constructed, so that the lateral tilt angle state and the heading angle state can be tracked onto the desired trajectory curve within a limited time;

[0011] Furthermore, the LPV mechanical model in step 1 is:

[0012]

[0013] In the formula in and δ represent the lateral tilt angle of the frame and the handlebar angle, respectively, τ=(0,τ δ ) T is the control torque vector of the system, and τ δ is the control torque of the handlebar.

[0014] Furthermore, the above formula is rewritten into the form of spatial state equation:

[0015]

[0016] In the formula, the state variable is A and B are the state matrix and input matrix of the system respectively, and u(t) is the control input.

[0017] Furthermore, in step 2, the relationship between the lateral tilt angle, the handlebar angle and the heading angle of the unmanned bicycle is introduced as follows:

[0018]

[0019]

[0020] Where g is the acceleration of gravity, h is the height of the integrated center of mass of the unmanned bicycle, ω is the wheelbase of the two wheels, b is the horizontal distance from the integrated center of mass of the unmanned bicycle to the coordinate origin, V m is the constant forward speed during the bicycle's stabilization process,

[0021] When the unmanned bicycle enters a certain steady state, the dynamic changes of the lateral tilt angle and the handlebar angle of the unmanned bicycle are slow and the amplitude is very small, that is, By combining the above formulas, we can obtain the relationship between the lateral inclination angle and the heading angle of the unmanned bicycle, that is:

[0022]

[0023] Furthermore, the lateral inclination angle can be calculated The first-order derivative of the heading angle Function:

[0024]

[0025] Furthermore, a preset trajectory for the unmanned bicycle can be designed The desired lateral tilt angle can be obtained from the above formula:

[0026] In step 3, the nonlinear sliding film surface is:

[0027]

[0028] In the formula, Where β>0, p and q are odd numbers and satisfy 1 <p / q<2。

[0029] The control law of the non-singular terminal sliding film in step 3 is set as:

[0030]

[0031] In the formula is the total interference Estimated value, D>|d t |,d t is the unknown high-order interference term, In the above absolute inequality, is an unknown nonlinear function of the system, and tanh(s) is the hyperbolic tangent function of the sliding surface s.

[0032] In step 4, the input error is calculated by the formula in step 2 and then referred to the PID controller principle design: The controller output is τ:

[0033]

[0034] In step 5, the output u of the inner loop non-singular terminal sliding mode controller and the output τ of the outer loop PID controller act together on the unmanned bicycle control object;

[0035] Beneficial effects of the present invention:

[0036] 1. The inner and outer double ring design of the present invention has a clear structure. The inner ring is the vehicle body lateral balance ring, and the outer ring is the vehicle body heading tracking ring. The physical boundaries are clear, which reduces the coupling degree between variables, is conducive to independent adjustment of control parameters, and is easy for engineering application.

[0037] 2. In the inner-outer-loop control mechanism of the present invention, the outer-loop PID controller can compensate for the inner-loop model deviation of the inner-loop non-singular terminal sliding film controller, thereby reducing the disturbance caused by the accuracy error of the dynamic model of the controlled object and improving the control accuracy of the inner-loop controller.

[0038] 3. In the inner-outer-loop control mechanism of the present invention, the inner-loop non-singular terminal sliding film controller can accelerate the convergence speed of the outer-loop PID controller. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a system control block diagram of an unmanned bicycle heading control method based on inner and outer loop controllers;

[0040] Figure 2 Schematic diagram of modeling the unmanned bicycle LPV model (top view and front view)

[0041] Figure 3 The response curves of the lateral tilt angle, heading angle, and handlebar angle when the inner and outer loop controllers add interference to the example work together;

[0042] Figure 4 The response curves of the lateral tilt angle, heading angle, and handlebar angle when the inner loop controller with interference is added to the example. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objectives, technical solutions and advantages of the present invention more obvious and understandable, the present invention is further described below in conjunction with implementation cases, but the present invention is not limited to the following embodiments.

[0044] Implementation Cases:

[0045] To address the problems of low tracking accuracy and insufficient robustness of the current heading control algorithm of unmanned bicycles without auxiliary mechanisms under the influence of external interference, this implementation case proposes an unmanned bicycle heading trajectory tracking method based on inner and outer loop control.

[0046] The present invention is applied to the problem of heading control of an unmanned bicycle without auxiliary mechanisms. Its purpose is to make the vehicle body walk along a preset sinusoidal curve ("S" type curve) trajectory under the interference of the external environment (the sinusoidal curve can help researchers better understand and predict the dynamic behavior of the system due to its wide applicability). The specific steps are as follows:

[0047] Step 1: The structure diagram of the research object is as follows: Figure 2 As shown, an improved LPV model is established based on the physical parameters of the unmanned bicycle and its components shown in Table 1 and Table 2, and is rewritten into a state space equation:

[0048]

[0049] In the formula, under the condition of no load, it can be calculated by physical parameters:

[0050]

[0051] The v in matrix A is the vehicle speed, which is generally taken as a fixed constant value.

[0052] Table 1 Physical parameters of unmanned bicycle

[0053] parameter symbol Numeric Total mass of vehicle body (kg) <![CDATA[m f ]]> 14.850 Wheelbase between front and rear wheels (m) ω 0.885 Vehicle center of mass height (m) h 0.547 Horizontal distance from the center of mass of the vehicle to the origin of the coordinate system (m) b 0.344 Rear offset of the front wheel contact point (m) c 0.078 Fork inclination angle (0) ɑ 75.000 <![CDATA[Acceleration due to gravity (m / s 2 )]]> g 9.810

[0054] Table 2 Physical parameters of unmanned bicycle components

[0055]

[0056] Step 2: Design the tracking trajectory of the unmanned bicycle according to the corresponding relationship between the lateral tilt angle and the first-order derivative of the heading angle of the unmanned bicycle in a certain steady state:

[0057]

[0058]

[0059] Where V m =2.10m / s, ω=0.885m, g=9.81m / s 2 ,ρ=2.281, It is the preset heading trajectory of the unmanned bicycle.

[0060] Step 3: Design the nonlinear sliding surface function s:

[0061]

[0062] In the formula, p=5, q=3, β=0.2.

[0063] Furthermore, a non-singular terminal sliding mode control law is designed:

[0064]

[0065] Wherein, F=600, D=100, η=1.292.

[0066] Step 4: Design a heading controller as the outer loop controller of the unmanned bicycle control system based on the principle of PID control:

[0067]

[0068] In the formula, P=6, I=2, D=3.

[0069] Step 5: Reference Figure 2 , combining the non-singular terminal synovial inner loop controller described in step 2 and the PID outer loop controller described in step 3 to construct an inner and outer loop controller, so that the lateral inclination angle state and the heading angle state converge to the preset equilibrium point within a limited time.

[0070] In order to verify the robustness of this embodiment, the heading control effect of the inner and outer loop controllers of the present invention on the unmanned bicycle is verified based on the MATLAB simulation environment, taking into account the existence of interference between the external environment and the internal environment.

[0071] Simulation Group 1:

[0072] The inner loop controller part works together with the outer loop controller part. The simulation duration is set to 100s, the step size is 0.005s, and the other settings are as follows:

[0073] (1) Setting the initial state: At the initial moment, the frame inclination angle is 10°, the frame inclination speed is 0, and the handlebar angle and handlebar speed are 0;

[0074] (2) Set simulation conditions: Set vehicle speed V m =2.1m / s, set the desired heading angle to:

[0075] (3) Add external environmental interference: add frame lateral pulse torque interference, handlebar pulse torque interference, random handlebar torque interference, and random lateral torque interference to simulate the interference introduced by environmental changes and terrain changes;

[0076] From the data curve of simulation group 1 in the embodiment (attached Figure 3 ) It can be seen that:

[0077] (1) In general, the lateral tilt angle of the unmanned bicycle is within the range of [-0.150-0.200] rad, and the handlebar angle also oscillates around the straight line y=0, indicating that the unmanned bicycle has achieved lateral balance;

[0078] (2) In general, the error oscillation amplitude between the heading angle of the unmanned bicycle and its expected heading angle is basically within the range of [-0.050, 0.050] rad, and the actual heading angle curve can basically track the preset trajectory, indicating that the unmanned bicycle can track the preset trajectory;

[0079] (3) During the system operation (T=30s and T=40s), the system is subject to random interference set by the simulation. After the interference, the lateral tilt angle, handlebar angle and heading angle of the unmanned bicycle can quickly converge to the budgeted target state, indicating that the unmanned bicycle has strong robustness.

[0080] Simulation Group 2:

[0081] Remove the outer loop control part, and keep the rest of the simulation settings the same as those of simulation group 1;

[0082] From the data curve of simulation group 1 in the embodiment (attached Figure 4 ) It can be seen that:

[0083] (1) In the initial stage, the lateral tilt angle of the unmanned bicycle is within the range of [-0.250-0.200] rad, and the roll angle exceeds 0.200 rad. The unmanned bicycle is at risk of tipping over, and the heading value is larger than the preset heading value. The response speed is slower than that of the simulation group 1, indicating that under the same conditions, compared with the joint control of the inner and outer loops, the lateral balance ability of the unmanned bicycle controlled by the single inner loop is insufficient;

[0084] (2) In general, the error between the heading angle of the unmanned bicycle and its expected heading angle is obviously increasing, and the heading angle state is obviously divergent, indicating that the unmanned bicycle has not achieved the expected trajectory control;

[0085] (3) During the system operation (T=30s and T=40s), the system is subject to random interference set by the simulation. After being disturbed, the lateral tilt angle and handlebar angle of the unmanned bicycle can quickly converge to the budget target state, indicating that the unmanned bicycle has strong robustness.

[0086] In summary, it can be seen from the data curves of simulation group 1 and simulation group 2 that the inner-outer-loop control method provided by the present invention can realize that the outer-loop controller can compensate the inner-loop model deviation of the inner-loop controller, thereby reducing the disturbance caused by the accuracy error of the dynamic model of the controlled object, and can improve the control accuracy of the inner-loop controller, and also accelerate the convergence process of the inner-loop controller.

[0087] It should be noted that although the embodiments of the present invention described above are illustrative, they are not intended to limit the present invention, and therefore the present invention is not limited to the above specific embodiments. Without departing from the principles of the present invention, any other embodiments obtained by those skilled in the art under the guidance of the present invention are deemed to be within the protection of the present invention.

Claims

1. The method for tracking the heading trajectory of an unmanned bicycle based on inner and outer loop control according to claim 1, characterized in that: In step 2, when the unmanned bicycle enters a certain steady state, the dynamic changes of the lateral tilt angle and the handlebar angle of the unmanned bicycle are slow and the amplitude of change is very small, that is, By combining the above formulas, we can obtain the relationship between the lateral inclination angle and the heading angle of the unmanned bicycle, that is: Furthermore, the lateral inclination angle can be calculated The first-order derivative of the heading angle Function: Furthermore, a preset trajectory for the unmanned bicycle can be designed The target lateral inclination angle can be obtained from the above formula:

2. The method for tracking the heading trajectory of an unmanned bicycle based on inner and outer loop control according to claims 1 and 2 is characterized in that: The nonlinear sliding film surface in step 3 is: In the formula, Where β>0, p and q are odd numbers and satisfy 1 <p / q<2。 Furthermore, the control law of the non-singular terminal synovial in step 3 is set to: In the formula is the total interference Estimated value, D>|d t |,d t is the unknown high-order interference term, In the above absolute inequality, is the unknown nonlinear function of the system, tanh(s) is the hyperbolic tangent function about the sliding membrane surface s; In step 4, the input error is calculated by the formula in step 2 and then referred to the PID controller principle design: The controller output is τ:

Images