Integrated control method for active suspension and differential steering of distributed drive unmanned vehicle

Abstract

The application provides an integrated control method for active suspension and differential steering of a distributed drive unmanned vehicle, which realizes trajectory tracking of the unmanned vehicle through differential steering, realizes vehicle body roll during steering through active suspension, improves the speed of the unmanned vehicle during cornering, and improves the steering stability. First, a dynamic model of the unmanned vehicle including differential steering and active suspension is established, a reference model is established to obtain a reference yaw rate and a reference vehicle body roll angle, a single-point preview driver model based on fractional calculus theory is designed to obtain a reference front wheel steering angle required by the reference model, and an H ∞ A robust controller controls the yaw rate and the vehicle body roll angle of the differential steering unmanned vehicle to track the reference values, and differential torques and control forces of left and right active suspensions are obtained. The results show that the integrated control method can make the differential steering unmanned vehicle realize trajectory tracking and vehicle body posture control.

Description

Technical Field

[0001] This invention relates to a method for controlling vehicle posture during steering, and in particular to an integrated control method for active suspension and differential steering of a distributed-drive unmanned vehicle. Background Technology

[0002] CN114859733A discloses a trajectory tracking and attitude control method for a differential steering unmanned vehicle. While achieving trajectory tracking through differential steering, it proposes controlling the vehicle's attitude to improve handling stability. It establishes a dynamic and kinematic model of the differential steering and a roll model for the unmanned vehicle, selecting a linear three-degree-of-freedom vehicle model as a reference model to obtain the ideal roll angle. A model predictive controller is used to control the differential steering model to track the given reference trajectory to obtain the required differential torque and the resulting front wheel steering angle. A sliding mode controller is also designed to control the roll model to track the ideal roll angle and obtain the required roll torque.

[0003] Because it uses a three-degree-of-freedom vehicle model as a reference model and requires the design of a sliding mode controller, the entire control method is somewhat cumbersome and has a slow response time. Furthermore, this control method only obtains the roll moment required for vehicle body roll, without considering how this roll moment is obtained; in other words, this control method does not involve the control of the active suspension. Summary of the Invention

[0004] The purpose of this invention is to provide an integrated control method for distributed drive unmanned vehicles, which achieves trajectory tracking of the unmanned vehicle through differential steering, and improves the cornering speed of the unmanned vehicle by achieving body roll during steering through active suspension, thereby improving its handling stability and preventing vehicle rollover.

[0005] The technical solution adopted in this invention is as follows:

[0006] This integrated control method for a distributed-drive autonomous vehicle features active suspension and differential steering. The distributed autonomous vehicle has hub motors installed in all four wheels, and active suspension is used on both the front and rear axles. The integrated control method includes the following steps: constructing a dynamic model of the front-wheel differential steering system, a dynamic model of the front-wheel differential steering vehicle, and a reference model; obtaining the reference front wheel steering angle δ required for the reference model to track the desired path by using a fractional-order single-point preview driver model. fd This is then fed into the reference model to obtain the reference yaw rate γ. d and reference body camber angle φ d H ∞ The robust controller is based on the reference yaw rate γ d and reference body camber angle φ dThe autonomous vehicle is controlled by differential steering and active suspension to have the same yaw rate and body roll angle as the reference model, and to obtain the differential torque ΔM required for steering and the left and right active suspension control forces f1 and f2 required for body roll, so as to prevent the autonomous vehicle from rolling over when steering.

[0007] The above-described integrated control method uses a fractional-order single-point anti-aiming driver model:

[0008]

[0009] Establish a vehicle coordinate system xoy with the car's center of mass as the origin. The driver is located at the car's center of mass point o, and p(x,y) is the driver's aiming point. p It is the driver's aiming distance, i.e., the distance from point p to the y-axis, where y is the longitudinal speed of the car, and v is the distance from p to the y-axis. y Let y be the lateral speed of the car and the aiming deviation. ε δ represents the deviation between the target point p and the actual driving path of the car. f (t) represents the front wheel steering angle, t represents the current time, and T represents the current time. s K is the time delay. m For the driver's control gain, 'a' is the longitudinal distance from the preview point to the rear edge of the line of sight, and 'b' is the longitudinal distance from the preview point to the front edge of the line of sight. Let f(t) be a fractional calculus operator, where α and α' are the fractional orders, and f(t) is the desired path.

[0010] The aforementioned integrated control method provides the following dynamic model for the front-wheel differential steering system:

[0011]

[0012]

[0013] In the formula, J e b e and τ f Let represent the equivalent moment of inertia, steering damping, and frictional torque of the steering system, respectively; ΔT represents the difference in torques around the kingpin between the two front wheels; and τ represents the torques of the steering system. a k is the total self-aligning torque of the front wheels. f For the front wheel lateral stiffness, α f is the front wheel slip angle, and l is the tire contact patch width.

[0014] The dynamic equation of the front wheel differential steering system using the integrated control method described above is:

[0015]

[0016] In the formula, d1 is the interference term, and R cLet l be the wheel radius, ΔM be the differential torque between the left and right front wheels of the vehicle, and l be the wheel radius. f This is the distance from the center of mass to the front axle.

[0017] The integrated control method described above uses the following vehicle differential steering dynamics model:

[0018]

[0019] F yfl =F yfr =k f α f F yrl =F yrr =k r α r ,

[0020] In the formula, m is the total vehicle mass, m s Let F be the sprung mass, φ be the body roll angle, and F be the body roll angle. yfl F yrl F yfr and F yrr The lateral forces of the four wheels (front, rear, left, and right) are respectively, I z Let γ be the yaw moment of inertia of the entire vehicle, and l be the yaw angular velocity. s It is half the wheelbase, l f and l r d1 is the distance from the center of mass to the front and rear axes, and d2 is the disturbance term.

[0021] The vehicle roll dynamics model for the aforementioned integrated control method is as follows:

[0022]

[0023] z s1 =z c +l s φ, z s2 =z c -l s φ,

[0024] In the formula, z c I is the vertical displacement of the center of mass of the sprung mass, h is the distance from the center of mass to the tilt axis, and I is the vertical displacement of the sprung mass. x f1 is the sprung mass roll inertia, f2 is the left active suspension control force, and f3 is the right active suspension control force. s Half the wheel track, m u1 For the unsprung mass on the left, m u2 For the unsprung mass on the right side, k t1 For the stiffness of the left tire, k t2For the stiffness of the right tire, z u1 The vertical displacement of the unsprung mass on the left side, z u2 The vertical displacement of the unsprung mass on the right side, z r1 For the vertical excitation of the left tire, z r2 For the vertical excitation of the right tire, k s1 k is the stiffness of the left suspension spring. s2 c represents the stiffness of the right-side suspension spring. s1 For the left suspension damping, c s2 This refers to the damping of the right-side suspension.

[0025] The aforementioned integrated control method uses a two-degree-of-freedom vehicle model as a reference model; let the state-space variable x... d (t)=[β d ,γ d ] T The system input is the reference front wheel steering angle δ. fd , that is u d (t)=[δ fd Then the corresponding state equation of the reference model is:

[0026]

[0027]

[0028] In the formula, β d The sideslip angle representing the centroid of the reference model, i.e., the reference centroid sideslip angle; δ fd Represents the reference front wheel steering angle; γ d The yaw rate represents the reference model, i.e., the reference yaw rate; v xd The longitudinal velocity of the reference model, i.e., the reference longitudinal velocity, v xd Given in the reference model.

[0029] The aforementioned integrated control method, based on the principle that when a car turns, the active suspension controls the active inward tilt of the vehicle body, thereby reducing the roll moment M generated by the gravitational component. G The tilting moment M generated by centrifugal force C Based on the principle of mutual cancellation, the reference camber angle of the vehicle body during steering is obtained:

[0030]

[0031] The beneficial effects of this invention: This invention provides a method for trajectory tracking and vehicle attitude integrated control of an autonomous vehicle with differential steering. This method achieves trajectory tracking of the autonomous vehicle through differential steering, while simultaneously improving cornering speed and handling stability through active suspension to achieve vehicle inclination during steering. First, an autonomous vehicle dynamics model including differential steering and active suspension is established, and a reference model is established to obtain the reference yaw rate and reference vehicle inclination angle. A single-point pre-aiming driver model based on fractional calculus theory is designed to obtain the reference front wheel steering angle required by the reference model. Based on this, an H... ∞ The robust controller tracks the reference values ​​for the yaw rate and body roll angle of the differential steering autonomous vehicle, obtaining the required differential torque and the active suspension control forces on the left and right sides. Simulation results show that the fractional-order single-point anti-drone model and the H-type driver model... ∞ Robust controllers enable differential steering autonomous vehicles to achieve both trajectory tracking and vehicle attitude control. This integrated control method simplifies the model, reduces control steps and data computation, and improves response speed. It can increase the vehicle's cornering speed while effectively preventing the vehicle from tilting outwards under centrifugal force during steering, thus preventing rollovers. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the dynamic model of the front wheel differential steering system.

[0033] Figure 2 This is a schematic diagram of the differential steering dynamics model.

[0034] Figure 3 This is a schematic diagram of the roll dynamics model.

[0035] Figure 4 It is a block diagram of a hierarchical control system.

[0036] Figure 5 This is a schematic diagram of a single-point pre-aiming and tracking model.

[0037] Figure 6 This is a diagram illustrating the driver's steering pattern.

[0038] Figure 7 This is a diagram illustrating the driver's field of vision.

[0039] Figure 8 It is H ∞ Robust control block diagram.

[0040] Figure 9 It is the modified H ∞ Robust control block diagram.

[0041] Figure 10 It is a standard H ∞Robust control block diagram.

[0042] Figure 11 It is a curve comparing the expected path with the actual path of the reference model.

[0043] Figure 12 It is the reference front wheel steering angle curve of the reference model.

[0044] Figure 13 It is a curve comparing yaw rate.

[0045] Figure 14 It is a curve comparing the inward tilt angle of the vehicle body.

[0046] Figure 15 It is the differential torque curve of the left and right front wheels.

[0047] Figure 16 This is a diagram showing the change in the control force of the left-side active suspension.

[0048] Figure 17 This is a diagram showing the change in the active suspension control force on the right side. Detailed Implementation

[0049] This patent describes a distributed-drive differential steering autonomous vehicle where hub motors are installed in all four wheels, and active suspension is used on both the front and rear axles. When there is a difference in driving torque between the left and right front wheels, the autonomous vehicle steers towards the side with the smaller driving torque. Considering that a car tilts outwards due to centrifugal force during steering, this can affect cornering speed and even lead to rollover in severe cases. Furthermore, drastic changes in load on the left and right front wheels can directly cause differential steering failure. However, if the active suspension can control the car to tilt inwards like a bicycle turning, the effectiveness of differential steering can be guaranteed, improving cornering speed and preventing rollover. Therefore, this patent proposes an integrated control method for the active suspension and differential steering of a distributed-drive autonomous vehicle. This integrated control method is described in detail below.

[0050] For distributed drive differential steering autonomous vehicles, this patent designs as follows: Figure 4 The hierarchical control system shown includes upper-level control and lower-level control. The upper-level control obtains the reference front wheel steering angle δ required for the reference model to track the desired path by using a fractional-order single-point anti-aiming driver model. fd This is then fed into the reference model to obtain the reference yaw rate γ. d and reference body camber angle φ d Lower-level control is achieved through H ∞ The robust controller controls the unmanned vehicle to have the same yaw rate and body roll angle as the reference model through differential steering and active suspension, and obtains the differential torque ΔM required for steering and the left and right active suspension control forces f1 and f2 required for body roll.

[0051] 1. Establishment of the dynamic model

[0052] First, a front-wheel differential steering system, a front-wheel differential steering vehicle dynamics model, and a reference model are constructed to study the trajectory tracking and vehicle attitude control of a distributed drive differential steering unmanned vehicle.

[0053] 1.1 Dynamic Model of Front Wheel Differential Steering System

[0054] This paper studies the front-wheel differential steering system of a distributed drive electric vehicle, such as... Figure 1 As shown, where τ al and τ ar The self-centering torques T for the left and right front wheels are respectively. fl and T fr These are the driving torques for the left and right front wheels, respectively.

[0055] Depend on Figure 1 The established dynamic model of the front wheel differential steering system is as follows:

[0056]

[0057]

[0058] In the formula, J e b e and τ f δ represents the equivalent moment of inertia, steering damping, and frictional torque of the steering system, respectively. f Let ΔT be the front wheel steering angle, ΔT be the difference in torques of the two front wheels around the kingpin, and τ be the steering angle of the front wheels. a k is the total self-aligning torque of the front wheels. f For the front wheel lateral stiffness, α f is the front wheel slip angle, and l is the tire contact patch width.

[0059] Define the torque difference ΔT between the left and right front wheels around the kingpin as:

[0060] ΔT=τ dr -τ dl =(F xfr -F xfl )r σ

[0061] In the formula, r σ This is the lateral offset of the master pin.

[0062] Because of the longitudinal force F of the front wheel xfi =T fi / R c ,(i=l,r), then

[0063]

[0064] In the formula, R c Let denot be the wheel radius, and ΔM be the differential torque between the left and right front wheels of the vehicle.

[0065] Combining equations (1), (2), and (3), the dynamic equation of the front wheel differential steering system can be rewritten as:

[0066]

[0067] In the formula, d1 is the interference term, and

[0068] 1.2 Differential Steering Vehicle Dynamics Model

[0069] The planar dynamics model and roll dynamics model of the differential steering vehicle are as follows: Figure 2 and 3 As shown. Figure 3 In this context, y′ represents the direction of tilting of the sprung mass.

[0070] Depend on Figure 2 The established vehicle differential steering dynamics model is as follows:

[0071]

[0072] F yfl =F yfr =k f α f F yrl =F yrr =k r α r ,

[0073] In the formula, m is the total vehicle mass, m s Let F be the sprung mass, ψ be the yaw angle, φ be the body roll angle, and F be the body roll angle. yfl F yrl F yfr and F yrr The lateral forces of the four wheels (front, rear, left, and right) are respectively, I z Let γ be the yaw moment of inertia of the entire vehicle, and l be the yaw angular velocity. s It is half the wheelbase, l f and l r d1 is the distance from the center of mass to the front and rear axes, and d2 is the disturbance term.

[0074] Depend on Figure 3 The established vehicle roll dynamics model is as follows:

[0075]

[0076] z s1=z c +l s φ, z s2 =z c -l s φ,

[0077] In the formula, z c I is the vertical displacement of the center of mass of the sprung mass, h is the distance from the center of mass to the tilt axis, and I is the vertical displacement of the sprung mass. x f1 is the sprung mass roll inertia, f2 is the left active suspension control force, and f3 is the right active suspension control force. s Half the wheel track, m u1 For the unsprung mass on the left, m u2 For the unsprung mass on the right side, k t1 For the stiffness of the left tire, k t2 For the stiffness of the right tire, z u1 The vertical displacement of the unsprung mass on the left side, z u2 The vertical displacement of the unsprung mass on the right side, z r1 For the vertical excitation of the left tire, z r2 For the vertical excitation of the right tire, k s1 k is the stiffness of the left suspension spring. s2 c represents the stiffness of the right-side suspension spring. s1 For the left suspension damping, c s2 This refers to the damping of the right-side suspension.

[0078] Combining (4), (5), and (6), the dynamic model of a vehicle with hub motor-driven front wheel differential steering can be obtained as follows:

[0079]

[0080] Select w p =[d1,d2,δ f ,z r1 ,z r2 ] T u = [ΔM, f1, f2] T The dynamic model of the differential steering autonomous vehicle can then be expressed as:

[0081]

[0082] Among them, A p B is an 11x11 matrix. p1 B is an 11x5 matrix. p2 It is an 11-row, 3-column matrix. For A... p In terms of matrices, if we take A p (1,1) represents A p The elements in the first row and first column of the matrix are then:

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098] All other elements are zero.

[0099] For B p1 In terms of matrices, B p1 (1,1)=1,

[0100] All other elements are zero.

[0101] For B p2 In terms of matrices,

[0102]

[0103] All other elements are zero.

[0104] 1.3 Reference Model

[0105] Here, a typical two-degree-of-freedom vehicle model is selected as the reference model to obtain the reference yaw rate γ. d and reference body camber angle φ d Let the state-space variable x d (t)=[β d ,γ d ] T The system input is the reference front wheel steering angle δ. fd , that is u d (t)=[δ fd Then the corresponding state equation of the reference model is:

[0106]

[0107]

[0108] In the formula, β d The sideslip angle representing the centroid of the reference model, i.e., the reference centroid sideslip angle; δ fd Represents the reference front wheel steering angle; γ d The yaw rate represents the reference model, i.e., the reference yaw rate; v xd The longitudinal velocity of the reference model, i.e., the reference longitudinal velocity, v xd Given in the reference model.

[0109] The reference yaw rate γ can be obtained from (9). d If a car can actively control the body's tilt when turning through its active suspension, thus reducing the roll moment M generated by the component of gravity... G The tilting moment M generated by centrifugal force C By canceling each other out, the reference camber angle φ during steering can be obtained. d .

[0110] Depend on Figure 3 We can obtain,

[0111]

[0112] Solving the two equations above simultaneously, and considering the steady-state turning point... Reference body camber angle can be obtained

[0113]

[0114] 2. Design of a hierarchical control system

[0115] For distributed-drive differential steering autonomous vehicles, this paper designs the following... Figure 4 The hierarchical control system shown includes upper-level control and lower-level control. The upper-level control obtains the reference front wheel steering angle δ required for the reference model to track the desired path by using a fractional-order single-point anti-aiming driver model.fd This is then fed into the reference model to obtain the reference yaw rate γ. d and reference body camber angle φ d Lower-level control is achieved through H ∞ The robust controller controls the unmanned vehicle to have the same yaw rate and body roll angle as the reference model through differential steering and active suspension, and obtains the differential torque ΔM required for steering and the left and right active suspension control forces f1 and f2 required for body roll.

[0116] from Figure 4 It can be seen that the reference front wheel steering angle δ required to track the desired path can be obtained from the fractional-order single-point preview driver model based on the desired path. fd The reference model uses the front wheel steering angle δ as a reference. fd The input is the reference yaw rate γ. d and reference body camber angle φ d H ∞ The robust controller, on the other hand, is based on the reference yaw rate γ. d and reference body camber angle φ d The differential torque ΔM required for the differential steering autonomous vehicle model to track the reference model and the left and right active suspension control forces f1 and f2 required for the vehicle body tilt are calculated based on the actual yaw rate γ and actual body tilt angle φ output by the differential steering autonomous vehicle model.

[0117] 2.1 Design of the Upper-Level Controller

[0118] The driver first selects an appropriate vehicle aiming distance based on the vehicle speed, estimates the deviation between the vehicle's front and the desired path, and determines the required steering wheel angle based on this deviation, thereby changing the car's direction to reduce the deviation. The entire process is a continuous adjustment as the vehicle moves and the path changes. The driver's single-point aiming model is as follows: Figure 5 As shown.

[0119] Figure 5 In this context, XOY represents the geodetic coordinate system, and xoy represents the vehicle coordinate system (point O coincides with the car's center of mass). Assume the driver is located at the car's center of mass point O, and p(x,y) is the driver's aiming point. p It is the driver's aiming distance (i.e., the distance from point p to the y-axis). v is the lateral speed of the car. x Let y be the longitudinal speed of the car, then the aiming deviation is... ε The deviation between the target point p and the actual driving path of the car can be expressed as:

[0120]

[0121] The driver turns the steering wheel according to the anti-aiming deviation, that is, the front wheel angle changes the direction of travel to reduce the anti-aiming deviation. The required front wheel angle can be expressed as:

[0122] δ f (t)=k m y ε (tT s (13)

[0123] In the formula, t is the current time, T is the time delay, and K is the driver's control gain.

[0124] sm

[0125] Laplace transform on both sides of (13)

[0126]

[0127] In (14) Expanding this into a Taylor series, and considering that the time delay is small, we only need to take the first two terms of the series, which gives us:

[0128] (1+T s s)δ f (s)=k m yε(s) (15)

[0129] Performing an inverse Laplace transform on (15) yields:

[0130]

[0131] The above (12) and (16) are the driver's single-point preview model, but this model assumes that the driver only pays attention to the road information at the preview point. In fact, in addition to focusing on the road information at the preview point, the driver usually also pays attention to other road information in the field of vision. However, the road information at the preview point is the clearest, and the road information far away from the preview point will gradually become blurred as the distance increases.

[0132] Considering that fractional calculus inherently carries weights during calculation, and these weights gradually decrease from 1, this pattern aligns perfectly with the driver's field of vision. Therefore, fractional calculus can be introduced to comprehensively consider road information within the driver's field of vision. The maximum weight of the fractional calculus can be correlated with the driver's pre-aiming point p(x,y). The weights further away from the pre-aiming point will gradually decrease. Figure 6 As shown. Driver's field of vision is as follows. Figure 7 As shown.

[0133] from Figure 7 It can be seen that 'a' is the longitudinal distance from the preview point to the rear edge of the line of sight, and 'b' is the longitudinal distance from the preview point to the front edge of the line of sight. Introducing fractional calculus, 'y' can be expressed as:

[0134]

[0135] In the formula, Let f(t) be a fractional calculus operator, where α and α' are the fractional orders, f(t) is the desired path, and its function is used as the antiderivative of the fractional integral. The integration region corresponds to... Figure 7 The red-lined area in the text.

[0136] Substituting (17) into (12) and combining (16), we can obtain the fractional-order single-point pre-aiming driver model:

[0137]

[0138] 2.2 Design of the Lower-Level Controller

[0139] Lower-level control is mainly achieved through H ∞ The robust controller controls the unmanned vehicle to have the same yaw rate and body roll angle as the reference model through differential steering and active suspension, and obtains the differential torque ΔM required for differential steering and the left and right active suspension control forces f1 and f2 required for body roll. Figure 8 For H ∞ Robust control block diagram.

[0140] Let the reference input r = [γ] d ,φ d ] T The controller's transfer function and output signal are K(s) and u, respectively. The differential steering autonomous vehicle's transfer function is P(s), the external disturbance is d, and the actual output is v = [γ, φ]. T Let e ​​be the tracking error, i.e., e = rv, and introduce equivalent disturbances (ω1, ω2) and two measurement outputs z1 = rv, z2 = ρu. Figure 8 The control system shown becomes as follows Figure 9 The control system shown.

[0141] Our control objective is to minimize the tracking error e and the weighted control energy ρu under the influence of external disturbances. Minimum, which is the measured output signal z = [rv, ρu] T The 2-norm is the smallest. Figure 9 The control system shown can be converted into, for example... Figure 10 The standard H shown ∞ form.

[0142] Select z = [rv, ρu] T , y=[rv], w=[w1,w2,d1,d2,δ f ,zr1 ,z r2 ] T ,but Figure 10 The controlled system can be represented as:

[0143]

[0144] In the formula, D 21 =[0 2×7 ], D 22 =[0 2×3 ],

[0145]

[0146] Finally, the state-space expression of the controller K(s) can be obtained using the function hinfsyn(). Once the state-space expression of the controller K(s) is obtained, it is substituted into... Figure 10 From the simulation model shown, we can obtain u = [ΔM, f1, f2]. T The curves are used to obtain the differential torque ΔM and the left and right active suspension control forces f1 and f2 required for body roll.

[0147] 3. Simulation test

[0148] To verify the established single-point anti-shooting driver model and H ∞ The effectiveness of the robust controller is demonstrated in this section through simulation experiments using a Carsim B-type vehicle in MATLAB / Simulink. The main vehicle parameters are shown in Table 1.

[0149] Table 1 Main Vehicle Parameters

[0150]

[0151] The simulation was set to a single lane change condition, with a constant vehicle speed of 20 m / s, a road surface adhesion coefficient of 0.8, and a simulation time of 25 seconds. Figure 11 This is a comparison chart of the expected path and the actual path of the reference model. From Figure 11 It can be seen that, under the action of the upper-level controller, the deviation between the actual path and the expected path of the reference model is small, with the maximum deviation being 1.17m, which is less than the half-width of the road, 2m.

[0152] Under the control of the upper-level controller, the reference model determines the front wheel steering angle required for tracking the desired path, such as... Figure 12 As shown in the figure, the maximum front wheel steering angle is 0.043 rad (2.46°).

[0153] Under the control of the lower-level controller, the comparison curves of the yaw rate and body roll angle of the differential steering autonomous vehicle with the reference yaw rate and reference body roll angle are as follows: Figure 13 and 14 As shown. From Figure 13 and 14 It can be seen that, under the action of the lower-level controller, the differential steering unmanned vehicle can effectively track both the upper reference yaw rate and the reference body roll angle simultaneously.

[0154] Furthermore, under the control of the lower-level controller, the differential torque required for the left and right front wheels of the differential steering autonomous vehicle to track the reference yaw rate is as follows: Figure 15 As shown. From Figure 15 It can be seen that the maximum absolute value of this differential torque is 26 Nm.

[0155] Under the control of the lower-level controller, the left and right active suspension control forces required for the differential steering autonomous vehicle to track the reference inclination angle are respectively as follows: Figure 16 and 17 As shown. From Figure 16 and 17 It can be seen that the maximum absolute values ​​of the active suspension control forces on the left and right sides are 9×10. 4 N, 8×10 4 N.

[0156] In summary, the single-point pre-aiming driver model (i.e., the upper-level controller) designed in this patent, based on fractional calculus theory, can obtain the front wheel steering angle required by the reference model. The H-axis designed in this paper... ∞ The robust controller (i.e., the lower-level controller) can control the differential steering autonomous vehicle to track the upper reference yaw rate and the reference body roll angle, and obtain the required differential torque and the left and right active suspension control forces.

Claims

1. An integrated control method for active suspension and differential steering in a distributed-drive unmanned vehicle, wherein hub motors are installed in all four wheels of the distributed unmanned vehicle, and active suspension is used on both the front and rear axles; its characteristics are: The integrated control method includes the following steps: constructing a dynamic model of the front wheel differential steering system, a dynamic model of the front wheel differential steering vehicle, and a reference model; obtaining the reference front wheel steering angle required for the reference model to track the desired path by using a fractional-order single-point pre-aiming driver model. And provide it to the reference model to obtain the reference yaw rate. and reference body camber angle H ∞ Robust controller based on reference yaw rate and reference body camber angle The autonomous vehicle is controlled by differential steering and active suspension to achieve the same yaw rate and body roll angle as the reference model, and to obtain the differential torque required for steering. And the left and right active suspension control forces required for vehicle body roll. and To prevent the autonomous vehicle from overturning when turning; Fractional-order single-point pre-aiming driver model: （18） Establish a vehicle coordinate system xoy with the car's center of mass as the origin, and position the driver at the car's center of mass point o. It is the driver's aiming point. It is the driver's pre-aiming distance. Click Distance between axes The longitudinal speed of the car For the lateral speed of the car, the aiming deviation Pre-aiming point The deviation between the actual driving path and the actual driving path of the car. For the front wheel steering angle, For the current time, It is a time delay. For the driver's control gain, This is the longitudinal distance from the aiming point to the rear edge of the line of sight. The longitudinal distance from the aiming point to the front edge of the line of sight. Let be a fractional calculus operator, where α and α' are the fractional orders. This is the expected path.

2. The integrated control method as described in claim 1, characterized in that: the front wheel The dynamic model of the differential steering system is as follows: （1） ， （2） In the formula, , and These represent the equivalent moment of inertia, steering damping, and frictional torque of the steering system, respectively. This represents the difference in torque around the kingpin between the two front wheels. This is the total self-aligning torque of the front wheels. For the front wheel lateral stiffness, The front wheel slip angle, The width of the tire contact patch is half the width of the ground.

3. The integrated control method as described in claim 2, characterized in that: The dynamic equation of the front wheel differential steering system is: （4） In the formula, It is a distractor, and ; For the wheel radius, This refers to the differential torque between the left and right front wheels of the vehicle. This is the distance from the center of mass to the front axle.

4. The integrated control method as described in claim 3, characterized in that: The vehicle differential steering dynamics model is as follows: （5） ， ， ， In the formula, For the overall vehicle quality, For the sprung mass, The body roll angle, , , and These are the lateral forces of the four wheels: front, rear, left, and right. The moment of inertia of the vehicle's yaw motion. The yaw rate is angular velocity. It is half the track width. and This is the distance from the center of mass to the front and rear axles. This is a distractor.

5. The integrated control method as described in claim 4, characterized in that: The vehicle roll dynamics model is as follows: （6） ， ， ， ， ， In the formula, Let be the vertical displacement of the center of mass of the spring-loaded mass. The distance from the center of gravity to the roll axis. The moment of inertia of the sprung mass during tilting. For the left-side active suspension control force, The right-side active suspension control force, It is half the wheel track. The unsprung mass on the left side, The unsprung mass on the right side, This refers to the stiffness of the left tire. This refers to the stiffness of the right tire. This represents the vertical displacement of the unsprung mass on the left side. This represents the vertical displacement of the unsprung mass on the right side. The vertical excitation is for the left tire. The vertical excitation is for the right tire. This refers to the stiffness of the left suspension spring. This refers to the stiffness of the right-side suspension spring. For the left suspension damping, This refers to the damping of the right-side suspension.

6. The integrated control method as described in claim 1, characterized in that: A two-degree-of-freedom vehicle model is used as the reference model; state-space variables are defined. The system input is the reference front wheel steering angle. ,Right now Then the corresponding state equation of the reference model is: （9） ， ， ， ； In the formula, The sideslip angle representing the centroid of the reference model, i.e., the reference centroid sideslip angle; Represents the reference front wheel steering angle; This represents the yaw rate of the reference model, i.e., the reference yaw rate. Represents the longitudinal velocity of the reference model, i.e., the reference longitudinal velocity. Given in the reference model.

7. The integrated control method as described in claim 6, characterized in that: Based on the fact that when a car turns, the active suspension can control the body's active inward tilt, thus reducing the roll moment generated by the component of gravity. The tilting moment generated by centrifugal force Based on the principle of mutual cancellation, the reference camber angle of the vehicle body during steering is obtained: （10）。

Images