A drive anti-skid control method for a variable-structure three-axis unmanned vehicle in outdoor environments

Abstract

The present invention provides a drive anti-skid control method for a variable-structure three-axle unmanned vehicle in a field environment, characterized in that the method comprises the following steps: Step 1: collecting sensor information and determining the vehicle driving mode based on the sensor information; Step 2: selecting the optimal slip rate parameter under each driving mode based on the vehicle driving mode; Step 3: constructing a multi-agent architecture under each driving mode based on the vehicle driving mode and the optimal slip rate under each driving mode; Step 4: designing a slip rate tracking controller and controlling the slip rate and torque of the wheel based on the slip rate tracking controller. The present invention is conducive to adaptively correcting the slip rate control strategy of the variable-structure three-axle unmanned vehicle when the environment changes, so as to meet the requirements of safe and stable driving.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control, and in particular to a drive anti-skid control method for a variable-structure three-axis unmanned vehicle in a field environment. Background Art

[0002] Compared with fuel vehicles, electric vehicles can effectively save energy and reduce harm to the environment; with the development of motors and motor control technology, an electric vehicle that uses a hub or wheel-side motor directly or indirectly connected to the wheel has been proposed, also known as a distributed drive vehicle; distributed drive vehicles eliminate the mechanical transmission structure, improve the mechanical transmission efficiency, and the saved space makes the chassis layout more flexible and changeable. At the same time, the torque of each wheel is independently controllable, making the control of the entire vehicle more flexible and reliable. It has huge development potential in terms of safety, maneuverability and stability. Therefore, ground unmanned platforms often adopt a distributed drive control architecture; distributed drive vehicles have high requirements for slip rate control when performing tasks, because slip rate control is closely related to the safety and stability of autonomous driving; many slip rate control methods have been proposed, but most of them are aimed at paved road environments, and there is no corresponding slip rate control method for outdoor soft soil environments. Summary of the Invention

[0003] The purpose of the present invention is to address the current deficiencies and propose a variable structure three-axis unmanned vehicle drive anti-skid control method for field environments.

[0004] The present invention adopts the following technical solutions:

[0005] A variable-structure three-axis unmanned vehicle drive anti-skid control method for outdoor environments, characterized in that the method comprises the following steps:

[0006] Step 1: Collect sensor information and determine the vehicle driving mode based on the sensor information;

[0007] Step 2: Select the optimal slip ratio parameters for each driving mode according to the vehicle's driving mode;

[0008] Step 3: Based on the vehicle's driving mode and the optimal slip ratio under each driving mode, a multi-agent architecture is constructed for each driving mode.

[0009] Step 4: Design a slip tracking controller to control the slip rate and torque of the wheel.

[0010] Furthermore, the determination of the vehicle driving mode in step 1 is to determine the vehicle driving mode based on information obtained by the visual sensor and the suspension sensor; specifically comprising: determining the vehicle driving environment based on images obtained by the visual sensor, the vehicle driving environment including paved roads, sandy soil environments, and clay environments; determining whether the vehicle is in a four-wheel or six-wheel driving mode based on the retraction and extension of the intermediate shaft suspension obtained by the suspension sensor; combining the information obtained by the visual sensor and the information obtained by the suspension sensor to classify the vehicle driving mode into four driving modes, the four driving modes being a four-wheel driving mode on paved roads, a six-wheel driving mode on paved roads, a six-wheel driving mode in a sandy soil environment, and a six-wheel driving mode in a clay environment;

[0011] Furthermore, in step 2, different methods are used to obtain the optimal slip ratio parameter under different driving modes. If the vehicle's driving mode is a paved road four-wheel driving mode or a paved road six-wheel driving mode, the optimal slip ratio is obtained using the following magic tire formula:

[0012] Y(x)=Dsin{Carctan[Bx-E(Bx-arctan(Bx)]};

[0013] Where x is the slip ratio, Y(x) is the longitudinal force corresponding to the slip ratio x, B, C, D, and E are curve fitting factors, whose values are determined by the road surface type and the vertical load of the tire;

[0014] In the above formula, the slip ratio is input from 0 to 1 at intervals of 0.05 to obtain a curve showing the longitudinal force changing with the slip ratio. The highest point of the curve, that is, the slip ratio corresponding to the maximum longitudinal force, is selected as the optimal slip ratio output.

[0015] If the vehicle is in the six-wheel driving mode in a sandy environment or in the six-wheel driving mode in a clay environment, the optimal slip ratio is obtained by the following methods:

[0016] Build a ground mechanics model:

[0017]

[0018]

[0019] Among them, σ(θ) is the normal stress, k c is the soil cohesive modulus, b is the wheel width, is the soil deformation modulus, z(θ) is the settlement amount, n is the settlement coefficient; τ(θ) is the shear stress, c is the soil cohesion, is the internal friction angle, j(θ) is the tangential displacement, and K is the shear modulus; the soil cohesion modulus and soil deformation modulus are determined by the soil type; the settlement amount is the height difference between the lowest point of the tire when the vehicle is stationary and the vehicle when it is unloaded, and its value is determined by the vertical load; the settlement coefficient is the ratio of the settlement under unit load to the load, and its magnitude varies with the slip rate, and its value is obtained through experiments; the soil cohesion is determined by the soil type and the vertical load; the internal friction angle is the mutual friction angle between the particles inside the soil, and its magnitude varies with the slip rate, and its value is obtained through experiments; the tangential displacement is the relative displacement of the wheel along the road surface during driving, and its magnitude varies with the slip rate, and its value is obtained through experiments; the shear modulus is the deformation capacity of the soil when it is subjected to shear stress, and its magnitude varies with the slip rate, and its value is obtained through experiments;

[0020] Decompose the normal stress and shear stress in the direction of wheel movement to obtain two force components in that direction. Combine these two force components to get the longitudinal force on the wheel, also known as traction.

[0021] When the driving mode is the six-wheel driving mode in a sandy environment, the slip ratio is input from 0 to 1 at intervals of 0.05. The parameters affected by the slip ratio are substituted into the model to obtain a curve showing the longitudinal force changing with the slip ratio. The slip ratio when the longitudinal force reaches 1000N is selected as the optimal slip ratio output.

[0022] When the driving mode is the six-wheel driving mode in a clay environment, the slip ratio is input from 0 to 1 at intervals of 0.05. The parameters affected by the slip ratio are substituted into the model to obtain a curve showing the longitudinal force changing with the slip ratio. The highest point of the curve is extracted, i.e., the slip ratio corresponding to the maximum longitudinal force. It is then determined whether the slip ratio at this time is greater than 0.6. If the slip ratio at this time is greater than 0.6, 0.6 is output as the optimal slip ratio. If the slip ratio at this time is not greater than 0.6, the slip ratio at this time is output as the optimal slip ratio.

[0023] Furthermore, step 3 constructs a multi-agent architecture for the corresponding driving mode according to different driving modes; when the driving mode is a paved road four-wheel driving mode, the optimal slip rate under this driving mode is set as the virtual leader agent 0, and the four wheels of the front and rear axles are respectively set as connected agents 1, 2, 3, and 4. The difference in slip rates of adjacent agents and the difference in differential of slip rates are used as evaluation indicators and substituted into the next level of slip rate control; agents 0, 1, 2, 3, and 4 are constructed into a multi-agent architecture for the paved road four-wheel driving mode;

[0024] When the driving mode is a six-wheel driving mode on paved roads, the optimal slip rate in this driving mode is set as the virtual leader agent 0, and the six wheels on the front, center, and rear axles are respectively set as connected agents 1, 2, 3, 4, 5, and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level. Agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture for the six-wheel driving mode on paved roads.

[0025] When the driving mode is a six-wheel driving mode in a sandy soil environment, the optimal slip rate in this driving mode is used as the virtual leader agent, the front axle wheels are set as connected agents 1 and 2, and the middle and rear axle wheels are set as connected agents 3, 4, 5, and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level; the agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture in the six-wheel driving mode in a sandy soil environment;

[0026] When the driving mode is a six-wheel driving mode in a clay environment, the optimal slip rate in this driving mode is used as the virtual leader agent, the front axle wheels are set as connected agents 1 and 2, the middle axle wheels are set as connected agents 3 and 4, and the rear axle wheels are set as connected agents 5 and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the next level of slip rate control. Agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture in the six-wheel driving mode in a clay environment.

[0027] Furthermore, the slip tracking controller in step 4 includes a model-free adaptive controller, a sliding mode controller, and a collective controller. The control formulas of the model-free adaptive controller, the sliding mode controller, and the collective controller are obtained as follows:

[0028] For a single wheel there exists:

[0029] e1(k)=λ a (k)-λ d (k);

[0030] Where e1(k) is the error between the actual slip rate of the intelligent agent where the wheel is located at time k and the optimal slip rate, λ a (k) is the actual slip rate at time k, λ d (k) is the optimal slip ratio at time k;

[0031] e2(k)=(λa1 (k)-λ a2 (k))+(λ a1 (k)-λ a3 (k))+…+(λ a1 (k)-λ an (k));

[0032] Where e2(k) is the difference in slip rates of the adjacent agents of the wheel at time k, λ a1 (k) is the actual slip rate of the intelligent agent where the wheel is located at time k, λ ai (k) is the slip rate of the agent connected to the agent where the wheel is located at time k, satisfying 2≤i≤n, where n is the number of agents connected to the wheel;

[0033]

[0034] Among them, e3(k) is the time derivative of the difference in slip rates of the adjacent intelligent bodies of the wheel at time k, is the time derivative of the slip rate of the intelligent body where the wheel is located at time k, is the time derivative of the slip rate of the agent connected to the agent where the wheel is located at time k, satisfying 2≤i≤n, where n is the number of agents connected to the wheel;

[0035] Establish a discrete-time nonlinear system that satisfies the following equation:

[0036] y(k+1)=f[y(k),y(k-1),...,y(kn y ),u(k),u(k-1),...,u(kn u )]; (1)

[0037] Among them, y(k) is the output of the system at time k, u(k) is the input of the system at time k, y(k+1) is the output of the system at the next time k, and n y and n u are two unknown positive integers, f(...) is an unknown nonlinear function;

[0038] When the system has continuous partial derivatives in f(...) and formula (1) satisfies the generalized Lipschiz condition, there are pseudo partial derivatives An equivalent data model is obtained through formula (1):

[0039]

[0040] Where Δy(k+1) represents the output change at the next moment after time k, satisfying Δy(k+1)=y(k+1)-y(k); Δμ(k) represents the input change at time k, satisfying Δμ(k)=μ(k)-μ(k-1);

[0041] For system input, consider the following estimation criteria:

[0042] J(u(k))=|y * (k+1)-y(k+1)| 2 +λ|u(k)-u(k-1)| 2 ; (3)

[0043] Among them, y * (k+1) is the expected output signal at the next moment after time k, and λ is the weight factor used to penalize the change of the transition control input;

[0044] Substituting equation (2) into equation (3), taking the derivative of u(k) and setting it equal to 0, we can obtain the control formula of the model-free adaptive controller:

[0045]

[0046] Where ρ is the step size factor, satisfying 0<ρ≤1;

[0047] For pseudo partial derivatives, consider the following estimation criteria:

[0048]

[0049] Regarding formula (5) Find the extreme value and get the updated model of pseudo partial derivative:

[0050]

[0051] in, represents the pseudo partial derivative at time k, represents the pseudo partial derivative at time k-1, η is the step size factor, and its value range is (0, 1); Δμ(k-1) represents the input change at the moment before time k, Δy(k) represents the output change at time k, and satisfies Δy(k)=y(k)-y(k-1); y(k-1) represents the output at the moment before time k; μ is the time-varying weight factor, and satisfies u>0; For formula (6), if or |Δμ(k-1)|≤ε or When is the initial value of the pseudo partial derivative, and ε is a very small positive number;

[0052] The control formula of the sliding mode controller is:

[0053] s(k)=e1(k)+e2(k)+e3(k); (7)

[0054] s(k+1)-s(k)=-α1×s(k)-α2×sign[s(k)]; (8)

[0055] Where s(k+1) is the difference between the actual value and the expected value in the sliding mode control at the next moment at time k, s(k) is the difference between the actual value and the expected value in the sliding mode control at time k, α1 and α2 are the sliding mode control parameters, satisfying 0<α1<1, α2>0;

[0056] The collective controller is obtained by fusing the model-free adaptive controller and the sliding mode controller to obtain the collective control formula:

[0057]

[0058] The beneficial effects achieved by the present invention are:

[0059] The present invention adopts different methods to obtain the optimal wheel slip rate according to different driving modes, establishes corresponding analysis models according to different road surface information, and ensures the accuracy of optimal slip rate selection; by taking multi-agent factors into account in model-free adaptive sliding mode control, nonlinear interference is reduced through data driving, the response speed of the system is increased, and multiple wheels work together, overcoming the characteristics of traditional control methods that are unable to adapt to variable structures and changing environments, thereby realizing coordinated control of the slip rate of each wheel under multi-mode and multi-working conditions of a variable-structure three-axle unmanned vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but rather the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0061] Figure 1 Schematic diagram of the control method of the present invention.

[0062] Figure 2 This is a schematic diagram of the driving mode judgment of the present invention.

[0063] Figure 3 Schematic diagram of the optimal slip rate selection in the four-wheel and six-wheel driving modes on paved roads of the present invention.

[0064] Figure 4 Schematic diagram of the ground mechanics model of the present invention.

[0065] Figure 5 This is a schematic diagram of the optimal slip rate selection in the six-wheel driving mode in a sandy environment of the present invention.

[0066] Figure 6 This is a schematic diagram of the optimal slip rate selection in the six-wheel driving mode in a clay environment of the present invention.

[0067] Figure 7 Schematic diagram of the multi-agent architecture in the four-wheel driving mode on paved roads of the present invention.

[0068] Figure 8 Schematic diagram of the multi-agent architecture in the six-wheel driving mode on paved roads of the present invention.

[0069] Figure 9 Schematic diagram of the multi-agent architecture in the six-wheel driving mode on sandy roads of the present invention.

[0070] Figure 10 Schematic diagram of the multi-agent architecture in the six-wheel driving mode on clay roads of the present invention.

[0071] Figure 11 This is a control block diagram of the present invention in the four-wheel driving mode on paved roads.

[0072] Figure 12 This is a control block diagram of the six-wheel driving mode on paved roads according to the present invention.

[0073] Figure 13 This is a control block diagram of the six-wheel driving mode on sandy roads of the present invention.

[0074] Figure 14 This is a control block diagram of the six-wheel driving mode on clay roads according to the present invention. DETAILED DESCRIPTION

[0075] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with its embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention; for those skilled in the art, after reviewing the following detailed description, other systems, methods and / or features of the present embodiment will become apparent; it is intended that all such additional systems, methods, features and advantages are included in this specification; included within the scope of the present invention and protected by the appended claims; additional features of the disclosed embodiments are described in the following detailed description, and these features will be apparent from the following detailed description.

[0076] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", etc. indicating directions or positional relationships, they are based on the directions or positional relationships shown in the drawings. This is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or component referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0077] Example 1.

[0078] like Figure 1 As shown, this embodiment provides a variable-structure three-axis unmanned vehicle drive anti-skid control method for field environments, the method comprising the following steps:

[0079] Step 1: Collect sensor information and determine the vehicle driving mode based on the sensor information;

[0080] Step 2: Select the optimal slip ratio parameters for each driving mode according to the vehicle's driving mode;

[0081] Step 3: Based on the vehicle's driving mode and the optimal slip ratio under each driving mode, a multi-agent architecture is constructed for each driving mode.

[0082] Step 4: Design a slip tracking controller to control the slip rate and torque of the wheel.

[0083] like Figure 2 As shown, the specific implementation of step 1 includes:

[0084] Determining the vehicle's driving environment based on images acquired by a visual sensor, including paved roads, sandy soil environments, and clay environments; determining whether the vehicle is in a four-wheel or six-wheel driving mode based on the retraction and extension of the intermediate axle suspension acquired by a suspension sensor; and combining the information acquired by the visual sensor with the information acquired by the suspension sensor to classify the vehicle's driving mode into four driving modes, namely, a four-wheel driving mode on paved roads, a six-wheel driving mode on paved roads, a six-wheel driving mode in a sandy soil environment, and a six-wheel driving mode in a clay environment;

[0085] like Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 As shown, the specific implementation of step 2 includes:

[0086] If the vehicle's driving mode is paved road four-wheel driving mode or paved road six-wheel driving mode, the optimal slip ratio is obtained using the following magic tire formula:

[0087] Y(x)=Dsin{Carctan[Bx-E(Bx-arctan(Bx)]};

[0088] Where x is the slip ratio, Y(x) is the longitudinal force corresponding to the slip ratio x, B, C, D, and E are curve fitting factors whose values are determined by the road surface type and the vertical load of the tire. In this embodiment, the vertical load is set to 4000N.

[0089] In the above formula, the slip ratio is input from 0 to 1 at intervals of 0.05 to obtain a curve showing the longitudinal force changing with the slip ratio. The highest point of the curve, that is, the slip ratio corresponding to the maximum longitudinal force, is selected as the optimal slip ratio output.

[0090] If the vehicle is in the six-wheel driving mode in a sandy environment or in the six-wheel driving mode in a clay environment, the optimal slip ratio is obtained by the following methods:

[0091] Build a ground mechanics model:

[0092]

[0093]

[0094] Among them, σ(θ) is the normal stress, k c is the soil cohesive modulus, b is the wheel width, is the soil deformation modulus, z(θ) is the settlement amount, n is the settlement coefficient; τ(θ) is the shear stress, c is the soil cohesion, is the internal friction angle, j(θ) is the tangential displacement, and K is the shear modulus; the soil cohesion modulus and soil deformation modulus are determined by the soil type; the settlement amount is the height difference between the lowest point of the tire when the vehicle is stationary and the vehicle when it is unloaded, and its value is determined by the vertical load; the settlement coefficient is the ratio of the settlement under unit load to the load, and its magnitude varies with the slip rate, and its value is obtained through experiments; the soil cohesion is determined by the soil type and the vertical load; the internal friction angle is the mutual friction angle between the particles inside the soil, and its magnitude varies with the slip rate, and its value is obtained through experiments; the tangential displacement is the relative displacement of the wheel along the road surface during driving, and its magnitude varies with the slip rate, and its value is obtained through experiments; the shear modulus is the deformation capacity of the soil when it is subjected to shear stress, and its magnitude varies with the slip rate, and its value is obtained through experiments;

[0095] Decompose the normal stress and shear stress in the direction of wheel movement to obtain two force components in that direction. Combine these two force components to get the longitudinal force on the wheel, also known as traction.

[0096] When the driving mode is the six-wheel driving mode in a sandy environment, the slip ratio is input from 0 to 1 at intervals of 0.05. The parameters affected by the slip ratio are substituted into the model to obtain a curve showing the longitudinal force changing with the slip ratio. The slip ratio when the longitudinal force reaches 1000N is selected as the optimal slip ratio output.

[0097] When the driving mode is the six-wheel driving mode in a clay environment, the slip ratio is input from 0 to 1 at intervals of 0.05. The parameters affected by the slip ratio are substituted into the model to obtain a curve showing the longitudinal force changing with the slip ratio. The highest point of the curve is extracted, i.e., the slip ratio corresponding to the maximum longitudinal force. It is then determined whether the slip ratio at this time is greater than 0.6. If the slip ratio at this time is greater than 0.6, 0.6 is output as the optimal slip ratio. If the slip ratio at this time is not greater than 0.6, the slip ratio at this time is output as the optimal slip ratio.

[0098] like Figure 7 、 Figure 8 、 Figure 9 、 Figure 10 As shown, the specific implementation of step 3 is:

[0099] When the driving mode is the paved road four-wheel driving mode, the optimal slip rate in this driving mode is set as the virtual leader agent 0, and the four wheels on the front and rear axles are respectively set as the connected agents 1, 2, 3, and 4. The difference in the slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level. Agents 0, 1, 2, 3, and 4 form a multi-agent architecture for the paved road four-wheel driving mode.

[0100] When the driving mode is a six-wheel driving mode on paved roads, the optimal slip rate in this driving mode is set as the virtual leader agent 0, and the six wheels on the front, center, and rear axles are respectively set as connected agents 1, 2, 3, 4, 5, and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level. Agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture for the six-wheel driving mode on paved roads.

[0101] When the driving mode is a six-wheel driving mode in a sandy soil environment, the optimal slip rate in this driving mode is used as the virtual leader agent, the front axle wheels are set as connected agents 1 and 2, and the middle and rear axle wheels are set as connected agents 3, 4, 5, and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level; the agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture in the six-wheel driving mode in a sandy soil environment;

[0102] When the driving mode is a six-wheel driving mode in a clay environment, the optimal slip rate in this driving mode is used as the virtual leader agent, the front axle wheels are set as connected agents 1 and 2, the middle axle wheels are set as connected agents 3 and 4, and the rear axle wheels are set as connected agents 5 and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the next level of slip rate control. Agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture in the six-wheel driving mode in a clay environment.

[0103] The specific implementation of step 4 is:

[0104] The slip tracking controller in step 4 includes a model-free adaptive controller, a sliding mode controller, and a collective controller. The control formulas of the model-free adaptive controller, the sliding mode controller, and the collective controller are obtained as follows:

[0105] For a single wheel there exists:

[0106] e1(k)=λ a (k)-λ d (k);

[0107] Where e1(k) is the error between the actual slip rate of the intelligent agent where the wheel is located at time k and the optimal slip rate, λ a (k) is the actual slip rate at time k, λ d (k) is the optimal slip ratio at time k;

[0108] e2(k)=(λ a1 (k)-λ a2 (k))+(λ a1 (k)-λ a3 (k))+…+(λ a1 (k)-λ an (k));

[0109] Where e2(k) is the difference in slip rates of the adjacent agents of the wheel at time k, λ a1(k) is the actual slip rate of the intelligent agent where the wheel is located at time k, λ ai (k) is the slip rate of the agent connected to the agent where the wheel is located at time k, satisfying 2≤i≤n, where n is the number of agents connected to the wheel;

[0110]

[0111] Among them, e3(k) is the time derivative of the difference in slip rates of the adjacent intelligent bodies of the wheel at time k, is the time derivative of the slip rate of the intelligent body where the wheel is located at time k, is the time derivative of the slip rate of the agent connected to the agent where the wheel is located at time k, satisfying 2≤i≤n, where n is the number of agents connected to the wheel;

[0112] Establish a discrete-time nonlinear system that satisfies the following equation:

[0113] y(k+1)=f[y(k),y(k-1),...,y(kn y ),u(k),u(k-1),...,u(kn u )]; (1)

[0114] Among them, y(k) is the output of the system at time k, u(k) is the input of the system at time k, y(k+1) is the output of the system at the next time k, and n y and n u are two unknown positive integers, f(…) is an unknown nonlinear function;

[0115] When the system has continuous partial derivatives in f(...) and formula (1) satisfies the generalized Lipschiz condition, there are pseudo partial derivatives An equivalent data model is obtained through formula (1):

[0116]

[0117] Where Δy(k+1) represents the output change at the next moment after time k, satisfying Δy(k+1)=y(k+1)-y(k); Δμ(k) represents the input change at time k, satisfying Δμ(k)=μ(k)-μ(k-1);

[0118] For system input, consider the following estimation criteria:

[0119] J(u(k))=|y * (k+1)-y(k+1)| 2 +λ|u(k)-u(k-1)| 2 ; (3)

[0120] Among them, y * (k+1) is the expected output signal at the next moment after time k, and λ is the weight factor used to penalize the change of the transition control input;

[0121] Substituting equation (2) into equation (3), taking the derivative of u(k) and setting it equal to 0, we can obtain the control formula of the model-free adaptive controller:

[0122]

[0123] Where ρ is the step size factor, satisfying 0<ρ≤1;

[0124] For pseudo partial derivatives, consider the following estimation criteria:

[0125]

[0126] Regarding formula (5) Find the extreme value and get the updated model of pseudo partial derivative:

[0127]

[0128] in, represents the pseudo partial derivative at time k, represents the pseudo partial derivative at time k-1, η is the step size factor, and its value range is (0, 1); Δμ(k-1) represents the input change at the moment before time k, Δy(k) represents the output change at time k, and satisfies Δy(k)=y(k)-y(k-1); y(k-1) represents the output at the moment before time k; μ is the time-varying weight factor, and satisfies u>0; For formula (6), if or |Δμ(k-1)|≤ε or When is the initial value of the pseudo partial derivative, and ε is a very small positive number;

[0129] The control formula of the sliding mode controller is:

[0130] s(k)=e1(k)+e2(k)+e3(k); (7)

[0131] s(k+1)-s(k)=-α1×s(k)-α2×sign[s(k)]; (8)

[0132] Where s(k+1) is the difference between the actual value and the expected value in the sliding mode control at the next moment at time k, s(k) is the difference between the actual value and the expected value in the sliding mode control at time k, α1 and α2 are the sliding mode control parameters, satisfying 0<α1<1, α2>0;

[0133] The collective controller is obtained by fusing the model-free adaptive controller and the sliding mode controller to obtain the collective control formula:

[0134]

[0135] This embodiment obtains the optimal wheel slip rate in different ways according to different driving modes, establishes corresponding analysis models according to different road surface information, and ensures the accuracy of the optimal slip rate selection; constructs multi-agent architectures in different situations according to different road surface types. For example, when four wheels are driving on paved roads, they have the advantages of easy control and energy saving. At this time, the number of multi-agents connected in the architecture is 4. When six wheels are driving on paved roads, the number of multi-agents connected in the multi-agent architecture is 6. However, on sandy roads, there is a small amount of subsidence at this time, and the soil particles do not undergo plastic deformation. The vehicle only compacts the soil when driving. Therefore, priority is given to ensuring the slip rate of the front wheels, because the front wheels will compact the sand and soil, and the settlement and deformation caused by the rear wheels are small. Therefore, the number of multi-agents connected in the multi-agent architecture is 2+4. On clay roads, due to the plastic deformation of soil particles, the settlement is large. Even if the front wheels pass, the rear wheels will still be compressed by a large amount. Therefore, the number of multi-agents connected in the multi-agent architecture is 2+2+2; the control method of combining the multi-agent architecture and model-free adaptive sliding mode control to control the slip rate of four or six wheels can achieve the control target and has good tracking characteristics and robustness.

[0136] Example 2.

[0137] This embodiment should be understood to include at least all the features of any of the aforementioned embodiments and be further improved thereon;

[0138] This embodiment provides a variable-structure three-axle unmanned vehicle anti-slip control method for outdoor environments. The specific control methods for controlling the slip rate and torque of the wheels according to the slip rate tracking controller include:

[0139] like Figure 11 As shown in the figure, when the vehicle is in the four-wheel driving mode on paved roads, in order to reduce unnecessary energy loss, the intermediate shaft active suspension is retracted, and only the front and rear four wheels are driven. Because the road surface does not deform, the working conditions of the four wheels are basically the same. Taking the left front wheel as an example, the front left wheel is equivalent to the right front wheel, the left rear wheel, and the right rear wheel.

[0140] At this time, the difference between the optimal slip rate and the actual slip rate of the left front wheel is used as the input of the model-free adaptive controller, and the desired torque of the left front wheel is used as the output of the model-free adaptive control. The difference between the optimal slip rate and the actual slip rate of the left front wheel, as well as the difference between the left front wheel slip rate and the actual slip rates of the right front wheel, the left rear wheel, and the right rear wheel, and the corresponding time-derivative differences are used as the input of the sliding mode controller. The above two controllers are integrated, and the desired torque of the left front wheel is used as the input of the collective controller, and the desired slip rate of the left front wheel is used as the output to achieve control of the slip rate of this wheel.

[0141] like Figure 12 As shown, when the vehicle is in the six-wheel driving mode on paved roads, in order to obtain stronger driving force, the intermediate shaft active suspension is lowered, and the front, middle and rear axles are driven by six wheels. Because the road surface does not deform, the working conditions of the six wheels are basically the same. Taking the left front wheel as an example, the left front wheel is equivalent to the right front wheel, the left middle wheel, the right middle wheel, the left rear wheel and the right rear wheel.

[0142] At this time, the difference between the optimal slip rate and the actual slip rate of the left front wheel is used as the input of the model-free adaptive controller, and the desired torque of the left front wheel is used as the output of the model-free adaptive control; the difference between the optimal slip rate and the actual slip rate of the left front wheel, as well as the difference between the left front wheel slip rate and the actual slip rates of the right front wheel, left middle wheel, right middle wheel, left rear wheel and right rear wheel, and the corresponding time-derivative differences are used as the input of the sliding mode controller; the above two controllers are integrated, and the desired torque of the left front wheel is used as the input of the collective controller, and the desired slip rate of the left front wheel is used as the output to achieve control of the slip rate of this wheel;

[0143] like Figure 13 As shown in the figure, when the vehicle driving mode is the six-wheel driving mode in a sandy soil environment, the six-wheel mode is usually adopted on such soft soil roads to increase driving safety. At this time, the soil particles are sandy soil particles, which only undergo elastic deformation but not plastic deformation after being squeezed. Therefore, the soil becomes more compact after the front wheel passes by. When the middle wheel and the rear wheel pass by again, it can be regarded as the same driving condition. Taking the left front wheel as an example, the left front wheel is equivalent to the right front wheel at this time.

[0144] At this time, the difference between the optimal slip rate and the actual slip rate of the left front wheel is used as the input of the model-free adaptive controller, and the desired torque of the left front wheel is used as the output of the model-free adaptive controller. The difference between the optimal slip rate and the actual slip rate of the left front wheel, the difference between the slip rate of the left front wheel and the actual slip rate of the right front wheel, and the corresponding time-derivative difference are used as the input of the sliding mode controller. The above two controllers are integrated, and the desired torque of the left front wheel is used as the input of the collective controller, and the desired slip rate of the left front wheel is used as the output to achieve control of the slip rate of this wheel.

[0145] At the same time, taking the left middle wheel as an example, the left middle wheel is equivalent to the right middle wheel, the left rear wheel and the right rear wheel;

[0146] At this time, the difference between the optimal slip rate and the actual slip rate of the left middle wheel is used as the input of the model-free adaptive controller, and the desired torque of the left middle wheel is used as the output of the model-free adaptive control. The difference between the optimal slip rate and the actual slip rate of the left middle wheel, as well as the difference between the slip rate of the left middle wheel and the actual slip rates of the right middle wheel, the left rear wheel, and the right rear wheel, and the corresponding time-derivative differences are used as the input of the sliding mode controller. The above two controllers are integrated, and the desired torque of the left middle wheel is used as the input of the collective controller, and the desired slip rate of the left middle wheel is used as the output to achieve control of the slip rate of this wheel.

[0147] like Figure 14 As shown in the figure, when the vehicle driving mode is the six-wheel driving mode in a clay environment, the six-wheel mode is usually used on such clay roads to increase driving safety. At this time, the soil particles are clay particles, which will undergo plastic deformation after being squeezed. Therefore, after the front wheels pass by, the clay can still undergo significant deformation and sinking when the middle wheels and rear wheels pass by. Taking the left front wheel as an example, the left front wheel is equivalent to the right front wheel at this time.

[0148] At this time, the difference between the optimal slip rate and the actual slip rate of the left front wheel is used as the input of the model-free adaptive controller, and the desired torque of the left front wheel is used as the output of the model-free adaptive controller. The difference between the optimal slip rate and the actual slip rate of the left front wheel, the difference between the slip rate of the left front wheel and the actual slip rate of the right front wheel, and the corresponding time-derivative difference are used as the input of the sliding mode controller. The above two controllers are integrated, and the desired torque of the left front wheel is used as the input of the collective controller, and the desired slip rate of the left front wheel is used as the output to achieve control of the slip rate of this wheel.

[0149] At the same time, taking the left middle wheel as an example, the left middle wheel is equivalent to the right middle wheel;

[0150] At this time, the difference between the optimal slip rate and the actual slip rate of the left middle wheel is used as the input of the model-free adaptive controller, and the desired torque of the left middle wheel is used as the output of the model-free adaptive control. The difference between the optimal slip rate and the actual slip rate of the left middle wheel, the difference between the slip rate of the left middle wheel and the actual slip rate of the right middle wheel, and the corresponding time-derivative difference are used as the input of the sliding mode controller. The above two controllers are integrated, and the desired torque of the left middle wheel is used as the input of the collective controller, and the desired slip rate of the left middle wheel is used as the output to achieve control of the slip rate of this wheel.

[0151] At the same time, taking the left rear wheel as an example, the left rear wheel is equivalent to the right rear wheel;

[0152] At this time, the difference between the optimal slip rate and the actual slip rate of the left rear wheel is used as the input of the model-free adaptive controller, and the desired torque of the left rear wheel is used as the output of the model-free adaptive control; the difference between the optimal slip rate and the actual slip rate of the left rear wheel, the difference between the slip rate of the left rear wheel and the actual slip rate of the right rear wheel, and the corresponding time-derivative difference are used as the input of the sliding mode controller; the above two controllers are integrated, and the desired torque of the left rear wheel is used as the input of the collective controller, and the desired slip rate of the left rear wheel is used as the output to achieve control of the slip rate of this wheel.

[0153] This embodiment aims to track the optimal wheel slip rate and coordinate the control of the slip rate of each wheel. It does not require a precise vehicle model, uses a data-driven control algorithm, and adopts a control method based on multi-agent theory and model-free adaptive sliding mode control to control the vehicle. It is universal in the field of longitudinal control of vehicles and ground unmanned platforms, eliminates interference from factors such as nonlinearity and model-freeness, and improves response speed and agent coordination.

[0154] The contents disclosed above are only preferred feasible embodiments of the present invention and do not limit the scope of protection of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention description and drawings are included in the scope of protection of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. A variable-structure three-axis unmanned vehicle anti-skid control method for outdoor environments, characterized by: The method comprises the following steps: Step 1: Collect sensor information and determine the vehicle driving mode based on the sensor information; Step 2: Select the optimal slip ratio parameters for each driving mode according to the vehicle's driving mode; Step 3: Based on the vehicle's driving mode and the optimal slip ratio under each driving mode, a multi-agent architecture is constructed for each driving mode. Step 4: Design a slip tracking controller. The slip tracking controller includes a model-free adaptive controller, a sliding film controller, and a collective controller. The slip rate and torque of the wheel are controlled according to the slip tracking controller. The determination of the vehicle driving mode in the step 1 is to determine the vehicle driving mode based on the information obtained by the visual sensor and the suspension sensor; specifically comprising: determining the vehicle driving environment based on the image obtained by the visual sensor, the vehicle driving environment including paved roads, sandy soil environments and clay environments; determining whether the vehicle is in a four-wheel or six-wheel driving mode based on the retraction and extension of the intermediate shaft suspension obtained by the suspension sensor; combining the information obtained by the visual sensor with the information obtained by the suspension sensor to divide the vehicle driving mode into four driving modes, the four driving modes being a paved road four-wheel driving mode, a paved road six-wheel driving mode, a sandy soil environment six-wheel driving mode and a clay environment six-wheel driving mode; when the vehicle driving mode is a paved road four-wheel driving mode, retracting the intermediate shaft active suspension and only driving the front and rear four wheels; when the vehicle driving mode is a paved road six-wheel driving mode, a sandy soil environment six-wheel driving mode and a clay environment six-wheel driving mode, lowering the intermediate shaft active suspension and driving the front, middle and rear axles six wheels; In the paved road four-wheel driving mode, the difference between the optimal slip rate and the actual slip rate of the left front wheel is used as the input of the model-free adaptive controller, and the expected torque of the left front wheel is used as the output of the model-free adaptive control; the difference between the optimal slip rate and the actual slip rate of the left front wheel, the difference between the slip rate of the left front wheel and the actual slip rate of the right front wheel, the left rear wheel and the right rear wheel, and the derivative of the difference between the slip rate of the left front wheel and the actual slip rate of the right front wheel, the left rear wheel and the right rear wheel with respect to time are used as the input of the sliding mode controller; the above-mentioned model-free adaptive controller and the sliding mode controller are integrated to obtain a collective controller, the expected torque of the left front wheel is used as the input of the collective controller, and the expected slip rate of the left front wheel is used as the slip rate output of the left front wheel, the right front wheel, the left rear wheel and the right rear wheel; in the paved road six-wheel driving mode , the difference between the optimal slip rate and the actual slip rate of the left front wheel is used as the input of the model-free adaptive controller, and the desired torque of the left front wheel is used as the output of the model-free adaptive control; the difference between the optimal slip rate and the actual slip rate of the left front wheel, the difference between the slip rate of the left front wheel and the actual slip rates of the right front wheel, the left middle wheel, the right middle wheel, the left rear wheel and the right rear wheel, and the derivative of the difference between the slip rate of the left front wheel and the actual slip rates of the right front wheel, the left middle wheel, the right middle wheel, the left rear wheel and the right rear wheel with respect to time are used as the input of the sliding mode controller; the above-mentioned model-free adaptive controller and the sliding mode controller are fused to obtain a collective controller, the desired torque of the left front wheel is used as the input of the collective controller, and the desired slip rate of the left front wheel is used as the slip rate output of the left front wheel, the right front wheel, the left middle wheel, the right middle wheel, the left rear wheel and the right rear wheel.

2. A variable-structure three-axis unmanned vehicle anti-skid control method for outdoor environments according to claim 1, characterized in that: In step 2, different methods are used to obtain the optimal slip ratio parameter under different driving modes. If the vehicle's driving mode is a paved road four-wheel driving mode or a paved road six-wheel driving mode, the optimal slip ratio is obtained using the following magic tire formula: ； in, is the slip rate, The slip rate is The corresponding longitudinal force is 、 、 and is the curve fitting factor, and its value is determined by the road surface type and the vertical load of the tire; In the above formula, the slip ratio is input from 0 to 1 at intervals of 0.05 to obtain a curve showing the longitudinal force changing with the slip ratio. The highest point of the curve, that is, the slip ratio corresponding to the maximum longitudinal force, is selected as the optimal slip ratio output. If the vehicle is in six-wheel driving mode in a sandy or clayy environment, the optimal slip ratio is obtained by: Build a ground mechanics model: ； ； in, is the normal stress, is the soil cohesive modulus, is the wheel width, is the soil deformation modulus, is the amount of subsidence, is the subsidence coefficient; is the shear stress, is the soil cohesion, is the internal friction angle, is the tangential displacement, is the shear modulus; the soil cohesion modulus and soil deformation modulus are determined by the soil type; the settlement is the height difference between the lowest point of the tire when the vehicle is stationary and the vehicle when it is unloaded, and its value is determined by the vertical load; the settlement coefficient is the ratio of the settlement under unit load to the load, its size varies with the slip rate, and its value is obtained through experiments; the soil cohesion is determined by the soil type and the vertical load; the internal friction angle is the mutual friction angle between the particles inside the soil, its size varies with the slip rate, and its value is obtained through experiments; the tangential displacement is the relative displacement of the wheel along the road surface during driving, its size varies with the slip rate, and its value is obtained through experiments; the shear modulus is the deformation capacity of the soil when it is subjected to shear stress, its size varies with the slip rate, and its value is obtained through experiments; Decompose the normal stress and shear stress in the direction of wheel movement to obtain two force components in that direction. Combine these two force components to get the longitudinal force on the wheel, also known as traction. When the driving mode is the six-wheel driving mode in a sandy environment, the slip ratio is input from 0 to 1 at intervals of 0.

05. The parameters affected by the slip ratio are substituted into the model to obtain a curve showing the longitudinal force changing with the slip ratio. The slip ratio when the longitudinal force reaches 1000N is selected as the optimal slip ratio output. When the driving mode is the six-wheel driving mode in a clay environment, the slip rate is input from 0 to 1 at intervals of 0.

05. The parameters affected by the slip rate are substituted into the model to obtain a curve showing the change of longitudinal force with slip rate. The highest point of the curve, that is, the slip rate corresponding to the maximum longitudinal force, is extracted to determine whether the slip rate at this time is greater than 0.

6. If the slip rate at this time is greater than 0.6, 0.6 is output as the optimal slip rate. If the slip rate at this time is not greater than 0.6, the slip rate at this time is output as the optimal slip rate.

3. A variable-structure three-axis unmanned vehicle anti-slip control method for outdoor environments according to claim 2, characterized in that: Step 3 constructs a multi-agent architecture under different driving modes according to the different driving modes; When the driving mode is the paved road four-wheel driving mode, the optimal slip rate in this driving mode is set as the virtual leader agent 0, and the four wheels on the front and rear axles are respectively set as the connected agents 1, 2, 3, and 4. The difference in the slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level. Agents 0, 1, 2, 3, and 4 form a multi-agent architecture for the paved road four-wheel driving mode. When the driving mode is a six-wheel driving mode on paved roads, the optimal slip rate in this driving mode is set as the virtual leader agent 0, and the six wheels on the front, center, and rear axles are respectively set as connected agents 1, 2, 3, 4, 5, and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level. Agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture for the six-wheel driving mode on paved roads. When the driving mode is a six-wheel driving mode in a sandy soil environment, the optimal slip rate in this driving mode is used as the virtual leader agent, the front axle wheels are set as connected agents 1 and 2, and the middle and rear axle wheels are set as connected agents 3, 4, 5, and 6. The difference in slip rates of adjacent agents and the difference in the differential of the slip rates are used as evaluation indicators and substituted into the slip rate control of the next level; the agents 0, 1, 2, 3, 4, 5, and 6 form a multi-agent architecture in the six-wheel driving mode in a sandy soil environment; When the driving mode is the six-wheel driving mode in a clay environment, the optimal slip rate in this driving mode is used as the virtual leader agent, the front axle wheels are set to be connected to agents 1 and 2, the middle axle wheels are set to be connected to agents 3 and 4, and the rear axle wheels are set to be connected to agents 5 and 6. The difference in slip rates of adjacent agents and the difference in differentials of slip rates are used as evaluation indicators and substituted into the slip rate control of the next level; the agents 0, 1, 2, 3, 4, 5 and 6 form a multi-agent architecture in the six-wheel driving mode in a clay environment.

4. A variable structure three-axis unmanned vehicle anti-skid control method for outdoor environments according to claim 3, characterized in that: The control formulas of the model-free adaptive controller, sliding mode controller and collective controller in step 4 are obtained as follows: For a single wheel there exists: ； in, For The error between the actual slip rate of the intelligent body where the wheel is located and the optimal slip rate at that moment, For The actual slip rate at time , For The optimal slip ratio at the moment; ； in, For The difference in slip rates of adjacent agents on the wheel at the moment, For The actual slip rate of the agent where the wheel is located at the moment, For The slip rate of the agent connected to the agent where the wheel is located at any time satisfies 2 , is the number of agents connected to the wheel; ； in, For The time derivative of the difference in slip rates of the adjacent intelligent bodies at this wheel, For The time derivative of the slip rate of the intelligent body where the wheel is located at the moment, For The time derivative of the slip rate of the agent connected to the agent where the wheel is located satisfies 2 , is the number of agents connected to the wheel; Establish a discrete-time nonlinear system that satisfies the following equation: ； (1) in, for The output of the time system, for The input of the time system, for The output of the system at the next moment, and are two unknown positive integers, is an unknown nonlinear function; The system When there are continuous partial derivatives and formula (1) satisfies the generalized Lipschiz condition, there are pseudo partial derivatives An equivalent data model is obtained through formula (1): ； (2) in, express The output change at the next moment satisfies ; express The input change at the moment satisfies ; For system input, consider the following estimation criteria: ； (3) in, For The expected output signal at the next moment, is a weight factor used to penalize changes in the transition control input; Substituting formula (2) into formula (3), we can get Taking the derivative and setting it equal to 0, we get the control formula of the model-free adaptive controller: ； (4) in, is the step size factor, satisfying ; For pseudo partial derivatives, consider the following estimation criteria: ；(5) Regarding formula (5) Find the extreme value and get the updated model of pseudo partial derivative: ； (6) in, express The pseudo partial derivative at time , express The pseudo partial derivative at time , is the step size factor, and its value range is ; express The input change at the moment before the moment, express The output change at the moment satisfies ; express The output of the previous moment; is a time-varying weight factor, satisfying ; For formula (6), if or or When ; is the initial value of the pseudo partial derivative, is a very small positive number; The control formula of the sliding mode controller is: ; (7) ； (8) in For The difference between the actual value and the expected value in the sliding mode control at the next moment, For The difference between the actual value and the expected value in the sliding mode control at the moment, and is the sliding mode control parameter, satisfying , ; The collective controller is obtained by fusing the model-free adaptive controller and the sliding mode controller to obtain the collective control formula: 。

Images