Distributed driving vehicle obstacle avoidance and stability coordination control method

Abstract

The invention discloses a distributed driving vehicle obstacle avoidance and stability coordination control method. An obstacle avoidance guiding track planning layer and an obstacle avoidance safety and stability coordination control layer are included. The obstacle avoidance safety and stability coordination control layer establishes an optimization problem based on a model prediction control framework, and comprises a prediction model, a constraint condition, a target function and an optimization solution link; the constraint conditions comprise obstacle avoidance boundary constraint, stable envelope constraint and execution boundary constraint; the obstacle avoidance boundary constraint is a vehicle drivable area constraint; the stable envelope constraint comprises yaw velocity stable boundary constraint and side slip angle stable boundary constraint; the execution boundary constraint comprises a control quantity constraint and a control quantity increment constraint; the objective function comprises an obstacle avoidance guide trajectory tracking precision item, a stable factor item, a control quantity item and a control quantity increment item; and safe and stable obstacle avoidance control is comprehensively realized.

Description

Technical Field

[0001] The present invention relates to the field of vehicle obstacle avoidance control, and in particular to the field of a method for coordinated control of obstacle avoidance safety and stability of a distributed drive vehicle. Background Art

[0002] In today's production and life, assisted driving and autonomous driving have gradually become part of our lives; the obstacle avoidance trajectory planning method is particularly important in the process of autonomous driving.

[0003] Traditional obstacle avoidance safety control and vehicle stability control are independent of each other. In practical applications, vehicle stability control is overly dependent on driver control, which can easily lead to a failure to maintain vehicle stability during obstacle avoidance due to improper driver control or to obstacle avoidance failure due to excessive driver intervention, posing a safety hazard. Model predictive control, with its ability to handle multi-objective and multi-constraint optimization problems, provides a technical means for coordinated control of obstacle avoidance safety and stability. However, existing research has not fully considered the need for coordinated optimization of obstacle avoidance safety and stability at the design level of obstacle avoidance optimization problems.

[0004] Therefore, how to invent a method for coordinated control of obstacle avoidance and stability of distributed drive vehicles to ensure the safety and stability of obstacle avoidance during the obstacle avoidance process has become a key issue that needs to be solved urgently. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a distributed drive vehicle obstacle avoidance and stability control method. Based on the model predictive control framework, coordinated control is achieved by constructing constraints and objective functions related to obstacle avoidance safety and stability, thereby comprehensively improving obstacle avoidance safety and vehicle stability during the obstacle avoidance process.

[0006] The present invention provides a method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle, which is characterized in that:

[0007] It includes obstacle avoidance guidance trajectory planning layer and obstacle avoidance safety and stability coordination control layer;

[0008] The obstacle avoidance guidance trajectory planning layer considers the actual obstacle avoidance conditions to generate a guidance trajectory and determines the expected values ​​of the vehicle's longitudinal speed, lateral position, and yaw angle during the obstacle avoidance process.

[0009] The obstacle avoidance safety and stability coordination control layer establishes an optimization problem based on the model predictive control framework, and the obstacle avoidance safety and stability coordination control layer includes a prediction model, constraints and objective functions, and an optimization solution link;

[0010] The prediction model is established based on a three-degree-of-freedom vehicle dynamics model; the prediction model obtains the vehicle state change in the predicted time domain based on the current state and control input, and assists in determining the optimized control quantity in combination with the constraint conditions and the objective function;

[0011] The constraints include obstacle avoidance boundary constraints, stability envelope constraints, and execution boundary constraints;

[0012] The obstacle avoidance boundary constraint is a constraint on the vehicle's drivable area; the stability envelope constraint includes a yaw rate stability boundary constraint and a center of mass sideslip angle stability boundary constraint; the execution boundary constraint includes a control quantity constraint condition and a control quantity increment constraint condition; the optimized control quantity is selected based on the constraints to enhance obstacle avoidance safety, vehicle stability, and control smoothness;

[0013] The objective function includes an obstacle avoidance trajectory tracking accuracy term, a stability factor term, a control amount term, and a control amount increment term;

[0014] The objective function quantitatively characterizes obstacle avoidance safety, vehicle stability, and control smoothness by softly constraining the vehicle state, control quantity, and control increment within the prediction time domain; the control quantity corresponding to the minimum value of the objective function is the optimal control quantity;

[0015] The vehicle state quantities in the predicted time domain include vehicle trajectory information and vehicle stability state information; the vehicle trajectory information includes longitudinal velocity, lateral position, and yaw angle, and obstacle avoidance safety is improved by approaching the expected value obtained based on the obstacle avoidance guidance trajectory; the vehicle stability state information includes yaw angular velocity and center of mass sideslip angle, and vehicle stability is enhanced by constructing a stability factor project objective function in combination with a stability envelope constraint;

[0016] The optimization solution uses a sequential quadratic programming solver and sets an iterative warm start to obtain the optimal control sequence in the control time domain, and its first item is used as the actual output optimized control quantity;

[0017] The control variables include the front wheel angle and the four wheel longitudinal forces, wherein the wheel longitudinal forces include the left front wheel longitudinal force, the right front wheel longitudinal force, the left rear wheel longitudinal force, and the right rear wheel longitudinal force;

[0018] The yaw rate stability boundary constraint is a stability boundary value for the vehicle's yaw rate w determined based on the road adhesion condition and the vehicle's speed. The yaw rate stability boundary constraint expression is:

[0019]

[0020] ω min ≤w≤ω max ;

[0021] The ω max is the upper limit of the yaw angular velocity, the ω min is the lower limit of yaw rate, V x is the longitudinal velocity of the vehicle, μ is the road adhesion coefficient, and g is the acceleration due to gravity;

[0022] The center of mass sideslip angle stability boundary constraint determines the stability boundary value of the center of mass sideslip angle β based on the road adhesion condition and tire saturation characteristics. The center of mass sideslip angle stability boundary constraint expression is:

[0023]

[0024] β min ≤β≤β max ;

[0025] The β max is the upper boundary of the center of mass sideslip angle, and the β min is the lower boundary of the center of mass side slip angle, b is the longitudinal distance between the center of mass of the vehicle and the rear axle, α sat is the tire saturation slip angle.

[0026] Compared with the existing technology, the present invention has the following advantages: based on the model predictive control framework, the constraints and objective functions related to obstacle avoidance safety and stability are designed to achieve coordinated optimization control of obstacle avoidance safety and vehicle stability, comprehensively improving the safety and stability of the obstacle avoidance process;

[0027] The stability envelope constraint includes the yaw rate stability boundary constraint and the center of mass sideslip angle stability boundary constraint. This constraint requires that the vehicle state always stays within the stability envelope constraint to fully utilize road adhesion to achieve the goal of safe obstacle avoidance. When the vehicle is within the stability envelope area, the vehicle is in a stable state, while when the vehicle exceeds the stability envelope boundary, instability may occur. To further quantitatively evaluate the vehicle's stability, this patent constructs a stability factor project objective function based on the relative position of the vehicle's expected state and the stability envelope to quantitatively evaluate the vehicle's stability characteristics, thus realizing soft constraints for vehicle stability control.

[0028] The obstacle avoidance trajectory planning layer generates the obstacle avoidance trajectory based on a quintic polynomial. It determines the expected values ​​of the vehicle's longitudinal velocity, lateral position, and yaw angle. Based on this, an obstacle avoidance safety objective function is designed to enhance obstacle avoidance safety by requiring the obstacle avoidance process to adhere as closely as possible to the obstacle avoidance trajectory.

[0029] Based on the obstacle avoidance guidance trajectory and stability envelope, the obstacle avoidance safety and stability coordination control layer further considers vehicle dynamics, stability envelope constraints, obstacle avoidance safety constraints, execution boundary constraints and other constraints based on the model predictive control framework, as well as objective functions such as obstacle avoidance safety, vehicle stability, and control smoothness to replan and control the obstacle avoidance trajectory, effectively improving the feasibility of the obstacle avoidance trajectory.

[0030] The vehicle's stability is characterized by yaw rate and center-of-mass slip angle. The yaw rate can intuitively represent the vehicle's stability and its changes under the influence of stability control schemes such as steering, driving, and braking. The center-of-mass slip angle can effectively characterize the vehicle's nonlinear characteristics. When in a stable state, the center-of-mass slip angle is small, and the vehicle's actual posture is consistent with the expected maneuverability. However, when the vehicle becomes unstable, the center-of-mass slip angle is large, indicating that the deviation between the vehicle's velocity direction and the vehicle's body direction increases. This significantly enhances the vehicle's nonlinearity, which can easily cause excessive panic in the driver and passengers, leading to improper operation and loss of vehicle control. By constructing a stability envelope constraint using the yaw rate stability bound and the center-of-mass slip angle stability bound, the vehicle's stability during obstacle avoidance is enhanced.

[0031] Furthermore, the method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle is characterized in that the control quantity constraint expression is:

[0032] u min ≤u(k+i)≤u max ;

[0033] The control quantity increment constraint expression is:

[0034] Δu min ≤Δu(k+i)≤Δu max ;

[0035] The control quantity mainly includes the front wheel angle δ f and the four wheel longitudinal forces F xfl 、F xfr 、F xrl 、F xrr ;u max ,u min are the maximum and minimum values ​​of the control quantity, respectively, which are determined by the physical limit boundary of the actuator; Δu max ,Δu min are the maximum and minimum values ​​of the control quantity increment respectively;

[0036] The obstacle avoidance constraint requires that the vehicle always travels within the feasible area during the obstacle avoidance process, that is, the positions of the four vertices of the vehicle are always within the safety boundary constraint range. Its expression is:

[0037]

[0038] Y fl 、Y fr 、Y rl 、Y rr Y is the predicted value of the lateral position of the four vertices of the vehicle, o is the lateral position of the vehicle's center of mass, a is the longitudinal distance between the vehicle's center of mass and the front axle, b is the longitudinal distance between the vehicle's center of mass and the rear axle, d is the vehicle's wheelbase, and ψ is the vehicle's yaw angle; Y Ufl 、YUfr 、Y Url 、Y Urr and Y Lfl 、Y Lfr 、Y Lrl 、Y Lrr They are the upper and lower boundary values ​​of the obstacle avoidance constraints corresponding to the four vertices of the vehicle.

[0039] The beneficial effect of the previous step is that it generates obstacle avoidance safety constraints based on obstacle information and road boundary information. This constraint strictly requires the vehicle to remain within the obstacle avoidance safety boundary during obstacle avoidance. Specifically, the positions of the vehicle's four vertices are always within the obstacle avoidance safety constraint. This constraint transforms the complex obstacle avoidance environment into the upper and lower boundaries of a rectangular convex space, linearizing the obstacle avoidance constraint and improving the computational efficiency of the optimization solution. Furthermore, by dilating the obstacles, the likelihood of collision between the vehicle and surrounding obstacles is reduced, thereby enhancing obstacle avoidance safety.

[0040] Furthermore, the objective function expression is:

[0041]

[0042] k is the sampling time; N p is the prediction time domain; N c To control the time domain; the first term of the objective function is the obstacle avoidance trajectory tracking accuracy term, and the first term of the objective function y p (k+i|k) is the predicted value of vehicle trajectory information in the prediction time domain, and the first term of the objective function y ref (k+i|k) is the expected value of the vehicle trajectory information in the predicted time domain determined based on the obstacle avoidance guidance trajectory; the second term u(k+i) is the control quantity term, the third term Δu(k+i) is the control quantity increment term; the fourth term S index (k+i|k) is the stability factor term;

[0043] Q, R, S, and P are four corresponding weight matrices;

[0044] Q is the weight matrix of the obstacle avoidance trajectory tracking accuracy term. The weight matrix Q includes the longitudinal velocity tracking bias weight, the vehicle lateral position tracking bias weight, and the vehicle yaw angle tracking bias weight. The longitudinal velocity tracking bias weight value is 2000-5000; the lateral displacement tracking bias weight value is 5000-8000; and the vehicle yaw angle tracking bias weight value is 5000-8000.

[0045] R is the weight matrix of the control quantity term, which includes the front wheel angle weight and the wheel longitudinal force weight. The front wheel angle weight value is 50,000-80,000; the wheel longitudinal force weight value is 10,000-20,000.

[0046] S is the weight matrix of the control quantity increment term, which includes the weight of the front wheel angle change rate and the weight of the wheel longitudinal force change rate; the weight value of the front wheel angle change rate is 50,000-80,000; the weight value of the wheel longitudinal force change rate is 10,000-20,000;

[0047] P is the weight of the stability factor item, and the weight value of the stability factor item is 50-500.

[0048] The beneficial effect of adopting the previous step is that the objective function part includes the obstacle avoidance trajectory tracking accuracy term, the control quantity amplitude term, the control quantity increment term, and the stability factor. The first term is mainly to make full use of the obstacle avoidance guidance trajectory to improve obstacle avoidance safety; the second term is mainly to minimize vehicle energy loss as much as possible; the third term is mainly to ensure that vehicle control is as stable as possible; the fourth term is a stability factor constructed based on the stability constraint boundary, which can quantitatively characterize the vehicle's stable state and reduce the risk of the vehicle exceeding the stability envelope boundary.

[0049] Furthermore, the stability factor S is determined based on the relative distance between the expected state of the vehicle and the stability envelope. index To quantitatively characterize the vehicle's stable state, the expression is:

[0050]

[0051] Furthermore, the prediction model includes the following formula:

[0052]

[0053] in,

[0054]

[0055] δ f is the front wheel angle, F xfl 、F xfr 、F xrl 、F xrr is the longitudinal force corresponding to the four wheels, namely the left front wheel, right front wheel, left rear wheel and right rear wheel, F yfl 、F yfr 、F yrl 、F yrr They are the lateral force of the left front wheel, the lateral force of the right front wheel, the lateral force of the left rear wheel, and the lateral force of the right rear wheel, respectively. drag is the longitudinal resistance, ΔM z is the additional yaw moment, a is the longitudinal distance between the vehicle's center of mass and the front axle, b is the longitudinal distance between the vehicle's center of mass and the rear axle, d is the vehicle's wheelbase, and I z is the moment of inertia of the vehicle around the z axis, m is the mass of the vehicle, v x is the longitudinal velocity, v yis the lateral velocity, ψ is the vehicle yaw angle, w is the vehicle yaw angular velocity, is the longitudinal velocity change rate, is the lateral velocity change rate, is the longitudinal velocity in geodetic coordinates, is the geodetic coordinate lateral velocity.

[0056] The beneficial effect of the previous step is that the prediction model can effectively improve obstacle avoidance safety, vehicle stability, and control smoothness by predicting the changes in the vehicle's longitudinal speed, lateral position, and yaw angular velocity over a period of time. Combined with the constraints and objective function, the feasibility of the obstacle avoidance control sequence is comprehensively evaluated.

[0057] Furthermore, the method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1 is characterized in that the step of generating the guidance trajectory and determining the expected value of the vehicle trajectory information in the obstacle avoidance guidance trajectory planning layer is as follows:

[0058] The vehicle obstacle avoidance guidance trajectory is the moving position reference value of the vehicle's center of mass (X(t), Y(t)) during the obstacle avoidance process;

[0059] Get the vehicle longitudinal velocity V x , size information, safety threshold information; the size information includes the vehicle width W v , the distance d between the center of mass of the vehicle and the front end f , obstacle lateral boundary position Y c ;

[0060] The safety threshold information includes the maximum lateral acceleration a ymax , a ymax =0.4*μ*g, where μ is the road adhesion coefficient and g is the acceleration due to gravity;

[0061] The vehicle obstacle avoidance guidance trajectory formula is established based on the fifth-order polynomial, and the safety threshold information is substituted into the obstacle avoidance guidance trajectory formula to obtain the obstacle avoidance guidance trajectory (X ref ,Y ref );

[0062] Combined with the obstacle avoidance guidance trajectory information, the vehicle yaw angle ψ is obtained by formula 1 ref Expected value;

[0063] The formula 1 is:

[0064]

[0065] k is the sampling time; ΔY ref (k) is the change in lateral position at adjacent moments; ΔX ref (k) is the change in longitudinal position between adjacent moments.

[0066] Furthermore, the obstacle avoidance guidance trajectory formula is:

[0067] Y(t)=a5*t 5 +a4*t 4 +a3*t 3 +a2*t 2 +a1*t+a0;

[0068] a0=0,a1=V x ,a2=0,

[0069] Y w =Y c +Δd ys -Y0;

[0070]

[0071] a0, a1, a2, a3, a4, a5 are the obstacle avoidance trajectory curve fitting coefficients, Δd ys is the lateral safety distance, Y w The lateral displacement required to avoid obstacles, Y c is the lateral boundary position of the obstacle, Δd ys is the lateral safety distance, and Y0 is the initial lateral displacement for obstacle avoidance.

[0072] Furthermore, the obstacle avoidance guidance trajectory is:

[0073]

[0074] X(t) is the longitudinal position of the center of mass of the vehicle at time t during the obstacle avoidance process, and Y(t) is the lateral position of the center of mass of the vehicle at time t during the obstacle avoidance process;

[0075] t f is the obstacle avoidance time, and the calculation formula is as follows

[0076]

[0077] Y w =Y c +Δd ys -Y0;

[0078]

[0079] Furthermore, sequential quadratic programming is used to solve the integrated nonlinear optimization problem, and the obtained optimized solution sequence U * (t-1) moves forward one step and serves as the initial solution U(t) for the next time step initial , thus forming an iterative warm restart;

[0080] BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 This is a comparison chart of the obstacle avoidance trajectories of Example 1, no stabilization control, and traditional stabilization control in a medium-speed and medium-attachment scenario;

[0082] Figure 2 1 is a comparison chart of the stability control effects of Example 1, no stability control, and traditional stability control in a medium-speed and medium-attachment scenario;

[0083] Figure 3 This is a comparison chart of the obstacle avoidance trajectories of Example 1, no stabilization control, and traditional stabilization control in a medium-speed and low-attachment scenario;

[0084] Figure 4 1 is a comparison chart of the stability control effects of Example 1, no stability control, and traditional stability control in a medium-speed and low-attachment scenario. DETAILED DESCRIPTION

[0085] In order to better understand the technical solution of the present invention, the present invention is further described below in conjunction with specific embodiments and the accompanying drawings.

[0086] Example 1:

[0087] According to this embodiment, a method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle is provided, comprising an obstacle avoidance guidance trajectory planning layer and an obstacle avoidance safety and stability coordinated control layer;

[0088] The obstacle avoidance guidance trajectory planning layer considers the actual obstacle avoidance conditions to generate a guidance trajectory and determines the expected values ​​of the vehicle's longitudinal speed, lateral position, and yaw angle during the obstacle avoidance process.

[0089] The obstacle avoidance safety and stability coordination control layer establishes an optimization problem based on the model predictive control framework, including prediction model, constraints and objective function, and optimization solution link;

[0090] The prediction model is established based on a three-degree-of-freedom vehicle dynamics model; the prediction model obtains vehicle state changes in the predicted time domain based on the current state and control input, and assists in determining the optimized control amount in combination with constraints and objective functions.

[0091] The prediction model includes the following formula:

[0092]

[0093] in,

[0094]

[0095] δ f is the front wheel angle, Fxfl 、F xfr 、F xrl 、F xrr is the longitudinal force of the four wheels, F yfl 、F yfr 、F yrl 、F yrr They are the lateral force of the left front wheel, the lateral force of the right front wheel, the lateral force of the left rear wheel, and the lateral force of the right rear wheel, respectively. drag is the longitudinal resistance, ΔM z is the additional yaw moment, a is the longitudinal distance between the vehicle's center of mass and the front axle, b is the longitudinal distance between the vehicle's center of mass and the rear axle, d is the vehicle's wheelbase, and I z is the moment of inertia of the vehicle around the z axis, m is the mass of the vehicle, v x is the longitudinal velocity, v y is the lateral velocity, ψ is the vehicle yaw angle, w is the vehicle yaw angular velocity, is the longitudinal velocity change rate, is the lateral velocity change rate, is the longitudinal velocity in geodetic coordinates, is the geodetic coordinate lateral velocity.

[0096] The steps of the obstacle avoidance guidance trajectory planning layer to generate the guidance trajectory and determine the expected value of the vehicle trajectory information are as follows:

[0097] The vehicle obstacle avoidance guidance trajectory is the moving position reference value of the vehicle's center of mass (X(t), Y(t)) during the obstacle avoidance process;

[0098] Get the vehicle longitudinal velocity V x , size information, safety threshold information; the size information includes the vehicle width W v , the distance d between the center of mass of the vehicle and the front end f , obstacle lateral boundary position Y c ;

[0099] The safety threshold information includes the maximum lateral acceleration a ymax , a ymax =0.4*μ*g, where μ is the road adhesion coefficient and g is the acceleration due to gravity;

[0100] The vehicle obstacle avoidance guidance trajectory formula is established based on the fifth-order polynomial, and the safety threshold information is substituted into the obstacle avoidance guidance trajectory formula to obtain the obstacle avoidance guidance trajectory (X ref ,Y ref );

[0101] Combined with the obstacle avoidance guidance trajectory information, the vehicle yaw angle ψ is obtained by formula 1 ref Expected value;

[0102] The formula 1 is:

[0103]

[0104] Among them, ΔY ref (k) = Y ref (k)-Y ref (k-1)

[0105] ΔX ref (k) = X ref (k)-X ref (k-1);

[0106] k is the sampling time; ΔY ref (k) is the change in lateral position at adjacent moments; ΔX ref (k) is the change in longitudinal position between adjacent moments.

[0107] The obstacle avoidance guidance trajectory formula is:

[0108] Y(t)=a5*t 5 +a4*t 4 +a3*t 3 +a2*t 2 +a1*t+a0

[0109] a0=0,a1=V x ,a2=0,

[0110] Among them, Y w =Y c +Δd ys -Y0,

[0111] a0, a1, a2, a3, a4, and a5 are the obstacle avoidance trajectory curve fitting coefficients; Δd ys It is the lateral safety distance.

[0112] The obstacle avoidance guidance trajectory is:

[0113]

[0114] X(t) is the longitudinal position of the center of mass of the vehicle at time t during the obstacle avoidance process, and Y(t) is the lateral position of the center of mass of the vehicle at time t during the obstacle avoidance process;

[0115] t f The obstacle avoidance time is calculated as follows:

[0116]

[0117] Y w =Y c +Δd ys -Y0,

[0118] The objective function includes an obstacle avoidance trajectory tracking accuracy term, a control quantity term, a control quantity increment term, and a stability factor term;

[0119] The objective function expression is:

[0120]

[0121] k is the sampling time; N p is the prediction time domain; N c To control the time domain; the first term of the objective function is the obstacle avoidance trajectory tracking accuracy term, and the first term of the objective function y p (k+i|k) is the predicted value of vehicle trajectory information in the prediction time domain, and the first term of the objective function y ref (k+i|k) is the expected value of the vehicle trajectory information in the predicted time domain determined based on the obstacle avoidance guidance trajectory; the second term u(k+i) is the control quantity term, the third term Δu(k+i) is the control quantity increment term; the fourth term S index (k+i|k) is the stability factor term;

[0122] Q, R, S, and P are four corresponding weight matrices;

[0123] Q is the weight matrix of the obstacle avoidance trajectory tracking accuracy term. The weight matrix Q includes the longitudinal velocity tracking bias weight, the vehicle lateral position tracking bias weight, and the vehicle yaw angle tracking bias weight. The longitudinal velocity tracking bias weight value is 3500; the lateral displacement tracking bias weight value is 6500; and the vehicle yaw angle tracking bias weight value is 6500.

[0124] R is the weight matrix of the control quantity term, which includes the front wheel angle weight and the wheel longitudinal force weight. The front wheel angle weight is 65000; the wheel longitudinal force weight is 15000.

[0125] S is the weight matrix of the control quantity increment term. The weight matrix S includes the weight of the front wheel angle change rate and the weight of the wheel longitudinal force change rate. The weight value of the front wheel angle change rate is 65000; the weight value of the wheel longitudinal force change rate is 15000.

[0126] P is the weight of the stability factor item, and the weight value of the stability factor item is 50ˉ500.

[0127] Determine the stability factor S based on the relative distance between the vehicle's expected state and the stability envelope index To quantitatively characterize the vehicle's stable state, the expression is:

[0128]

[0129] The constraints include obstacle avoidance boundary constraints, stability envelope constraints, and execution boundary constraints;

[0130] The obstacle avoidance boundary constraint is a constraint on the vehicle's drivable area; the stability envelope constraint includes a yaw rate stability boundary constraint and a center of mass sideslip angle stability boundary constraint; the execution boundary constraint includes a control quantity constraint condition and a control quantity increment constraint condition;

[0131] The yaw rate stability boundary constraint is a stability boundary value for the vehicle's yaw rate w determined based on the road adhesion condition and the vehicle's speed. The yaw rate stability boundary constraint expression is:

[0132]

[0133] ω min ≤w≤ω; max

[0134] The ω max is the upper limit of the yaw angular velocity, ω min is the lower limit of yaw rate, V x is the longitudinal velocity of the vehicle, μ is the road adhesion coefficient, and g is the acceleration due to gravity;

[0135] The center of mass sideslip angle stability boundary constraint determines the stability boundary value of the center of mass sideslip angle β based on the road adhesion condition and tire saturation characteristics. The center of mass sideslip angle stability boundary constraint expression is:

[0136]

[0137] β min ≤β≤β max ;

[0138] The β max is the upper boundary of the center of mass sideslip angle, and the β min is the lower boundary of the center of mass side slip angle, b is the longitudinal distance between the center of mass of the vehicle and the rear axle, α sat is the tire saturation slip angle.

[0139] The control quantity boundary constraint expression is:

[0140] u min ≤u(k+i)≤u max ;

[0141] The control quantity increment boundary constraint expression is:

[0142] Δu min ≤Δu(k+i)≤Δu max ;

[0143] The control quantity mainly includes the front wheel angle δ f and the four wheel longitudinal forces F xfl 、F xfr 、Fxrl 、F xrr ;u max ,u min are the maximum and minimum values ​​of the control quantity, respectively, which are determined by the physical limit boundary of the actuator; Δu max ,Δu min are the maximum and minimum values ​​of the control quantity increment respectively;

[0144] The obstacle avoidance constraint requires that the vehicle always travels within the feasible area during the obstacle avoidance process, that is, the positions of the four vertices of the vehicle are always within the safety boundary constraint range. Its expression is:

[0145]

[0146] Y fl 、Y fr 、Y rl 、Y rr Y is the predicted value of the lateral position of the four vertices of the vehicle, o is the lateral position of the vehicle's center of mass, a is the longitudinal distance between the vehicle's center of mass and the front axle, b is the longitudinal distance between the vehicle's center of mass and the rear axle, d is the vehicle's wheelbase, and ψ is the vehicle's yaw angle; Y Ufl 、Y Ufr 、Y Url 、Y Urr and Y Lfl 、Y Lfr 、Y Lrl 、Y Lrr They are the upper and lower boundary values ​​of the obstacle avoidance constraints corresponding to the four vertices of the vehicle.

[0147] The objective function quantitatively characterizes obstacle avoidance safety, vehicle stability, and control smoothness by softly constraining the vehicle state, control quantity, and control increment within the prediction time domain. The control quantity corresponding to the minimum value of the objective function is the optimal control quantity.

[0148] The vehicle state variables within the prediction time domain include vehicle trajectory information and vehicle stability information. Vehicle trajectory information includes longitudinal velocity, lateral position, and yaw angle, and obstacle avoidance safety is improved by keeping the values ​​as close as possible to the expected values ​​obtained based on the obstacle avoidance guidance trajectory. Vehicle stability information includes yaw rate and sideslip angle, and vehicle stability is enhanced by constructing a stability factor objective function in conjunction with stability envelope constraints.

[0149] The optimization solution uses a sequential quadratic programming solver and sets an iterative warm start to obtain the optimal control sequence in the control time domain, and its first item is used as the actual output optimized control quantity;

[0150] Sequential quadratic programming is used to solve the integrated nonlinear optimization problem, and the obtained optimized solution sequence U *(t-1) moves forward one step and serves as the initial solution U(t) for the next time step initial , thus forming an iterative warm restart;

[0151]

[0152] The controlled variables include the front wheel steering angle and the four wheel longitudinal forces, wherein the wheel longitudinal forces include the left front wheel longitudinal force, the right front wheel longitudinal force, the left rear wheel longitudinal force, and the right rear wheel longitudinal force.

[0153] Figure 1-4 For the double lane-changing test under different initial vehicle speeds and road adhesion conditions, the scheme of Example 1 was compared with the control groups with no stability control and traditional stability control. In the optimization problem without stability control, the objective function only considered the obstacle avoidance guidance trajectory tracking accuracy term, the control amount, and the control amount increment term, and the constraint condition only considered the execution boundary constraint.

[0154] V in medium speed and medium attachment scene x =20m / sμ=0.5; Example 1, no stability control, and traditional stability control can all achieve good obstacle avoidance effects. Without stability control, the vehicle state exceeds the stability envelope constraint boundary range and the risk of vehicle instability increases. Compared with traditional stability control, Example 1 can effectively promote the vehicle state to converge within the stability envelope constraint by adding a stability factor term in the objective function. The obstacle avoidance trajectory and stability control effect under this working condition are shown in the figure. Figure 1 and Figure 2 As shown;

[0155] V in medium speed and low attachment scene x =20m / sμ=0.3; the road adhesion coefficient decreases, resulting in a decrease in tire stability margin. Without stability control, the vehicle becomes significantly unstable and exceeds the road boundary; under traditional stability control, although the vehicle avoids obstacles, the actual state of the vehicle exceeds the steady-state envelope constraint boundary, resulting in an increased risk of instability; under the control of Example 1, the vehicle can avoid obstacles safely, and by adding a stability factor term to the objective function, it effectively promotes the vehicle state to always remain within the stable envelope constraint. The obstacle avoidance effect and stability control effect under this working condition are as follows: Figure 3 and Figure 4 As shown;

[0156] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention herein is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the inventive concept. For example, the above-mentioned features may have similar functions to (but not limited to) those disclosed in this application.

Claims

1. A method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle, characterized in that: It includes obstacle avoidance guidance trajectory planning layer and obstacle avoidance safety and stability coordination control layer; The obstacle avoidance guidance trajectory planning layer considers the actual obstacle avoidance conditions to generate a guidance trajectory and determines the expected values ​​of the vehicle's longitudinal speed, lateral position, and yaw angle during the obstacle avoidance process. The obstacle avoidance safety and stability coordination control layer establishes an optimization problem based on the model predictive control framework, and the obstacle avoidance safety and stability coordination control layer includes a prediction model, constraints and objective functions, and an optimization solution link; The prediction model is established based on a three-degree-of-freedom vehicle dynamics model; the prediction model obtains the vehicle state change in the predicted time domain based on the current state and control input, and assists in determining the optimized control quantity in combination with the constraint conditions and the objective function; The constraints include obstacle avoidance boundary constraints, stability envelope constraints, and execution boundary constraints; The obstacle avoidance boundary constraint is a constraint on the vehicle's drivable area; the stability envelope constraint includes a yaw rate stability boundary constraint and a center of mass sideslip angle stability boundary constraint; the execution boundary constraint includes a control quantity constraint condition and a control quantity increment constraint condition; the optimized control quantity is selected based on the constraints to enhance obstacle avoidance safety, vehicle stability, and control smoothness; The objective function includes an obstacle avoidance trajectory tracking accuracy term, a stability factor term, a control amount term, and a control amount increment term; The objective function quantitatively characterizes obstacle avoidance safety, vehicle stability, and control smoothness by softly constraining the vehicle state, control quantity, and control increment within the prediction time domain; the control quantity corresponding to the minimum value of the objective function is the optimal control quantity; The vehicle state quantities in the predicted time domain include vehicle trajectory information and vehicle stability state information; the vehicle trajectory information includes longitudinal velocity, lateral position, and yaw angle, and obstacle avoidance safety is improved by approaching the expected value obtained based on the obstacle avoidance guidance trajectory; the vehicle stability state information includes yaw angular velocity and center of mass sideslip angle, and vehicle stability is enhanced by constructing a stability factor project objective function in combination with a stability envelope constraint; The optimization solution uses a sequential quadratic programming solver and sets an iterative warm start to obtain the optimal control sequence in the control time domain, and its first item is used as the actual output optimized control quantity; The control variables include the front wheel angle and the four wheel longitudinal forces, wherein the wheel longitudinal forces include the left front wheel longitudinal force, the right front wheel longitudinal force, the left rear wheel longitudinal force, and the right rear wheel longitudinal force; The yaw rate stability boundary constraint is a stability boundary value for the vehicle's yaw rate w determined based on the road adhesion condition and the vehicle's speed. The yaw rate stability boundary constraint expression is: oh min ≤w≤ω max ; The ω max is the upper limit of the yaw angular velocity, the ω min is the lower limit of yaw rate, V x is the longitudinal velocity of the vehicle, μ is the road adhesion coefficient, and g is the acceleration due to gravity; The center of mass sideslip angle stability boundary constraint determines the stability boundary value of the center of mass sideslip angle β based on the road adhesion condition and tire saturation characteristics. The center of mass sideslip angle stability boundary constraint expression is: β min ≤β≤β max ； The β max is the upper boundary of the center of mass sideslip angle, and the β min is the lower boundary of the center of mass side slip angle, b is the longitudinal distance between the center of mass of the vehicle and the rear axle, α sat is the tire saturation slip angle.

2. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: The control quantity constraint expression is: in min ≤u(k+i)≤u max ; The control quantity increment constraint expression is: Δu min ≤Δu(k+i)≤Δu max ; The control variables mainly include the front wheel angle δ f and the four wheel longitudinal forces F xfl 、F xfr 、F xrl 、F xrr ;u max ,u min are the maximum and minimum values ​​of the control quantity, respectively, which are determined by the physical limit boundary of the actuator; Δu max ,Δu min are the maximum and minimum values ​​of the control quantity increment respectively; The obstacle avoidance constraint requires that the vehicle always travels within the feasible area during the obstacle avoidance process, that is, the positions of the four vertices of the vehicle are always within the safety boundary constraint range. Its expression is: Y fl 、Y fr 、Y rl 、Y rr Y is the predicted value of the lateral position of the four vertices of the vehicle, o is the lateral position of the vehicle's center of mass, a is the longitudinal distance between the vehicle's center of mass and the front axle, b is the longitudinal distance between the vehicle's center of mass and the rear axle, d is the vehicle's wheelbase, and ψ is the vehicle's yaw angle; Y Ufl 、Y Ufr 、Y Url 、Y Urr and Y Lfl 、Y Lfr 、Y Lrl 、Y Lrr They are the upper and lower boundary values ​​of the obstacle avoidance constraints corresponding to the four vertices of the vehicle.

3. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: The objective function expression is: k is the sampling time; N p is the prediction time domain; N c To control the time domain; the first term of the objective function is the obstacle avoidance trajectory tracking accuracy term, and the first term of the objective function y p (k+i|k) is the predicted value of vehicle trajectory information in the prediction time domain, and the first term of the objective function y ref (k+i|k) is the expected value of the vehicle trajectory information in the predicted time domain determined based on the obstacle avoidance guidance trajectory; the second term u(k+i) is the control quantity term, the third term Δu(k+i) is the control quantity increment term; the fourth term S index (k+i|k) is the stability factor term; Q, R, S, and P are four corresponding weight matrices; Q is the weight matrix of the obstacle avoidance trajectory tracking accuracy term. The weight matrix Q includes the longitudinal velocity tracking bias weight, the vehicle lateral position tracking bias weight, and the vehicle yaw angle tracking bias weight. The longitudinal velocity tracking bias weight value is 2000-5000; the lateral displacement tracking bias weight value is 5000-8000; and the vehicle yaw angle tracking bias weight value is 5000-8000. R is the weight matrix of the control quantity term, which includes the front wheel angle weight and the wheel longitudinal force weight. The front wheel angle weight value is 50,000-80,000; the wheel longitudinal force weight value is 10,000-20,000. S is the weight matrix of the control quantity increment term, which includes the weight of the front wheel angle change rate and the weight of the wheel longitudinal force change rate; The weight value of the front wheel angle change rate is 50,000-80,000; the weight value of the wheel longitudinal force change rate is 10,000-20,000; P is the weight of the stability factor item, and the weight value of the stability factor item is 50-500.

4. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 3, characterized in that: Determine the stability factor S based on the relative distance between the vehicle's expected state and the stability envelope index To quantitatively characterize the vehicle's stable state, the expression is:

5. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: The prediction model includes the following formula: in, δ f is the front wheel angle, F xfl 、F xfr 、F xrl 、F xrr is the longitudinal force corresponding to the four wheels, namely the left front wheel, right front wheel, left rear wheel and right rear wheel, F yfl 、F yfr 、F yrl 、F yrr They are the lateral force of the left front wheel, the lateral force of the right front wheel, the lateral force of the left rear wheel, and the lateral force of the right rear wheel, respectively. drag is the longitudinal resistance, ΔM z is the additional yaw moment, a is the longitudinal distance between the vehicle's center of mass and the front axle, b is the longitudinal distance between the vehicle's center of mass and the rear axle, d is the vehicle's wheelbase, and I z is the moment of inertia of the vehicle around the z axis, m is the mass of the vehicle, v x is the longitudinal velocity, v y is the lateral velocity, ψ is the vehicle yaw angle, w is the vehicle yaw angular velocity, is the longitudinal velocity change rate, is the lateral velocity change rate, is the longitudinal velocity in geodetic coordinates, is the geodetic coordinate lateral velocity.

6. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: The steps of the obstacle avoidance guidance trajectory planning layer to generate the guidance trajectory and determine the expected value of the vehicle trajectory information are as follows: The vehicle obstacle avoidance guidance trajectory is the moving position reference value of the vehicle's center of mass (X(t), Y(t)) during the obstacle avoidance process; Get the vehicle longitudinal velocity V x , size information, safety threshold information; the size information includes the vehicle width W v , the distance d between the center of mass of the vehicle and the front end f , obstacle lateral boundary position Y c ; The safety threshold information includes the maximum lateral acceleration a ymax , a ymax =0.4*μ*g, where μ is the road adhesion coefficient and g is the acceleration due to gravity; The vehicle obstacle avoidance guidance trajectory formula is established based on the fifth-order polynomial, and the safety threshold information is substituted into the obstacle avoidance guidance trajectory formula to obtain the obstacle avoidance guidance trajectory (X ref ,Y ref ); Combined with the obstacle avoidance guidance trajectory information, the vehicle yaw angle ψ is obtained by formula 1 ref Expected value; The formula 1 is: k is the sampling time; ΔY ref (k) is the change in lateral position at adjacent moments; ΔX ref (k) is the change in longitudinal position between adjacent moments.

7. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: The obstacle avoidance guidance trajectory formula is: Y(t)=a5*t 5 +a4*t 4 +a3*t 3 +a2*t 2 +a1*t+a0; Y w =Y c +Δd ys -Y0; a0, a1, a2, a3, a4, a5 are the obstacle avoidance trajectory curve fitting coefficients, Δd ys is the lateral safety distance, Y w The lateral displacement required to avoid obstacles, Y c is the lateral boundary position of the obstacle, Δd ys is the lateral safety distance, and Y0 is the initial lateral displacement for obstacle avoidance.

8. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: The obstacle avoidance guidance trajectory is: X(t) is the longitudinal position of the center of mass of the vehicle at time t during the obstacle avoidance process, and Y(t) is the lateral position of the center of mass of the vehicle at time t during the obstacle avoidance process; t f is the obstacle avoidance time, and the calculation formula is as follows Y w =Y c +Δd ys -Y 0； 9. The method for coordinated control of obstacle avoidance and stability of a distributed drive vehicle according to claim 1, characterized in that: Sequential quadratic programming is used to solve the integrated nonlinear optimization problem, and the obtained optimized solution sequence U * (t-1) moves forward one step and serves as the initial solution U(t) for the next time step initial , thus forming an iterative warm restart;

Images