Vehicle active suspension control method and system based on previewing ground surface features

Abstract

The present invention provides a vehicle active suspension control method based on previewed ground surface features, comprising the following steps: S1 extracts an elevation sequence from a previewed elevation map; S2 sets obstacle values; S3 determines whether active suspension usage conditions are met; if so, executes S4; if not, issues a warning and returns to S1; S4 discretizes the elevation sequence Ω containing the obstacle in S2 into a surface scale space #imgabs0#; S5 utilizes the elevation surface sequence to obtain target values ​​for active suspension control and control variables for the active suspension actuator. The present invention also provides a corresponding system, comprising: a GNSS / GPS combined inertial navigation system, a 3D ranging sensor, a vehicle status database, an elevation sequence acquisition module, an obstacle determination module, an active suspension activation module, an elevation surface sequence acquisition module, and an active suspension control module. Based on previewed terrain, the present invention ensures both safety and comfort when traversing uneven roads.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to a vehicle active suspension control method and system based on previewing ground surface features. Background Art

[0002] To improve driving comfort, many vehicles are equipped with active suspension components instead of passive ones. Active suspension components can adjust parameters based on the vehicle's driving state, effectively buffering impact loads transmitted from uneven roads to the frame or body, attenuating vertical vibrations of the vehicle body, and ensuring passenger comfort and vehicle handling stability. Active suspension control methods are primarily categorized into frequency domain and time domain. Frequency domain methods are currently in the simulation and optimization phase, while time domain methods, which are applicable in engineering, primarily focus on model predictive control.

[0003] The fundamental premise of active suspension control theory is the assumption that the vehicle and the road are in contact and non-slip. From a vehicle control perspective, behavioral models and controllers focus on complex dynamic models and state machines. At relatively high vehicle speeds, considering the time delay of mechanical actuator systems, effective control methods rely on preemptive control strategies. Limited to certain speed conditions and known spatial conditions of terrain and topography, currently known active suspension control methods rely on dynamic prediction of the vehicle's driving state and active suspension control based on the optimized driving state. The accuracy of vehicle state prediction is crucial. Because the dynamic response of vehicles varies at low and high speeds, complex dynamic models must account for the applicable range of vehicle-ground interaction at different speeds. However, constructing a complex and accurate state machine and solving the state vector are significant challenges in engineering applications. These methods are not applicable in high-speed driving situations with high timeliness requirements. The demands of application scenarios and the evolution of intelligent vehicle functional design are placing higher demands on speed optimization, steering, and active suspension control.

[0004] Real-time map generation technology is becoming increasingly mature. Maps contain a wealth of map data, especially elevation information in elevation maps, which clearly defines road conditions. Active suspension systems rely on road conditions to determine whether to activate the suspension. Combining elevation information with vehicle data would allow for more efficient and rapid control of the active suspension. However, a robust method for combining map elevation information with active suspension control is currently lacking. Summary of the Invention

[0005] In order to overcome the shortcomings of the existing technology, the present invention proposes a vehicle active suspension control method and system based on previewing the ground surface features, which adopts a practical and efficient method of using the predicted state vector of the state machine to approximate the optimal state set of engineering tests under the conditions of the current predicted terrain.

[0006] The present invention discloses a vehicle active suspension control method based on previewing ground surface features, and the specific steps are as follows:

[0007] S1: Extract elevation sequence from preview elevation map;

[0008] A preview elevation map centered on the vehicle body is generated using a GNSS / GPS combined inertial navigation system and a 3D ranging sensor. The elevation sequence Ω is extracted from the preview elevation map along the vehicle's travel direction.

[0009] S2: Set the value of the obstacle;

[0010] S21: extracting road surface features from the high-order sequence Ω obtained in S1 to obtain a road surface feature sequence L(s);

[0011] S22: By analyzing the obtained road feature sequence L(s), if there is an obstacle, save the obstacle R = {σ0; h k ; K}, where σ0 is the size of the obstacle, h k is the height of the obstacle, and K is the slope of the obstacle; otherwise, the value of obstacle R is empty;

[0012] S3: Determine whether the active suspension use conditions are met. If so, execute S4. If not, issue a warning and return to S1.

[0013] S4: Discretize the high-order sequence Ω containing obstacles in S2 into a surface scale space

[0014]

[0015] Surface scale space Contains the process of converting each elevation value in the elevation sequence Ω into its value. After equal division; The high-order sequence group composed of equal parts will be divided into The elevation sequence composed of equal parts is called elevation surface sequence, denoted by T j ;

[0016] S5: Initially set j = N, and use the elevation surface sequence to obtain the target value of the active suspension control and the control amount of the active suspension actuator, which specifically includes the following sub-steps:

[0017] S51: Elevation surface sequence Tj Perform cubic spline interpolation to obtain the elevation curve fj(x); obtain the elevation curve value sequence H={h x1 ,h x2 ,…,h xi ,…,h xn}, when the coordinate is x i When the corresponding elevation value is h xi ; When x i+1 is x i The next point coordinates, the corresponding elevation value is h x(i+1) ;

[0018] S52: Obtain elevation prediction value sequence using iteration

[0019] S53: Compare the elevation curve value sequence H obtained in S51 with the elevation prediction value sequence H obtained in S52 ～ , if the elevation prediction value sequence H ～ Each corresponding element in the elevation curve value sequence H satisfies Then execute S54; otherwise, set the value of j to j-1; if j is 0, execute S55; if j is not 0, return to S51;

[0020] S54: f j (x) is set as the target value of active suspension control, and the high-order sequence Ω and the current T j Subtract to obtain the control variable Γ of the active suspension actuator:

[0021]

[0022] S55: End the loop.

[0023] Preferably, the step S52 uses an iterative method to obtain a sequence of elevation prediction values. The specific iterative process is as follows:

[0024]

[0025]

[0026] Δθ ～ =ω i ×Δt+Δω×Δt (14)

[0027] Δh ～ =Δθ ～ ×L×cosθ i (15)

[0028]

[0029] Among them, the angular velocity change of the pitch angle is Δω, and the distance from the center of mass to the rear axle of the vehicle is l f , the mass of the vehicle is m, V is the vehicle speed; the moment of inertia of the vehicle around the rear axle is J M The angle between the straight line from the center of mass to the rear axle of the vehicle and the horizontal plane determined by the axle is δ, and the pitch angle of the vehicle body attitude angle is θ, θ i Represents x i The corresponding pitch angle, θ i+1 Represents x i+1 The corresponding pitch angle, Δθ represents θ i+1 and θ i The difference between the pitch angle and the angular velocity is ω, ω i Represents x i The angular velocity corresponding to the pitch angle; ω i+1 Represents x i+1 The angular velocity corresponding to the pitch angle; x i+1 is x i The coordinates of the next point; Δx represents x i+1 and x i Δt is the time required for the vehicle to pass Δx; Δh represents h x(i+1) and h xi The difference, Δh ～ express and The difference; h xi The coordinate x obtained from the elevation curve fj(x) is i The corresponding elevation value when h x(i+1) The coordinate x obtained from the elevation curve fj(x) is i+1 The corresponding elevation value when Δθ ～ , Δh ～ , hx (i+1) ～ All are predicted values, representing the predicted pitch angle increment predicted value, elevation increment predicted value, and elevation value predicted value respectively.

[0030] The parameters required for formulas (12) to (16) are obtained as follows:

[0031] cosδ=2l f / L (17)

[0032]

[0033]

[0034]

[0035]

[0036]

[0037] where k is the first derivative of the elevation curve fj(x) at x i+1 at that point.

[0038] Preferably, the elevation sequence Ω in S1 is specifically as follows:

[0039]

[0040]

[0041] In the formula: h i is the elevation value of the i-th cell; N SP is the data length of the elevation sequence; L p is the fixed preview length of the elevation map; Δl is the fixed interval of the elevation map; i is a positive integer.

[0042] Preferably, S21 extracts road surface features from the elevation sequence Ω to obtain a road surface feature sequence L(s); the specific process is as follows:

[0043] L(s) = [1, -2, 1] * G(s) * Ω (3)

[0044] where G(s) is a discrete value sequence used to smooth the original elevation sequence Ω and filter out high-frequency noise. The symbol '*' in formula (3) represents the convolution operation of discrete digital sequences.

[0045] Preferably, the specific process of S22 is as follows:

[0046] If the i-th element s i in the road surface feature sequence L(s) satisfies the following condition, then the i-th element s i is regarded as a feature point:

[0047] |L(s i )| > σ (5)

[0048] In the formula, the value of σ is an empirical value. There is a position correspondence relationship between the elements in the road surface feature sequence L(s) and the elevation sequence Ω. Therefore, when the p-th element s p and the q-th element s q in L(s) are greater than σ, then the p-th element and the q-th element in the elevation sequence Ω are regarded as ground inflection points. When the k-th element s k ​​​k , the slope of the obstacle Otherwise the value of obstacle R is empty.

[0049] Preferably, the specific judgment process of S3 is as follows:

[0050] S31: Determine whether there is an obstacle. If the value of obstacle R in S2 is not null, there is an obstacle, and proceed to S32. If the value of obstacle R in S2 is null, there is no obstacle, and return to S1 to continue driving without active suspension control.

[0051] S32: Determine the suspension limit travel H max Is it greater than the maximum height h of the obstacle? p , when the maximum height of the obstacle h p Less than the suspension height H max When the maximum height h of the obstacle is p Greater than or equal to suspension height H max When , a warning is given and the process returns to S1;

[0052] S33: Determine whether a grounding condition related to vehicle speed is met. When the vehicle speed grounding condition is met, continue to S4. When the wheel grounding condition is not met, give a warning and return to S1.

[0053] Preferably, the vehicle speed grounding condition in S33 is:

[0054] The square of the vehicle speed is less than the acceleration due to gravity multiplied by the wheelbase of the front and rear axles, multiplied by the rate of change of the obstacle's elevation; the rate of change of elevation is described as the square root of the sum of 1 plus the inverse of the square of the slope, expressed as r. a Indicates that r a The expression is:

[0055]

[0056] The vehicle speed grounding condition is expressed as the formula:

[0057] V 2 ≤g×r a ×L (9)

[0058] Where: V is the vehicle speed; L is the wheelbase between the front and rear axles; K is the slope of the obstacle; r a is the rate of change of elevation; g is the acceleration due to gravity.

[0059] The present invention also discloses a vehicle active suspension control system based on previewing ground surface features, and uses a vehicle active suspension control method based on previewing ground surface features, which specifically includes the following devices: a GNSS / GPS combined inertial navigation system, a 3D ranging sensor, a vehicle status database, an elevation sequence acquisition module, an obstacle judgment module, an active suspension starting module, an elevation surface sequence acquisition module and an active suspension control module.

[0060] The GNSS / GPS combined inertial navigation system and 3D ranging sensor generate a preview elevation map centered on the vehicle body and store it in the vehicle status database;

[0061] The elevation sequence acquisition module is configured to obtain a preview elevation map from a vehicle status database and then obtain an elevation sequence Ω according to the preview elevation map along the vehicle's travel direction;

[0062] The obstacle judgment module is used to extract road features according to the high-level sequence Ω and save the value of the obstacle R according to the road feature sequence. If an obstacle exists, the obstacle R is saved as {σ0; h k ; K}, where σ0 is the size of the obstacle, h k is the height of the obstacle, and K is the slope of the obstacle; otherwise, set the value of obstacle R to empty;

[0063] The active suspension starting module is used to determine whether the active suspension actuator needs to be started. If the active suspension actuator needs to be started, the active suspension starting module starts the elevation surface sequence acquisition module and the active suspension control module;

[0064] The elevation surface sequence acquisition module is used to discretize the elevation sequence into a surface scale space Surface scale space Contains N elevation surface sequences T j , surface scale space Save in the vehicle status database;

[0065] The active suspension control module is used to obtain the target value of the active suspension control and the control amount of the active suspension actuator; the active suspension control module includes an elevation curve value sequence acquisition submodule, an elevation prediction value sequence prediction submodule, a comparison submodule and an active suspension parameter storage submodule;

[0066] The elevation curve value sequence acquisition submodule is used to obtain the elevation surface sequence T stored in the vehicle status database. j , obtain the elevation curve fj(x); and obtain the elevation curve value sequence according to the elevation curve fj(x);

[0067] The elevation prediction value sequence prediction submodule is used to predict the trajectory in an iterative manner to obtain an elevation prediction value sequence;

[0068] The comparison submodule is used to compare the elevation curve value sequence H and the elevation prediction value sequence H ～ , if the elevation prediction value sequence H ～ Each corresponding element in the elevation curve value sequence H satisfies Then start the active suspension parameter saving submodule;

[0069] The active suspension parameter saving submodule saves the elevation curve fj(x) obtained in the elevation curve value sequence acquisition submodule as the target value of the active suspension control; the elevation sequence Ω and the current elevation surface sequence T j The control amount Γ of the active suspension actuator is obtained by subtraction, and the target value of the active suspension control and the control amount of the active suspension actuator are stored in the vehicle state database.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. The present invention is based on active suspension control based on the previewed terrain. It can judge the maximum driving speed of the vehicle passing through the terrain and the dynamic state of the vehicle passing through the terrain in advance according to the previewed terrain, thereby ensuring the safety of the vehicle.

[0072] 2. The parameters used in the active suspension control process of the present invention are the physical parameters of the vehicle and the geometric dimensions of the raised obstacle. It is simple to implement, has a small amount of calculation, high accuracy, and good robustness. It is a practical and efficient method that can adapt to different vehicle models and raised obstacles.

[0073] 3. Under the condition of the currently predicted terrain, the present invention can control the active suspension more effectively and quickly by using the elevation curve as the target value of the active suspension control and determining the control amount of the active suspension.

[0074] 4. The solution proposed in this invention is used to perform active suspension adjustment to increase vehicle passability and comfort on uneven roads and prevent the vehicle from bottoming out or colliding on special roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a flow chart of a control method of a vehicle active suspension control system based on previewing ground surface features according to an embodiment of the present invention;

[0076] Figure 2 Schematic diagram of the structure of a vehicle active suspension control system based on previewing ground surface features according to an embodiment of the present invention;

[0077] Figure 3Schematic diagram of the state change of a vehicle when passing an obstacle in an embodiment of the present invention. DETAILED DESCRIPTION

[0078] To better understand the technical solutions of the present invention, the specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and embodiments. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0079] The present invention discloses a control method for a vehicle active suspension control system based on previewing ground surface features, such as Figure 1 The specific implementation steps are as follows:

[0080] S1: Extract elevation sequence from preview elevation map.

[0081] A preview elevation map centered on the vehicle is generated using a combination of a GNSS (Global Navigation Satellite System) / GPS (Global Positioning System) inertial navigation system and a 3D ranging sensor. Generating preview elevation maps while a vehicle is in motion is a common technique, as detailed in the paper "Statistical terrain model with geometric feature detection based on GPU using LiDAR on vehicles" published in the journal Measurement Science and Technology (June 8, 2022).

[0082] Along the vehicle's travel direction, the elevation sequence is extracted from the preview elevation map and represented by Ω. The elevation sequence Ω is specifically:

[0083]

[0084]

[0085] Where: h i is the elevation value of the i-th cell; i=1,2,3,…,N SP ; N SP L is the data length of the high-order sequence, which is a positive integer; p is the fixed preview length of the elevation map; Δl is the fixed interval of the elevation map; i is a positive integer.

[0086] The elevation sequence Ω is directly obtained from the preview elevation map. The elevation value in this example is based on the flat road surface. The fixed preview length and fixed interval of the elevation map are set when the preview elevation map is generated. For example, the fixed preview length of the elevation map is set to L p = 10m, the fixed interval of the elevation map Δl = 5cm; then the preview elevation map along the vehicle's travel direction has 200 cells from 1 to 200, that is, N SP = 200. Each cell has its own elevation value. By previewing the elevation map, we can obtain the elevation sequence Ω along the vehicle's travel direction. In actual use, the specific value selection method and number of elements in the elevation sequence can be adjusted according to the elevation map and actual needs.

[0087] S2: Set the value of the obstacle.

[0088] S21: Extract road surface features from the high-order sequence Ω obtained in S1 using the following method:

[0089] L(s)=[1,-2,1]*G(s)*Ω (3)

[0090]

[0091] Where s is the length of the static contact area between the tire and the ground; r takes values ​​between -3s and 3s at intervals of Δl, resulting in a discrete value sequence G(s). The discrete value sequence G(s) is used in formula (3) to better extract road surface features, while the purpose of formula (4) is to smooth the original high-frequency sequence Ω and filter out high-frequency noise. Therefore, a pre-set discrete value sequence G(s) or a discrete value sequence G(s) obtained by other means can also be used directly. L(s) is the extracted road surface feature sequence. The symbol '*' in formula (3) represents the convolution operation of the discrete digital sequence.

[0092] S22: Analyze the obtained road feature sequence L(s) and save the value of the obstacle R.

[0093] The i-th element s in the road surface feature sequence L(s) i If the following conditions are met, then the i-th element s i are considered as feature points:

[0094] |L(s i )|>σ (5)

[0095] In the formula, σ is an empirical value, usually between 0.5 and 0.8. Because the road surface feature sequence L(s) is obtained according to the high-order sequence Ω, the elements in the two have a positional correspondence. Therefore, when the pth element s in L(s) p and the qth element s qWhen it is greater than σ, the p-th element and the q-th element in the high-order sequence Ω are regarded as ground inflection points, and the k-th element s in L(s) k When it is less than -σ, then the k-th element in the high-order sequence Ω is regarded as the maximum value in the high-order sequence Ω, and p < k < q. The distance corresponding to the two front and rear ground inflection points is denoted as σ0, and σ0 is the size of the obstacle, σ0 = (q - p) × Δl; the height of the obstacle is the k-th element h in the high-order sequence Ω k , and K is used to represent the slope of the obstacle:

[0096]

[0097] The obstacle is represented by R, and the data structure of R is as follows:

[0098] R = {σ0; h k ; K} (7)

[0099] The present invention mainly deals with the case where the road surface is a monotonic convex surface, temporarily ignoring other cases. Therefore, only when there are s p , s q and s k in the road surface feature sequence L(s), the value of the obstacle R is saved, otherwise the value of the obstacle R is empty.

[0100] S3: Judge whether the active suspension usage conditions are met. The usage conditions include: whether there is an obstacle, whether the suspension limit stroke H max is greater than the maximum height h of the obstacle p and whether the wheel grounding condition is met. The specific steps are as follows:

[0101] S31: Judge whether there is an obstacle. When the value of the obstacle R in S2 is non-empty, there is an obstacle, and continue with S32. When the value of the obstacle R in S2 is empty, there is no obstacle, return to S1, continue driving, and there is no active suspension control;

[0102] S32: Judge whether the suspension limit stroke H max is greater than the maximum height h of the obstacle p . When the maximum height h of the obstacle p is less than the suspension height H max , continue with S33. When the maximum height h of the obstacle p is greater than or equal to the suspension height H max , give a warning and return to S1. The warning can be a prompt statement, a pattern, a sound alarm or a combination of multiple ones. After receiving the warning, the driver can perform braking or steering avoidance operations.

[0103] S33: Determine whether the ground contact condition related to vehicle speed is met. If the ground contact condition related to vehicle speed is met, proceed to S4. If the ground contact condition related to vehicle speed is not met, give a warning and return to S1. The ground contact condition related to vehicle speed is specifically: the square of the vehicle speed is less than the acceleration due to gravity multiplied by the wheelbase of the front and rear axles, multiplied by the elevation change rate of the obstacle; the elevation change rate is described as the square root of the sum of 1 and the inverse of the square of the slope, expressed as r. a Indicates that r a The expression is:

[0104]

[0105] The vehicle speed-related ground contact condition is expressed as:

[0106] V 2 ≤g×r a ×L (9)

[0107] Where: V is the vehicle speed; L is the wheelbase between the front and rear axles; K is the slope of the obstacle; r a is the rate of change of elevation; g is the acceleration due to gravity.

[0108] S4: Discretize the high-order sequence Ω containing obstacles in S2 into a surface scale space

[0109]

[0110] Surface scale space Contains the process of converting each elevation value in the elevation sequence Ω into its value. After equal division; The high-order sequence group is composed of equal parts, where j = N, N-1, ..., 2, 1. According to the value of j, The elevation sequence composed of equal parts is called elevation surface sequence, denoted by T j .

[0111] If N is set to 16, then the surface scale space for:

[0112]

[0113] Then the value of j is from 16 to 1, and we get T 16 、T 15 , ..., T2, T1, a total of 16 elevation surface sequences, the difference between two adjacent elevation surface sequences is High-level series.

[0114] S5: Initially set j = N, and use the elevation surface sequence to obtain the target value of active suspension control and the control amount of active suspension actuator. According to the state change when the vehicle passes the obstacle, such as Figure 3 As shown, it can be implemented in the following ways:

[0115] S51: Elevation surface sequence T j Perform cubic spline interpolation to obtain the elevation surface sequence T j The curve expression function is referred to as the elevation curve fj(x); the elevation curve value sequence is obtained according to the elevation curve fj(x);

[0116] According to the elevation curve fj(x), the elevation value corresponding to the x coordinate of any point can be obtained. Set the coordinate sequence X, X={x1,x2,…,x i ,…,x n}, n is a positive integer, and the corresponding elevation curve value sequence H = {h x1 ,h x2 ,…,h xi ,…,h xn}, when the coordinate is x i When the corresponding elevation value is h xi ; When x i+1 is x i The next point coordinates, the corresponding elevation value is h x(i+1) ;

[0117] S52: Obtaining a sequence of elevation prediction values; the specific process is as follows:

[0118] Elevation prediction value series Elevation prediction value sequence H ～ It is a prediction of the trajectory, obtained iteratively; the iterative process is as follows:

[0119]

[0120]

[0121] Δθ ～ =ω i ×Δt+Δω×Δt (14)

[0122] Δh ～ =Δθ ～ ×L×cosθ i (15)

[0123]

[0124] Among them, the angular velocity change of the pitch angle is Δω, and the distance from the center of mass to the rear axle of the vehicle is l f , the mass of the vehicle is m, V is the vehicle speed; the moment of inertia of the vehicle around the rear axle is J MThe angle between the straight line from the center of mass to the rear axle of the vehicle and the horizontal plane determined by the axle is δ, and the pitch angle of the vehicle body attitude angle is θ, θ i Represents x i The corresponding pitch angle, θ i+1 Represents x i+1 The corresponding pitch angle, Δθ represents θ i+1 and θ i The difference between the pitch angle and the angular velocity is ω, ω i Represents x i The angular velocity corresponding to the pitch angle; ω i+1 Represents x i+1 The angular velocity corresponding to the pitch angle; x i+1 is x i The coordinates of the next point; Δx represents x i+1 and x i Δt is the time required for the vehicle to pass Δx; Δh represents h x(i+1) and h xi The difference, Δh = h x(i+1) -h xi ; Δh ～ express and The difference, The difference; h xi According to the elevation curve f j (x) The coordinate obtained is x i The corresponding elevation value when Δθ ～ , Δh ～ , h x(i+1) ～ All are predicted values, representing the predicted pitch angle increment predicted value, elevation increment predicted value, and elevation value predicted value respectively.

[0125] The parameters required for formulas (12) to (16) are obtained as follows:

[0126] cosδ=2l f / L (17)

[0127]

[0128]

[0129]

[0130]

[0131]

[0132] Where k is x i+1 The first derivative of the elevation curve fj(x) at .

[0133] S53: Compare the elevation curve value sequence H obtained in S51 with the elevation prediction value sequence H obtained in S52 ～ , if the elevation prediction value sequence H ～ Each corresponding element in the elevation curve value sequence H satisfies Then execute S54; otherwise, set the value of j to j-1; if j is 0, execute S55; if j is not 0, return to S51;

[0134] S54: f j (x) is set as the target value of active suspension control, and the high-order sequence Ω and the current T j Subtract to obtain the control variable Γ of the active suspension actuator:

[0135]

[0136] a i is the i-th element in the control quantity Γ; i=1,2,3,…,N SP ; N SP The data length of the high-order sequence;

[0137] S55: End the loop from S51 to S54.

[0138] The present invention also provides a vehicle active suspension control system based on previewing ground surface features, such as Figure 2 As shown, it specifically includes: a GNSS / GPS combined inertial navigation system 1, a 3D ranging sensor 2, a vehicle status database 3, an elevation sequence acquisition module 4, an obstacle judgment module 5, an active suspension starting module 6, an elevation surface sequence acquisition module 7 and an active suspension control module 8.

[0139] The GNSS / GPS combined inertial navigation system 1 and the 3D ranging sensor 2 generate a preview elevation map centered on the vehicle body and store it in the vehicle status database 3. The GNSS / GPS combined inertial navigation system 1 is fixed to the vehicle body, and the 3D ranging sensor 2 is placed at the front of the vehicle.

[0140] The elevation sequence acquisition module 4 is used to obtain the preview elevation map from the vehicle status database and obtain the elevation sequence Ω according to the preview elevation map along the vehicle's travel direction;

[0141] The obstacle judgment module 5 is used to extract the road surface features according to the high-level sequence Ω and save the value of the obstacle R according to the road surface feature sequence. If an obstacle exists, the obstacle R is saved as {σ0; h k ; K}, where σ0 is the size of the obstacle, h k is the height of the obstacle, and K is the slope of the obstacle; otherwise, set the value of obstacle R to empty.

[0142] The active suspension starting module 6 is used to determine whether the active suspension actuator needs to be started. The determination of starting the active suspension actuator includes: whether there is an obstacle, the suspension limit travel H max Is it greater than the maximum height h of the obstacle? p and whether the wheel grounding condition is met, the specific implementation method adopts step S3; if the active suspension actuator needs to be started, the active suspension starting module starts the elevation surface sequence acquisition module 7 and the active suspension control module 8.

[0143] The elevation surface sequence acquisition module 7 is used to discretize the elevation sequence Ω into a surface scale space Surface scale space Contains N elevation surface sequences T j , surface scale space Stored in the vehicle status database 3.

[0144] The active suspension control module 8 is used to obtain the target value of the active suspension control and the control amount of the active suspension actuator; the active suspension control module includes an elevation curve value sequence acquisition submodule 81, an elevation prediction value sequence prediction submodule 82, a comparison submodule 83 and an active suspension parameter storage submodule 84.

[0145] The elevation curve value sequence acquisition submodule 81 is used to obtain the elevation curve value sequence T according to the elevation curve sequence T stored in the vehicle status database 3. j , and obtain the elevation curve fj(x); and obtain the elevation curve value sequence based on the elevation curve fj(x).

[0146] The elevation prediction value sequence prediction submodule 82 is used to predict the trajectory in an iterative manner to obtain an elevation prediction value sequence, and the prediction is implemented in step S52.

[0147] Comparison submodule 83, used to compare the elevation curve value sequence H and the elevation prediction value sequence H ～ , if the elevation prediction value sequence H ～ Each corresponding element in the elevation curve value sequence H satisfies Then the active suspension parameter saving submodule 84 is started.

[0148] The active suspension parameter saving submodule 84 saves the elevation curve fj(x) obtained in the current course curve value sequence acquisition submodule as the target value of the active suspension control; and saves the elevation sequence Ω and the current elevation surface sequence T j The control variable Γ of the active suspension actuator is obtained by subtraction, and the target value of the active suspension control and the control variable of the active suspension actuator are stored in the vehicle state database 3.

[0149] The embodiments described above are merely descriptions of preferred implementations of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A vehicle active suspension control method based on previewing ground surface features, characterized in that: The specific steps are as follows: S1: Extract elevation sequence from preview elevation map; Use the GNSS / GPS combined inertial navigation system and 3D ranging sensor to generate a preview elevation map centered on the vehicle body; Extract the elevation sequence Ω from the preview elevation map along the vehicle's travel direction; S2: Set the value of the obstacle; S21: extracting road surface features from the high-order sequence Ω obtained in S1 to obtain a road surface feature sequence L(s); S22: By analyzing the obtained road feature sequence L(s), if there is an obstacle, save the obstacle R = {σ0; h k ; K}, where σ0 is the size of the obstacle, h k is the height of the obstacle, and K is the slope of the obstacle; otherwise, the value of obstacle R is empty; S3: Determine whether the active suspension use conditions are met. If so, execute S4. If not, issue a warning and return to S1. S4: Discretize the high-order sequence Ω containing obstacles in S2 into a surface scale space Surface scale space Contains the process of converting each elevation value in the elevation sequence Ω into its value. After equal division; The high-order sequence group composed of equal parts will be divided into The elevation sequence composed of equal parts is called elevation surface sequence, denoted by T j ;h i is the elevation value of the i-th cell; i=1,2,3,…,N SP ; N SP The data length of the high-order sequence is a positive integer; S5: Initially set j = N, and use the elevation surface sequence to obtain the target value of the active suspension control and the control amount of the active suspension actuator, which specifically includes the following sub-steps: S51: Elevation surface sequence T j Perform cubic spline interpolation to obtain the elevation curve f j (x); according to the elevation curve f j (x) Get the elevation curve value sequence H = {h x1 ,h x2 ,…,h xi ,…,h xn }, when the coordinate is x i When the corresponding elevation value is h xi ; When x i+1 is x i The next point coordinates, the corresponding elevation value is h x(i+1) ; S52: Obtain elevation prediction value sequence using iteration The specific iterative process is as follows: Dth ～ =ω i ×Δt+Δω×Δt (14) Dh ～ =Δθ ～ ×L×cosθ i (15) Among them, the angular velocity change of the pitch angle is Δω, and the distance from the center of mass to the rear axle of the vehicle is l f , the mass of the vehicle is m, V is the vehicle speed; the moment of inertia of the vehicle around the rear axle is J M The angle between the straight line from the center of mass to the rear axle of the vehicle and the horizontal plane determined by the axle is δ, and the pitch angle of the vehicle body attitude angle is θ, θ i Represents x i The corresponding pitch angle, θ i+1 Represents x i+1 The corresponding pitch angle, Δθ represents θ i+1 and θ i The difference between the pitch angle and the angular velocity is ω, ω i Represents x i The angular velocity corresponding to the pitch angle; ω i+1 Represents x i+1 The angular velocity corresponding to the pitch angle; x i+1 is x i The coordinates of the next point; Δx represents x i+1 and x i Δt is the time required for the vehicle to pass Δx; Δh represents h x(i+1) and h xi The difference, Δh ～ express and The difference; h xi According to the elevation curve f j (x) The coordinate obtained is x i The corresponding elevation value when h x(i+1) According to the elevation curve f j (x) The coordinate obtained is x i+1 The corresponding elevation value when Δθ ～ , Δh ～ , hx (i+1) ～ are all predicted values, representing the predicted pitch angle increment, elevation increment, and elevation value respectively; L is the wheelbase between the front and rear axles of the vehicle body; The parameters required for formulas (12) to (16) are obtained as follows: cosδ=2l f / L (17) Where k is x i+1 Elevation curve f j (x) first-order derivative; S53: Compare the elevation curve value sequence H obtained in S51 with the elevation prediction value sequence H obtained in S52 ～ , if the elevation prediction value sequence H ～ Each corresponding element in the elevation curve value sequence H satisfies Then execute S54; otherwise, set the value of j to j-1; if j is 0, execute S55; if j is not 0, return to S51; S54: f j (x) is set as the target value of active suspension control, and the high-order sequence Ω and the current T j Subtract to obtain the control variable Γ of the active suspension actuator: a i is the i-th element in the control quantity Γ; i=1,2,3,…,N SP ; N SP The data length of the high-order sequence; S55: End the loop.

2. The vehicle active suspension control method based on previewing ground surface features according to claim 1, characterized in that: The high-order sequence Ω in S1 is specifically: Where: h i is the elevation value of the i-th cell; N SP L is the data length of the high-order sequence; p is the fixed preview length of the elevation map; Δl is the fixed interval of the elevation map; i is a positive integer.

3. The vehicle active suspension control method based on previewing ground surface features according to claim 1, characterized in that: The S21 extracts road surface features from the high-order sequence Ω to obtain a road surface feature sequence L(s). The specific process is as follows: L(s)=[1,-2,1]*G(s)*Ω (3) Among them, G(s) is a discrete value sequence used to smooth the original high-order sequence Ω and filter out high-frequency noise. The symbol '*' in formula (3) represents the convolution operation of the discrete digital sequence.

4. The vehicle active suspension control method based on previewing ground surface features according to claim 1, characterized in that: The specific process of step S22 is as follows: The i-th element s in the road surface feature sequence L(s) i If the following conditions are met, then the i-th element s i are considered as feature points: |L(s i )|>σ (5) Wherein, the value of σ is an empirical value. There is a position correspondence relationship between the pavement feature sequence L(s) and the elements in the high-order sequence Ω. Therefore, when the p-th element s p and the q-th element s q in L(s) are greater than σ, the p-th element and the q-th element in the high-order sequence Ω are regarded as ground inflection points. When the k-th element s k in L(s) is less than -σ, then the k-th element in the high-order sequence Ω is regarded as the maximum value in the high-order sequence Ω. When there are three points p, k, and q, and p < k < q, the obstacle is represented by R = {σ0; h k ; K}; the size of the obstacle σ0 = (q - p) × Δl; the height of the obstacle is the k-th element h k in the high-order sequence Ω, and the slope of the obstacle Otherwise, the value of the obstacle R is empty.

5. The vehicle active suspension control method based on previewing ground surface features according to claim 1, characterized in that: The specific judgment process of S3 is as follows: S31: Determine whether there is an obstacle. If the value of obstacle R in S2 is not null, there is an obstacle, and proceed to S32. If the value of obstacle R in S2 is null, there is no obstacle, and return to S1 to continue driving without active suspension control. S32: Determine the suspension limit travel H max Is it greater than the maximum height h of the obstacle? p , when the maximum height of the obstacle h p Less than the suspension height H max When the maximum height h of the obstacle is p Greater than or equal to the suspension height H max When , a warning is given and the process returns to S1; S33: Determine whether a grounding condition related to vehicle speed is met. When the vehicle speed grounding condition is met, continue to S4. When the wheel grounding condition is not met, give a warning and return to S1.

6. The vehicle active suspension control method based on previewing ground surface features according to claim 5, characterized in that: The vehicle speed grounding condition in S33 is: The square of the vehicle speed is less than the acceleration due to gravity multiplied by the wheelbase of the front and rear axles, multiplied by the rate of change of the obstacle's elevation; the rate of change of elevation is described as the square root of the sum of 1 plus the inverse of the square of the slope, expressed as r. a Indicates that r a The expression is: The vehicle speed grounding condition is expressed as the formula: In 2 ≤g×r d ×L (9) Where: V is the vehicle speed; L is the wheelbase between the front and rear axles; K is the slope of the obstacle; r a is the rate of change of elevation; g is the acceleration due to gravity.

7. A vehicle active suspension control system based on previewing ground curved surface features, using the vehicle active suspension control method based on previewing ground curved surface features as described in claims 1-6, characterized in that: Specifically, it includes the following devices: GNSS / GPS combined inertial navigation system, 3D ranging sensor, vehicle status database, elevation sequence acquisition module, obstacle judgment module, active suspension activation module, elevation surface sequence acquisition module and active suspension control module; The GNSS / GPS combined inertial navigation system and 3D ranging sensor generate a preview elevation map centered on the vehicle body and store it in the vehicle status database; The elevation sequence acquisition module is configured to obtain a preview elevation map from a vehicle status database and then obtain an elevation sequence Ω according to the preview elevation map along the vehicle's travel direction; The obstacle judgment module is used to extract road features according to the high-level sequence Ω and save the value of the obstacle R according to the road feature sequence. If an obstacle exists, the obstacle R is saved as {σ0; h k ; K}, where σ0 is the size of the obstacle, h k is the height of the obstacle, and K is the slope of the obstacle; otherwise, set the value of obstacle R to empty; The active suspension starting module is used to determine whether the active suspension actuator needs to be started. If the active suspension actuator needs to be started, the active suspension starting module starts the elevation surface sequence acquisition module and the active suspension control module; The elevation surface sequence acquisition module is used to discretize the elevation sequence into a surface scale space Surface scale space Contains N elevation surface sequences T j , surface scale space Save in the vehicle status database; The active suspension control module is used to obtain the target value of the active suspension control and the control amount of the active suspension actuator; the active suspension control module includes an elevation curve value sequence acquisition submodule, an elevation prediction value sequence prediction submodule, a comparison submodule and an active suspension parameter storage submodule; The elevation curve value sequence acquisition submodule is used to obtain the elevation surface sequence T stored in the vehicle status database. j , get the elevation curve f j (x); and according to the elevation curve f j (x) obtain the elevation curve value sequence; The elevation prediction value sequence prediction submodule is used to predict the trajectory in an iterative manner to obtain an elevation prediction value sequence; The comparison submodule is used to compare the elevation curve value sequence H and the elevation prediction value sequence H ～ , if the elevation prediction value sequence H ～ Each corresponding element in the elevation curve value sequence H satisfies Then start the active suspension parameter saving submodule; The active suspension parameter saving submodule obtains the elevation curve f obtained in the curve value sequence submodule. j (x) is saved as the target value of active suspension control; the elevation sequence Ω and the current elevation surface sequence T j The control amount Γ of the active suspension actuator is obtained by subtraction, and the target value of the active suspension control and the control amount of the active suspension actuator are stored in the vehicle state database.

Images