Control method of rigid-flexible coupled active suspension for aerial platform fire truck

Abstract

The application provides a kind of rigid-flex coupling active suspension control method of aerial-jet fire fighting truck of in-transit operation, it is related to fire truck control field;Its steps include: S1, rigid-flex coupling dynamics model is constructed using Lagrange method, rigid-flex coupling state matrix and state input matrix are obtained;S2, state observer is constructed and rigid-flex coupling state estimation vector is calculated;S3, segmented flexible working arm is rigidly equivalent, rigid reference gain is constructed;S4, using the dynamics coupling relationship between rigid vehicle body and segmented flexible working arm, bounded pointing increment compensation is constructed;S5, six-channel actual actuation power is formed within the execution constraint range of active suspension and fed back to state observer, closed-loop control of active suspension is formed;The application solves the problems of difficult description, stability and difficult to consider by establishing rigid-flex coupling model, hierarchical control and space projection error index, and brings the effect of double promotion of in-transit spraying precision and vehicle stability.

Description

Technical Field

[0001] This invention relates to the field of fire truck control, and specifically to a rigid-flexible coupling active suspension control method for a fire truck operating while in motion. Background Technology

[0002] Aerial fire trucks typically need to extend a long boom during firefighting and rescue operations, spraying water or other extinguishing agents through a fire monitor mounted at the end of the boom. As firefighting and rescue missions evolve towards complex terrain, high-rise buildings, high-volume spraying, and continuous operations while on the move, maintaining the stability of the fire monitor's spray direction while the vehicle is in motion has become a critical issue affecting the accuracy of fire suppression coverage and operational safety.

[0003] Existing active suspension control methods, such as those described in patent CN101916113B, primarily target the vertical, pitch, and roll movements of the vehicle body, aiming to improve ride comfort and handling stability by adjusting the active suspension. However, this type of method has the following shortcomings when applied to aerial ladder fire trucks: 1. Existing active suspension dynamics models simplify the boom to a rigid added mass, focusing on the rigid motion of the vehicle body as the primary modeling object. They fail to incorporate the flexible boom as a continuous structure with elastic degrees of freedom into a unified modeling framework, neglecting the description of the dynamic response of the segmented flexible boom. Furthermore, for external inputs such as the active suspension's dynamics, road surface excitation, and jet reaction force, existing models fail to integrate them with the flexible boom deformation into a unified rigid-flexible coupling system, making it difficult to describe the coupling relationship between the rigid motion of the vehicle and the dynamic response of the flexible boom during operation.

[0004] 2. Existing control strategies employ a single-level control architecture, focusing solely on vehicle attitude stability and failing to construct a layered control structure tailored to the spraying characteristics of aerial ladder fire trucks. Specifically, existing methods neither establish a rigid reference gain to maintain basic vehicle attitude stability nor utilize a rigid-flexible coupling model to generate limited pointing corrections on top of the reference control. Consequently, while maintaining basic vehicle stability, the active suspension control cannot simultaneously address spray pointing accuracy and boom dynamic response, making it difficult to effectively suppress residual pointing errors caused by flexible modes.

[0005] 3. Existing methods for evaluating the vertical pointing error of the spray remain at the level of simple geometric relationships. They fail to construct pointing error indicators based on the spatial projection relationship between vehicle pitch angle, vehicle roll angle, the nominal spray yaw direction of the fire monitor, and the flexible angle of the boom end. This results in the inability to uniformly quantify the impact of vehicle attitude changes and flexible deformation of the working boom on the vertical pointing of the spray. Especially under non-zero yaw conditions, existing technologies struggle to reflect the combined effects of vehicle pitch, roll, and flexible deformation of the boom end on the spray direction, and there is a lack of direct correlation between the active suspension control target and the pointing stability requirements of spraying operations while in motion.

[0006] In summary, existing active suspension control methods mainly use vehicle attitude as the object of control and evaluation, making it difficult to balance vehicle attitude stability and arm-end injection direction stability under the physical constraints of the suspension. Summary of the Invention

[0007] To address the shortcomings of the prior art, the present invention aims to provide a rigid-flexible coupling active suspension control method for a fire truck operating while in motion. This method addresses the problems of existing technologies, such as the inability to uniformly describe the rigid-flexible coupling relationship, the inability to balance attitude stability and spray pointing accuracy, and the inability to quantify the combined effects of multiple factors on high and low pointing.

[0008] Specifically, the present invention provides a rigid-flexible coupling active suspension control method for a fire truck operating while in motion, which includes the following steps: S1. Analyze the contribution relationship between the rigid body, unsprung wheel ends, and segmented flexible working arm to the vehicle system energy; analyze the fire monitor's operating state and static equilibrium state to construct the jet reaction force increment; analyze the relationship between the active suspension's dynamics, road excitation, and jet reaction force increment on the virtual work of external forces; construct a rigid-flexible coupling dynamic model using the Lagrange method, and obtain the rigid-flexible coupling state matrix and state input matrix after first-order state space transformation and mass matrix transformation; S2. The vehicle side measurement signals during the vehicle's driving process are collected using the vehicle's own sensors. The injection reaction force increment is used as a known disturbance input to construct a state observer and calculate the rigid-flexible coupling state estimation vector. S3. Based on the vehicle structure and operating parameters consistent with the rigid-flexible coupling dynamic model, the segmented flexible working arm is rigidly equivalent. The basic attitude of the whole vehicle and the suspension stability requirements are used as the benchmark control targets to construct the extended-dimensional rigid benchmark gain. S4. Based on the rigid-flexible coupling state matrix and state input matrix, establish the jetting high and low direction pointing error according to the spatial projection relationship between the vehicle pitch angle, vehicle roll angle, fire monitor nominal spray yaw direction and arm end flexible rotation angle; on the extended dimension rigid reference gain, combined with the high and low direction pointing error, use the dynamic coupling relationship between the rigid vehicle body and the segmented flexible working arm to construct bounded pointing increment compensation. S5. Based on the rigid-flexible coupling state estimation vector, the extended-dimensional rigid reference gain and the bounded pointing increment compensation are fused together, and a six-channel actual driving force is formed within the active suspension execution constraint range and fed back to the state observer, so that the active suspension can correct the injection direction while maintaining the vehicle attitude stability, thus forming a closed-loop control of the active suspension.

[0009] Preferably, S1 includes: S11. Establish a three-dimensional Cartesian inertial coordinate system, a three-dimensional rectangular coordinate system fixed to the vehicle body, and a local axial coordinate system for the working arm; divide the working arm in the extended state into four segments along the axial direction, and obtain the natural circular frequency and normalized mode function through numerical eigenvalue analysis. S12. Based on the normalized modal function, suspension dynamic travel vector, suspension dynamic travel rate of change and tire dynamic deformation vector, calculate the vehicle system kinetic energy, suspension dissipated energy and system potential energy one by one, and construct the whole vehicle parameter matrix. S13. Based on the three-dimensional rectangular coordinate system of the vehicle body, select the static equilibrium point under the combined action of vehicle gravity, working arm gravity and average injection reaction force as the static equilibrium state, and calculate the injection reaction force increment near the static equilibrium state. S14. Based on the principle of virtual work, combined with the suspension geometric mapping relationship, tire dynamic deformation vector and the position and direction of action of the arm end injection reaction force, construct the corresponding generalized input matrix respectively; S15. Based on the vehicle parameter matrix obtained in S12 and the generalized input matrix obtained in S14, and combined with the injection reaction force increment obtained in S13, a rigid-flexible coupled dynamic model is constructed using the Lagrange method. After first-order state space transformation and mass matrix transformation, the rigid-flexible coupled state matrix and state input matrix are obtained.

[0010] Preferably, the vehicle parameter matrix of S12 includes: a vehicle mass parameter matrix, a vehicle damping parameter matrix, and a vehicle stiffness parameter matrix; the steps include: S121. Based on the normalized mode function, the system kinetic energy is constructed by combining the rigid body mass and moment of inertia, the mass at the ends of the six unsprung wheels, the distributed mass of the working arm, and the equivalent concentrated mass at the arm end. The system kinetic energy is then rearranged according to the square and cross terms of each generalized velocity to obtain the vehicle mass parameter matrix including inertial coupling terms. : ; in, Here are the mass and moment of inertia matrices. The coupling matrix obtained from the cross kinetic energy term. The mass matrix consists of the masses at the ends of six unsprung wheels. Let T be the modal mass matrix of the flexible working arm, and T be the transpose vector. S122. The suspension energy dissipation is constructed using the suspension dynamic travel change rate and the preset suspension damping coefficient. Modal damping is constructed by combining the natural circular frequency and the preset modal damping ratio. The two are then rearranged according to the generalized velocity to obtain the vehicle damping parameter matrix. : ; in, For geometric mapping matrix, This is the suspension damping matrix composed of six suspension damping coefficients. The modal damping matrix of the flexible working arm; S123. Based on the suspension dynamic travel vector and the tire dynamic deformation vector, construct the suspension elastic potential energy and the tire elastic potential energy respectively. Combine the natural circular frequency to construct the modal elastic potential energy. Construct the gravitational potential energy according to the mass parameters of the working arm and arm end. Linearize the corresponding generalized gravitational force near the static equilibrium posture to obtain the gravitational tangential stiffness correction. Organize the system potential energy according to the generalized displacement to obtain the vehicle stiffness parameter matrix including stiffness coupling terms. : ; in, This is the suspension stiffness matrix composed of six suspension stiffness coefficients. Correction for the gravitational tangential stiffness corresponding to the rigid body motion of the vehicle body. This is the gravitational stiffness coupling term between the rigid body motion and the flexible modal motion of the vehicle body. This is a tire stiffness matrix consisting of six tire stiffness coefficients. The modal stiffness matrix of the flexible working arm. This is a correction for the gravitational tangential stiffness corresponding to flexible modal motion.

[0011] Preferably, S13 includes: S131. Based on the three-dimensional rectangular coordinate system fixed to the vehicle body, and combined with the quantitative relationship between the effective cross-sectional area of ​​the fire monitor nozzle and the spray volume flow rate, the total spray reaction force of the fire monitor is obtained. S132. Select a unit time interval, take the static equilibrium point under the combined action of vehicle weight, working arm weight and average injection reaction force as the static equilibrium state, and calculate the average injection reaction force. S133. Calculate the difference between the total jet reaction force and the average jet reaction force of the fire monitor to obtain the jet reaction force increment relative to the static equilibrium state.

[0012] Preferably, the generalized input matrix of S14 includes the generalized input matrix of the active suspension, the generalized input matrix of the road surface, and the generalized input matrix of the injection reaction force; the steps include: S141. Based on the position and direction of action of the six active suspension actuators, the six-channel active suspension action force is mapped to the rigid body channel of the vehicle body and the six wheel end channels through the virtual work relationship to obtain the generalized input matrix of the active suspension. S142. Based on the relative relationship between wheel end displacement and road surface roughness in tire dynamic deformation, the tire elastic effect is expressed as the influence of wheel end displacement on the vehicle stiffness and the external excitation effect of road surface roughness on the system. The effect corresponding to wheel end displacement is included in the vehicle stiffness parameter matrix, and the effect corresponding to road surface roughness forms the road surface generalized input matrix. S143. Based on the three-dimensional rectangular coordinate system fixed to the vehicle body, the position of the arm end relative to the center of mass of the vehicle body is quantified. According to the virtual work done by the arm end jet reaction force on the vertical, pitch, roll and flexible modes of the vehicle body, the generalized input matrix of the jet reaction force is obtained.

[0013] Preferably, the relationship between the vehicle-side measurement signals in S2 is as follows: ; in, This is the vehicle-side measurement vector, and it is a 10-dimensional real vector. The vehicle body pitch angle, The body roll angle, , Angular velocity, This is the suspension dynamic travel vector. For vehicle-side measurement matrix, For rigid-flexible coupling state vectors, This is a 10-dimensional measurement noise vector.

[0014] Preferably, S4 includes: S41. Based on the three-dimensional Cartesian inertial coordinate system and the vehicle body fixed three-dimensional rectangular coordinate system, according to the projection relationship of the vehicle body pitch angle, vehicle body roll angle, fire monitor yaw direction and arm end flexible rotation angle in the fire monitor spray height direction, the spray height pointing error of the fire monitor relative to the nominal spray direction is obtained. S42. Based on the linear relationship between normalized energy and rigid-flexible coupling state vector, and combined with the projection relationship of injection altitude and direction pointing errors, construct the pointing performance output matrix; based on the normalized modal function and rigid-flexible coupling state vector, obtain the first mode arm end deflection and the first mode arm end flexible rotation angle; combined with the projection relationship of injection altitude and direction pointing errors, obtain the first mode pointing proxy quantity; and construct the pointing performance output matrix. : ; in, For standard partial derivatives, The first mode arm end deflection, The first mode arm has a flexible rotation angle. The target first modality points to the proxy quantity. This serves as the non-zero normalized reference value for the deflection at the end of the first mode arm. This is a non-zero normalized reference value for the flexible rotation angle at the end of the first mode arm. The non-zero normalized reference value for the first modality pointing to the surrogate quantity; S43. Set the basic state weight matrix based on the basic control requirements of vehicle posture, wheel ends, suspension travel, and tire dynamic deformation; set the output pointing performance weight matrix based on the importance of the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing proxy quantity; map the pointing performance weights to the rigid-flexible coupling state space using the pointing performance output matrix, and add it to the basic state weight matrix to obtain the pointing perception state weight matrix. ; ; in, Based on the basic state weight matrix, This refers to the performance output matrix. To point to the performance weight matrix, This refers to the state weight increment formed after mapping the performance weights to the rigid-flexible coupling state space; S44. Input the rigid-flexible coupling state matrix, the state input matrix, and the pointing sensing state weight matrix into the linear quadratic regulator, and obtain the pointing sensing feedback gain offline. S45. Generate bounded pointing increment compensation based on the difference between the pointing perception feedback gain and the extended dimension rigid reference gain.

[0015] Preferably, S42 includes: S421. Based on the normalized mode function, extract the first flexible mode coordinates from the rigid-flexible coupling state vector, combine the two to obtain the first mode arm end deflection and the first mode arm end flexible rotation angle. S422. Based on the projection relationship of the jet elevation and elevation pointing errors, the vehicle pitch angle, vehicle roll angle, fire monitor yaw direction, and first mode arm end flexible rotation angle are combined to obtain the target first mode pointing proxy quantity. ; ; in, The vehicle body pitch angle, The body roll angle, To fix the nominal yaw angle of the fire monitor; S423. Normalize the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing proxy quantity using non-zero reference quantities to obtain normalized energy. Based on the linear relationship between normalized energy and rigid-flexible coupling state vector, construct the pointing performance output matrix.

[0016] Preferably, S5 includes: Based on the extended-dimensional rigid benchmark gain and estimated state, the fundamental dynamics are generated. : ; in, For extended-dimensional rigid reference gain, This is the state estimation vector for rigid-flexible coupling. The bounded pointing increment compensation is multiplied by the rigid-flexible coupling state estimation vector, and after adjusting the pointing increment compensation coefficients and limiting the amplitude of each channel, a six-channel pointing increment compensation vector is obtained. : ; in, The upper limit of the directional increment compensation amplitude for a single active suspension actuator; To restrict the input of each of the six channels to a range Inside, This refers to the incremental compensation coefficient. This refers to the gain of the sensor feedback. The basic working force and the six-channel directional incremental compensation vector are added together, and constraints on the rate of change of working force and the final amplitude are applied to obtain the actual working force of the six channels. : ; in, This represents the upper limit of the final actuation force amplitude of a single active suspension actuator. To ultimately perform channel-by-channel saturation processing, This represents the upper limit of the rate of change of the actuating force of a single active suspension actuator. To limit the rate of change of dynamics for each channel.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention integrates the rigid body, unsprung wheel ends, segmented flexible working arm, active suspension as power, road excitation and jet reaction force into a rigid-flexible coupled dynamic model, which can describe the coupling relationship between the rigid body motion of the vehicle and the dynamic response of the flexible working arm during the operation of the aerial fire truck.

[0018] 2. This invention constructs a hierarchical control structure consisting of a rigid reference gain and bounded pointing increment compensation. The rigid reference gain maintains the basic attitude stability of the vehicle, and on this basis, a rigid-flexible coupling model is used to generate a limited pointing correction, so that the active suspension control can maintain the basic stability of the vehicle while taking into account the injection pointing and dynamic response of the working arm.

[0019] 3. This invention constructs the spray elevation and elevation pointing error based on the spatial projection relationship between the vehicle pitch angle, vehicle roll angle, nominal spray yaw direction of the fire monitor, and flexible rotation angle at the boom end. This allows for the unified quantification of the influence of vehicle attitude changes and flexible deformation of the working boom on the spray elevation and elevation pointing. It can reflect the combined influence of vehicle pitch, roll, and flexible deformation of the boom end on the spray direction under non-zero yaw conditions, and directly link the active suspension control target with the pointing stability requirements of spraying operations while in motion. Attached Figure Description

[0020] Figure 1 This is a flowchart of the rigid-flexible coupling active suspension control method for a fire truck operating while in motion according to the present invention. Figure 2 This is a schematic diagram of the rigid-flexible coupling dynamics model of the aerial spray fire truck of the present invention; Figure 3 This is a comparison diagram of the vehicle pitch angle response under the combined action of random road surface and jet reaction force according to the present invention; Figure 4 This is a comparison diagram of the vehicle roll angle response under the combined action of random road surface and jet reaction force according to the present invention; Figure 5 This is a comparison diagram of the jetting vertical pointing error response of the present invention under the combined action of random road surface and jetting reaction force; Figure 6 This is a comparison chart of the coordinate model output and observation estimation of the first-order flexible mode under the combined action of random road surface and jet reaction force according to the present invention; Figure 7 This diagram shows the dynamic constraint results of the active suspension system under the combined action of random road surface and jet reaction force according to the present invention. Figure 8 This is a comparison diagram of the vehicle pitch angle response under the combined action of a single-sided convex road surface and jet reaction force. Figure 9 This is a comparison diagram of the vehicle roll angle response under the combined action of a single-sided convex road surface and jet reaction force according to the present invention; Figure 10 This is a comparison diagram of the jetting height and direction pointing error response of the present invention under the combined action of a single-sided convex road surface and jetting reaction force; Figure 11 This is a comparison chart of the output and observation estimation of the first-order flexible modal coordinate model under the combined action of a single-sided convex road surface and jet reaction force according to the present invention; Figure 12 This diagram shows the dynamic constraint results of the active suspension system under the combined action of a single-sided convex road surface and jet reaction force according to the present invention. Detailed Implementation

[0021] The following is for reference Figure 1 The embodiments of the present invention will be described.

[0022] On the other hand, the present invention provides a rigid-flexible coupling active suspension control method for a fire truck operating while in motion, comprising: S1. Analyze the contribution relationship between the rigid body, unsprung wheel ends, and segmented flexible working arm to the vehicle system energy; analyze the fire monitor's operating state and static equilibrium state to construct the jet reaction force increment; analyze the relationship between the active suspension's dynamics, road excitation, and jet reaction force increment on the virtual work of external forces; construct a rigid-flexible coupling dynamic model using the Lagrange method, and obtain the rigid-flexible coupling state matrix and state input matrix after first-order state space transformation and mass matrix transformation.

[0023] Specifically, the energy contribution relationship of the vehicle system is described using a vehicle parameter matrix, which includes: vehicle mass parameter matrix, vehicle damping parameter matrix, and vehicle stiffness parameter matrix; the interaction of virtual work of external forces is described using a generalized input matrix, which includes the generalized input matrix of active suspension, the generalized input matrix of road surface, and the generalized input matrix of injection reaction force.

[0024] S11. Establish a three-dimensional Cartesian inertial coordinate system, a three-dimensional rectangular coordinate system fixed to the vehicle body, and a local axial coordinate system for the working arm; divide the working arm in the extended state into four segments along the axial direction, and obtain the natural circular frequency and normalized mode function through numerical eigenvalue analysis. S111. Establish a three-dimensional Cartesian inertial coordinate system, a three-dimensional rectangular coordinate system fixed to the body with the body centroid of the rigid body as the origin, and a local axial coordinate system of the working arm along the axis of the working arm in the extended state. like Figure 2 As shown, a three-dimensional Cartesian inertial coordinate system, a vehicle-mounted three-dimensional rectangular coordinate system, and an axial coordinate system are established. Specifically, in the vehicle-mounted three-dimensional rectangular coordinate system, the x-axis is along the longitudinal direction of the vehicle, the y-axis is along the transverse direction of the vehicle, and the z-axis is vertically upward. The vehicle pitch angle is defined using the right-hand rule about the positive y-axis, and the vehicle roll angle is defined using the right-hand rule about the positive x-axis.

[0025] S112. Based on the vehicle body's fixed three-dimensional Cartesian coordinate system, under small angle conditions, the relationship between the vehicle suspension and the vehicle body's center of gravity is analyzed, and the suspension dynamic travel vector is calculated. Suspension dynamic travel change rate With tire dynamic deformation vector : ; ; ; in, For geometric mapping matrix, Let be the generalized coordinate vector of the vehicle body rigid body. Let V be the generalized velocity vector of the vehicle body. These are the vertical displacement vectors of the six unsprung wheel ends. These are the vertical velocity vectors of the six unsprung wheel ends. This is the road surface input vector.

[0026] Specifically, geometric mapping matrix Defined as: ; in, and The first The coordinates of the suspension mounting position relative to the vehicle's center of gravity in the longitudinal and lateral directions of the vehicle's fixed three-dimensional Cartesian coordinate system. ; Used to map the generalized coordinates of the vehicle body rigid body to the vertical displacement at the six suspension mounting positions.

[0027] Generalized coordinate vector of rigid body of vehicle body and the generalized velocity vector of the car body rigid body They are defined as follows: , ; in, This represents the vertical displacement of the vehicle's center of gravity. The vehicle body pitch angle, This refers to the vehicle body roll angle; , and These are the corresponding vertical velocity of the vehicle's center of gravity, pitch angular velocity, and roll angular velocity.

[0028] Vertical displacement vectors of the six unsprung wheel ends and the velocity vectors of the six unsprung wheel ends They are defined as follows: , ; in, Let be the vertical displacement of the b-th unsprung wheel end. Let be the vertical velocity corresponding to the b-th unsprung wheel end. .

[0029] Road surface input vector Defined as: ; in, Input the vertical unevenness of the road surface at the position corresponding to the b-th unsprung wheel end. .

[0030] S113. Based on the local axial coordinate system of the working arm, the flexible working arm is divided into four segments along the axial direction. The boundary conditions and connection continuity conditions of each segment of the working arm are analyzed based on the natural vibration characteristic equation of the segmented beam. The natural circular frequency and modal function are obtained through numerical eigenvalue analysis. ; in, For the local axial coordinates of the working arm, The elastic modulus of the working arm, The inertia matrix of the working arm section is... Let j be the mass per unit length of the working arm, and j be the modal order index. Let j be the natural angular frequency. This is the j-th order mode function obtained from numerical eigenvalue analysis.

[0031] For mode functions Mass normalization is performed to obtain the j-th normalized mode function. ,satisfy: ; Where i is the first modal order arbitrarily selected from the modal order index j, and k is the second modal order arbitrarily selected from the modal order index j. Let i be the normalized mode function of order i. Let k be the normalized mode function. For the equivalent concentrated mass at the arm end, For the i-th normalized mode function at the arm end The value at that location, For the k-th normalized mode function at the arm end The value at point L is the total axial length of the working arm. For Kronecker symbols.

[0032] S12. Based on the normalized modal function, suspension dynamic travel vector, suspension dynamic travel rate of change and tire dynamic deformation vector, the kinetic energy of the vehicle system, the energy dissipated by the suspension and the potential energy of the system are calculated one by one to construct the whole vehicle parameter matrix.

[0033] Specifically, the vehicle parameter matrix includes: the vehicle mass parameter matrix, the vehicle damping parameter matrix, and the vehicle stiffness parameter matrix.

[0034] S121. Based on the normalized mode function obtained in S113, the system kinetic energy is constructed by combining the rigid body mass and moment of inertia, the mass at the ends of the six unsprung wheels, the distributed mass of the working arm, and the equivalent concentrated mass at the arm end. The system kinetic energy is then rearranged according to the square and cross terms of each generalized velocity to obtain the vehicle mass parameter matrix including inertial coupling terms. for: ; in, This represents the mass and moment of inertia matrix formed by the vehicle body and the boom in the vertical, pitch, and roll paths of the vehicle body. This is the coupling matrix obtained from the cross-kinetic energy term between the generalized kinetic energy of the rigid body and the generalized kinetic energy of the flexible modes after substituting the normalized modal function into the system kinetic energy. The mass matrix consists of the masses at the ends of six unsprung wheels. The modal mass matrix of the flexible working arm; Specifically, the normalized mode functions of each order obtained in S113 are combined into a normalized mode function vector. The corresponding arm-end normalized mode function vector is denoted as The modal mass matrix of the flexible working arm for: ; Where I is the identity matrix corresponding to the retained flexible mode order. To form a normalized mode function vector for each order of normalized mode function, For the arm-end normalized mode function vector, The equivalent concentrated mass at the end of the arm.

[0035] S122. Using the suspension dynamic travel change rate obtained in S112 and the preset suspension damping coefficient, the suspension energy dissipation is constructed. Combined with the natural circular frequency obtained in S113 and the preset modal damping ratio, modal damping is constructed. The two are then rearranged according to the generalized velocity to obtain the vehicle damping parameter matrix: Specifically, based on the suspension dynamic travel change rate obtained from S11 and the suspension damping matrix consisting of six suspension damping coefficients. Energy dissipation through suspension design: ; According to the generalized velocity vector of the rigid body of the vehicle Vertical velocity vectors of the six unsprung wheel ends After processing, the block form of the suspension damping in the whole vehicle damping parameter matrix is ​​obtained: ; Specifically, based on the natural circular frequencies and corresponding modal damping ratios obtained from S113, the modal damping matrix of the flexible working arm is obtained. : ; in, To arrange the items within the parentheses into a diagonal matrix according to the j-th order modal order index, Let j be the damping ratio of the j-th mode. Let be the j-th natural angular frequency.

[0036] By combining the suspension damping blocks with the flexible working arm modal damping according to the generalized velocity arrangement relationship, the whole vehicle damping parameter matrix is ​​obtained. : ; in, This is the suspension damping matrix composed of six suspension damping coefficients. This is the modal damping matrix for the flexible working arm.

[0037] S123. Based on the suspension dynamic travel vector and tire dynamic deformation vector obtained in S11, the suspension elastic potential energy and tire elastic potential energy are constructed respectively. Combined with the natural circular frequency obtained in S113, modal elastic potential energy is constructed. Gravitational potential energy is constructed according to the mass parameters of the working arm and arm end. The corresponding generalized gravity force is linearized near the static equilibrium posture to obtain the gravity tangential stiffness correction. The system potential energy is rearranged according to the generalized displacement to obtain the vehicle stiffness parameter matrix including stiffness coupling terms. Specifically, based on the suspension dynamic travel vector obtained from S11 and the suspension stiffness matrix consisting of six suspension stiffness coefficients. The elastic potential energy of the suspension structure: ; According to the generalized coordinate vector of the rigid body of the vehicle and the vertical displacement vectors of the six unsprung wheel ends After processing, the suspension stiffness is represented in a block-like form within the vehicle stiffness parameter matrix: ; Specifically, based on the tire dynamic deformation vector in S11 and a tire stiffness matrix consisting of six tire stiffness coefficients. Constructing the elastic potential energy of the tire: ; Among them, with The corresponding elastic action forms the tire stiffness term in the vehicle stiffness parameter matrix, which is related to road surface roughness. The corresponding function is used to construct the generalized input matrix of the road surface in S142.

[0038] Specifically, based on the natural circular frequencies obtained from S113, modal elastic potential energy is constructed, and the modal stiffness matrix of the flexible working arm is obtained: ; Gravitational potential energy is constructed based on the mass parameters of the boom and boom tip, and after linearization near the static equilibrium posture, it forms the gravity tangential stiffness correction corresponding to the rigid body motion of the vehicle body. Gravitational stiffness coupling term between rigid body motion and flexible modal motion of the vehicle body and the gravitational tangential stiffness correction corresponding to flexible modal motion. ; Combining the above stiffness effects according to the generalized displacement relationship, we obtain the vehicle stiffness parameter matrix. : ; in, This is the suspension stiffness matrix composed of six suspension stiffness coefficients. Correction for the gravitational tangential stiffness corresponding to the rigid body motion of the vehicle body. This is the gravitational stiffness coupling term between the rigid body motion and the flexible modal motion of the vehicle body. This is a tire stiffness matrix consisting of six tire stiffness coefficients. The modal stiffness matrix of the flexible working arm. This is a correction for the gravitational tangential stiffness corresponding to flexible modal motion.

[0039] S13. Based on the three-dimensional rectangular coordinate system of the vehicle body, the static equilibrium point under the combined action of vehicle gravity, working arm gravity and average injection reaction force is selected as the static equilibrium state, and the injection reaction force increment near the static equilibrium state is calculated.

[0040] S131. Based on the vehicle body's fixed three-dimensional rectangular coordinate system, and combining the quantitative relationship between the effective cross-sectional area of ​​the fire monitor nozzle and the jet volume flow rate, the total jet reaction force of the fire monitor is obtained. : ; in, The density of the sprayed medium, This refers to the injection volumetric flow rate; The effective cross-sectional area of ​​the fire monitor nozzle. The nominal spray direction unit vector of the fire monitor.

[0041] The nominal spray direction unit vector of the fire monitor is: ; in, The fixed nominal spray elevation angle of the fire monitor relative to the center of gravity of the vehicle body. This is the fixed nominal yaw angle of the fire monitor's spray direction relative to the vehicle's center of gravity.

[0042] S132. Select a unit time interval, take the static equilibrium point under the combined action of vehicle weight, boom weight, and average injection reaction force as the static equilibrium state, and calculate the average injection reaction force. : ; in, Let T be the starting point of a unit time interval, and T be the unit time. This represents the end point of a unit time interval.

[0043] S133. Calculate the difference between the total and average jet reaction forces of the fire monitor to obtain the jet reaction force increment relative to the static equilibrium state. : ; S14. Based on the principle of virtual work, combined with the suspension geometric mapping relationship, tire dynamic deformation vector and the position and direction of the arm end injection reaction force, construct the corresponding generalized input matrix respectively.

[0044] Specifically, the generalized input matrix includes the generalized input matrix of the active suspension, the generalized input matrix of the road surface, and the generalized input matrix of the injection reaction force.

[0045] S141. Based on the position and direction of action of the six active suspension actuators, the six-channel active suspension action force is mapped to the vehicle body rigid body channel and the six wheel end channels through the virtual work relationship to obtain the generalized input matrix of the active suspension. : ; in, It is a sixth-order identity matrix, representing that the six active suspensions directly act on the six corresponding unsprung wheel ends; The zero matrix is ​​matched with the dimension of the flexible modal coordinates, indicating that the active suspension does not directly act on the flexible modal coordinates of the working arm.

[0046] S142. Based on the relative relationship between wheel end displacement and road surface roughness in tire dynamic deformation, the tire elastic effect is expressed as the influence of wheel end displacement on the vehicle stiffness and the external excitation effect of road surface roughness on the system. The effect corresponding to wheel end displacement is included in the vehicle stiffness parameter matrix, and the effect corresponding to road surface roughness forms the road surface generalized input matrix. Specifically, the construction process involves inputting terms related to wheel-end displacement into the vehicle stiffness parameter matrix K constructed in S123, and moving terms related to road surface input to the right side of the equation to form the generalized road surface input matrix. : ; in, for The zero-order matrix indicates that road surface roughness does not directly affect the generalized coordinates of the vehicle body rigid body; This is a tire stiffness matrix consisting of six tire stiffness coefficients; The zero matrix is ​​matched to the dimension of the flexible modal coordinates, indicating that road surface unevenness does not directly affect the flexible modal coordinates of the boom.

[0047] S143. Based on the vehicle body's fixed three-dimensional Cartesian coordinate system, quantify the position of the boom tip relative to the vehicle body's center of mass. Based on the virtual work done by the boom tip's jet reaction force on the vehicle body's vertical, pitch, roll, and flexible modes, obtain the generalized input matrix of the jet reaction force. : ; in, Let x be the position coordinate of the arm tip relative to the center of mass of the vehicle body along the x-axis in the three-dimensional Cartesian coordinate system fixed to the vehicle body. Let be the position coordinates of the arm tip relative to the center of mass of the vehicle body along the y-axis in the three-dimensional Cartesian coordinate system fixed to the vehicle body. Let be the position coordinates of the arm tip relative to the center of mass of the vehicle body along the z-axis in the three-dimensional Cartesian coordinate system fixed to the vehicle body. for The zero-order matrix indicates that the jet reaction force at the arm end does not directly act on the six unsprung wheel ends. For the arm-end normalized mode function vector, The flexible deformation direction of the working arm is represented by the unit direction vector in the three-dimensional Cartesian coordinate system fixed to the vehicle body.

[0048] in, , The nominal amplitude of the working arm.

[0049] S15. Based on the vehicle parameter matrix obtained in S12 and the generalized input matrix obtained in S14, and combined with the injection reaction force increment obtained in S13, a rigid-flexible coupled dynamic model is constructed using the Lagrange method. After first-order state space transformation and mass matrix transformation, the rigid-flexible coupled state matrix and state input matrix are obtained.

[0050] Specifically, the generalized coordinate vector of the rigid body of the vehicle body defined by S11 is adopted. Vertical displacement vectors of the six unsprung wheel ends and the flexible modal coordinate vector of the working arm Construct the system's generalized coordinate vector : ; The rigid-flexible coupling dynamic model is as follows: ; in, The input vector is composed of the power components of the six active suspensions. The road surface input vector, For the dynamic increment of jet reaction force, For the system's generalized velocity vector, Let be the system's generalized acceleration vector.

[0051] Specifically, the rigid-flexible coupling state vector is constructed using the system's generalized coordinate vector and its corresponding generalized velocity. : ; Based on the rigid-flexible coupling state vector, the above second-order rigid-flexible coupling dynamic equations are transformed into a first-order state-space form; whereby the vehicle mass parameter matrix is ​​used. The inverse matrix is ​​used to perform mass matrix transformations on the stiffness term, damping term, and generalized input terms of active suspension, road roughness, and injection reaction force, respectively, to obtain the rigid-flexible coupled state-space model. : ; in, The state matrix is ​​a rigid-flexible coupling matrix. This is the input matrix for the active suspension state. The input matrix is ​​the road surface state. The input matrix is ​​the injection reaction force state matrix.

[0052] Among them, the rigid-flexible coupling state matrix for: ; Active suspension state input matrix Road surface state input matrix and the injection reaction state input matrix They are respectively: ; S2. The vehicle's own sensors are used to collect vehicle-side measurement signals during the vehicle's driving process. The injection reaction force increment is used as a known disturbance input to construct a state observer and calculate the rigid-flexible coupling state estimation vector.

[0053] Vehicle-side measurement signals include vehicle-side measurement vectors and vehicle-side measurement matrices, and their relationship is as follows: ; in, This is the vehicle-side measurement vector, and it is a 10-dimensional real vector. The vehicle body pitch angle, The body roll angle, , Angular velocity, This is the suspension dynamic travel vector. for 3D vehicle-side measurement matrix For rigid-flexible coupling state vectors, This is a 10-dimensional measurement noise vector.

[0054] Specifically, in actual operation, the vehicle pitch angle, vehicle roll angle, and their angular velocity are obtained by the vehicle inertial measurement unit; the six suspension travels are obtained by suspension displacement sensors installed between the vehicle body and the corresponding unsprung wheel ends; the vehicle side measurement matrix uses ten vehicle side measurement signals, with the first four measurements directly corresponding to the rigid-flexible coupling state vector. The vehicle pitch angle, roll angle, and angular velocity are measured in the six-channel system based on the suspension dynamic travel relationship in S11. Sure.

[0055] Using the injection reaction force increment as a known disturbance input, and combining the vehicle-side measurement signal with the rigid-flexible coupling state matrix and state input matrix, a state observer is constructed: ; in, This is the state estimation vector for rigid-flexible coupling. The first derivative of the state estimation vector of a rigid-flexible coupling system with respect to time. This is the input matrix for the active suspension state. The input matrix is ​​the injection reaction force state. The observer gain matrix; This is the observation residual between the measured output and the estimated output.

[0056] In practice, the rigid-flexible coupling state estimation vector calculated by the observer is... The complete state estimation vector is obtained by integrating and updating over time. Observer gain matrix Based on the rigid-flexible coupling state matrix Measurement matrix The settings for process disturbances and measurement noise are obtained offline using a continuous-time linear quadratic estimation method; before designing the observer, the matrix pairs are... Perform observability or detectability verification. The observer uses a six-channel active suspension as the actual power source. As a known control input. When the injection reaction force increment can be calculated based on the injection flow rate, nozzle area, and injection direction, the injection reaction force increment will be... As a known disturbance input. Due to the six-wheel road surface input vector. Difficult to measure directly, and not included in the observer equation. The project treats the six-wheel road surface input, jet reaction force calculation error, unmodeled dynamics, and other external disturbances as process disturbances.

[0057] S3. Based on the vehicle structure and operating parameters consistent with the rigid-flexible coupling dynamic model, the segmented flexible working arm is rigidly equivalent. The basic attitude of the whole vehicle and the suspension stability requirements are used as the benchmark control targets to construct the extended-dimensional rigid benchmark gain.

[0058] S31. Based on a coordinate system consistent with the rigid-flexible coupling dynamic model, the working arm is regarded as a rigid body. According to the total mass of the working arm, the position of the center of mass, the moment of inertia, the position of the jet reaction force, and the influence of the tangential stiffness of gravity, the state matrix of the rigid reference model is established. With active suspension as the power input matrix : in, for 3D rigid baseline model state matrix, for S32. Based on the influence of the vehicle's vertical displacement, vehicle pitch angle, vehicle roll angle, six wheel-end vertical displacements, and corresponding speed states in the rigid reference state vector on the vehicle's basic attitude and suspension stability, set the rigid reference state weight matrix, and set the action power weight matrix according to the usage requirements of the six-channel active suspension.

[0059] S33. Based on the state matrix and active suspension power input matrix obtained in S31, and the rigid reference state weight matrix and power input matrix set in S32, input them into the linear quadratic regulator to obtain the rigid reference gain offline.

[0060] ; Specifically, for 3D rigid reference gain matrix, This is for calculating the gain of a linear quadratic regulator. The rigid reference state weight matrix is... To construct the dynamic weight matrix.

[0061] S34. According to the arrangement order of the rigid-flexible coupling state vectors, zero-padding is performed on the corresponding columns of the flexible modes corresponding to the rigid reference gain matrix to obtain the extended-dimensional rigid reference gain with the same dimension as the rigid-flexible coupling state.

[0062] Due to rigid reference gain for 3D matrix, and rigid-flexible coupling state vector The dimensions are inconsistent, so they cannot be directly compared with... Dimensional pointing perception feedback gain Compare or subtract. Therefore, according to the rigid-flexible coupling state vector The arrangement order, in the rigid reference gain The corresponding flexible modal coordinate vector and flexible modal velocity vector Insert a zero column at the specified position to get... Dimensional expansion rigid reference gain The columns corresponding to the vehicle body and wheel end states are inherited from the rigid reference gain. The corresponding columns for flexible modal coordinates and flexible modal velocities are zero.

[0063] Dimension expansion processing is only used to make the rigid reference gain consistent with the dimension of the complete state, without changing the original rigid reference gain law.

[0064] Specifically, rigid reference gain Extended-dimensional rigid reference gain and direction-aware feedback gain All of these were obtained during the offline design phase of the controller, and the rigid arm vehicle baseline model and the complete rigid-flexible coupling model do not need to be run simultaneously when the vehicle is running online.

[0065] S4. Based on the rigid-flexible coupling state matrix and state input matrix, establish the jetting high and low direction pointing error according to the spatial projection relationship between the vehicle pitch angle, vehicle roll angle, nominal jetting yaw direction of the fire monitor and flexible rotation angle at the arm end; on the extended-dimensional rigid reference gain, combined with the high and low direction pointing error, utilize the dynamic coupling relationship between the rigid vehicle body and the segmented flexible working arm to construct a bounded pointing increment compensation.

[0066] S41. Based on the three-dimensional Cartesian inertial coordinate system and the vehicle-mounted three-dimensional rectangular coordinate system, and according to the projection relationship between the vehicle pitch angle, vehicle roll angle, fire monitor yaw direction, and boom end flexible rotation angle on the fire monitor's spray elevation direction, the spray elevation pointing error of the fire monitor relative to the nominal spray direction is obtained. : ; in, This refers to the deviation between the actual elevation angle of the fire monitor and the nominal spray elevation angle. The vehicle body pitch angle, For flexible angle at the arm end, The body roll angle, The first geometric projection coefficient, , which is the second geometric projection coefficient.

[0067] Specifically, the vehicle pitch angle and roll angle are obtained from the rigid-flexible coupling state of the flexible boom, and the flexible rotation angle at the boom end is... The fixed nominal yaw angle of the fire monitor is obtained by combining the slopes of each modal function at the end of the working arm in S12 with the corresponding flexible modal coordinates. Determined by the current spraying operation posture; This refers to the pointing error component caused by the vehicle body pitch motion in the direction of jet elevation; This refers to the pointing error component caused by the vehicle body tilting motion in the injection height direction; This represents the pointing error component caused by the flexible rotation angle at the arm tip in the jet height direction. First geometric projection coefficient. Second geometric projection coefficient The fixed nominal yaw angle is determined by the fire monitor.

[0068] S42. Based on the linear relationship between normalized energy and rigid-flexible coupling state vector, and combined with the projection relationship of injection height and low pointing error, construct the pointing performance output matrix; based on the normalized modal function and rigid-flexible coupling state vector, obtain the first mode arm end deflection and the first mode arm end flexible rotation angle; combined with the projection relationship of injection height and low pointing error, obtain the first mode pointing proxy quantity; and construct the pointing performance output matrix.

[0069] S421. Based on the normalized mode function obtained in S1, and from the rigid-flexible coupling state vector... Extract the coordinates of the first flexible mode and combine them to obtain the deflection at the end of the first mode arm. and the flexible rotation angle at the end of the first modal arm : ; ; in, This represents the function value of the first-order normalized mode function at the end of the working arm. Let be the slope of the first-order normalized mode function at the end of the working arm. The coordinates of the first flexible mode extracted from the rigid-flexible coupling state vector.

[0070] S422. Based on the projection relationship of the jet elevation and elevation pointing errors, the vehicle pitch angle, vehicle roll angle, fire monitor yaw direction, and first mode arm end flexible rotation angle are combined to obtain the target first mode pointing proxy quantity. : ; in, To fix the nominal yaw angle for the fire monitor.

[0071] Specifically, the injection elevation and elevation pointing error is used to reflect the changes in injection elevation and elevation pointing caused by the vehicle body attitude and the first-order flexible mode of the target.

[0072] S423. Normalize the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing proxy quantity using non-zero reference quantities to obtain normalized energy. Based on the linear relationship between normalized energy and the rigid-flexible coupling state vector, construct the pointing performance output matrix. : ; in, For standard partial derivatives, This serves as the non-zero normalized reference value for the deflection at the end of the first mode arm. This is a non-zero normalized reference value for the flexible rotation angle at the end of the first mode arm. It is the non-zero normalized reference value for the first mode pointing to the surrogate quantity.

[0073] Specifically, for The pointing performance output matrix has three rows representing the normalized linear mapping relationships between the elastic deflection at the first modal arm end, the flexible rotation angle at the first modal arm end, and the pointing surrogate quantity of the first modal relative to the rigid-flexible coupling state. Under a fixed operating attitude, the nominal yaw angle is... All normalized reference quantities are known constants.

[0074] S43. Set the basic state weight matrix based on the basic control requirements of vehicle posture, wheel ends, suspension travel, and tire dynamic deformation; set the output pointing performance weight matrix based on the importance of the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing proxy quantity; map the pointing performance weights to the rigid-flexible coupling state space using the pointing performance output matrix, and add it to the basic state weight matrix to obtain the pointing perception state weight matrix. : ; in, The basic state weight matrix; The pointing performance output matrix is ​​used to establish a normalized linear mapping relationship between the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing surrogate quantity and the rigid-flexible coupling state vector; This is the performance weight matrix, used to set the importance of the above three performance parameters; This refers to the state weight increment formed after mapping the performance weights to the rigid-flexible coupled state space.

[0075] S44. Input the rigid-flexible coupling state matrix and state input matrix obtained in S1, and the pointing sensing state weight matrix obtained in S32, into the linear quadratic regulator to obtain the pointing sensing feedback gain offline. : ; in, for Dimensional pointing to the sensing feedback gain matrix.

[0076] S45. Generate bounded pointing increment compensation based on the difference between the pointing sensing feedback gain and the extended-dimensional rigid reference gain. : ; S5. Based on the rigid-flexible coupling state estimation vector, the extended-dimensional rigid reference gain and the bounded pointing increment compensation are fused together, and a six-channel actual driving force is formed within the active suspension execution constraint range and fed back to the state observer, so that the active suspension can correct the injection direction while maintaining the vehicle attitude stability, thus forming a closed-loop control of the active suspension.

[0077] Based on the extended-dimensional rigid benchmark gain and estimated state, the fundamental dynamics are generated. : ; The bounded pointing increment compensation is multiplied by the rigid-flexible coupling state estimation vector, and after adjusting the pointing increment compensation coefficients and limiting the amplitude of each channel, a six-channel pointing increment compensation vector is obtained. : ; in, This refers to the incremental compensation coefficient; The upper limit of the directional increment compensation amplitude for a single active suspension actuator; To restrict the input of each of the six channels to a range Inside.

[0078] The basic working force and the six-channel directional incremental compensation vector are added together, and constraints on the rate of change of working force and the final amplitude are applied to obtain the actual working force of the six channels. : ; in, To limit the rate of change of dynamics for each channel; The upper limit of the rate of change of the actuating force of a single active suspension actuator; To ultimately perform channel-by-channel saturation processing; This represents the upper limit of the final actuation force amplitude of a single active suspension actuator.

[0079] The actual dynamic force vector is fed back to the state observer, enabling the active suspension to correct the injection direction while maintaining vehicle attitude stability, thus forming a closed-loop control of the active suspension.

[0080] like Figure 3 and Figure 4 As shown, under the combined action of random road surface and jet reaction force, and Figure 8 and Figure 9 As shown, under the combined effects of a unilateral convex road surface and injection reaction force, both rigid reference feedback control and the control method of this invention can reduce the vehicle pitch and roll angle responses compared to passive suspension. Even after applying bounded pointing increment compensation, the vehicle pitch and roll responses remain within a controlled range, indicating that this invention, by increasing the injection pointing and flexible response control targets of the working arm, can still maintain the basic attitude stability of the vehicle.

[0081] like Figure 5 and Figure 10As shown, the spray elevation and directional pointing errors corresponding to the control method of the present invention are smaller than those of passive suspension and rigid reference feedback control. This result indicates that incorporating vehicle pitch, vehicle roll, arm-end flexible angle, and first-mode pointing proxy into the pointing perception performance index enables the active suspension control target to be directly correlated with the fire monitor's spraying task, and further improves the stability of the spraying direction while in motion.

[0082] like Figure 6 and Figure 11 As shown, the state observer's estimation of the first-order flexible mode coordinates can follow the corresponding output of the complete rigid-flexible coupled dynamic model. This result indicates that an effective estimate of the target's first-order flexible mode state can be obtained using ten vehicle-side measurement signals, and the working arm strain signal or arm end motion signal is not required as a feedback signal for the controller in this embodiment.

[0083] like Figure 7 and Figure 12 As shown, the incremental compensation for the six-channel steering is limited to a set amplitude of 3kN, and the actual working force of the six-channel active suspension does not exceed the final amplitude limit of 12kN. This result indicates that the present invention can improve vehicle attitude, injection steering, and arm end dynamic response while constraining the additional working force demand generated by steering correction, thus meeting the active suspension actuation capability requirements.

[0084] On the other hand, the present invention provides a system for a rigid-flexible coupling active suspension control method for a fire truck operating while in motion, which includes: a rigid-flexible coupling model construction module, a spray direction measurement construction module, a vehicle side state observation module, a rigid reference feedback module, a direction increment compensation module, and a six-channel active suspension actuation module.

[0085] The rigid-flexible coupling model building module is used to establish a rigid-flexible coupling dynamic model of the flexible boom vehicle under the dynamic incremental action of the jet reaction force and to normalize it to obtain the rigid-flexible coupling state matrix and state input matrix.

[0086] The spray pointing accuracy module establishes the spray elevation and elevation pointing error based on the spatial projection relationship between the vehicle pitch angle, vehicle roll angle, nominal spray yaw direction of the fire monitor, and flexible rotation angle at the boom end.

[0087] The vehicle-side state observation module includes a dynamic increment unit for injection reaction force and a state observer unit. The dynamic increment unit for injection reaction force calculates the dynamic increment of injection reaction force based on injection flow rate, nozzle area, and injection direction. The state observer unit calculates the rigid-flexible coupling state estimation vector based on vehicle pitch angle, vehicle roll angle, vehicle angular velocity, six suspension travel distances, six channels of actual action force, and the dynamic increment of injection reaction force.

[0088] The rigid reference feedback module is used to construct the rigid reference gain and expand its dimension to obtain the expanded rigid reference gain.

[0089] The pointing feedback gain module includes a feedback gain unit and a pointing increment compensation module. The feedback gain unit constructs the pointing sensing feedback gain based on the high and low pointing errors, the rigid-flexible coupling state matrix, and the state input matrix. The pointing increment compensation module generates bounded pointing increment compensation based on the difference between the pointing sensing feedback gain and the extended-dimensional rigid reference gain.

[0090] The six-channel active suspension actuation module integrates the extended-dimensional rigidity reference gain with the bounded pointing increment compensation to form a six-channel actual actuation force, which is fed back to the state observer. This enables the active suspension to correct the injection direction while maintaining vehicle attitude stability, thus forming a closed-loop control of the active suspension.

[0091] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A rigid-flexible coupling active suspension control method for a mobile aerial ladder fire truck, characterized in that, It includes: S1. Analyze the contribution relationship between the rigid body, unsprung wheel ends, and segmented flexible working arm to the energy of the vehicle system. The incremental jet reaction force is constructed by analyzing the fire monitor in both its operational and static equilibrium states. The relationship between the active suspension's dynamics, road surface excitation, and the incremental injection reaction force on the virtual work done by the external forces is analyzed. A rigid-flexible coupling dynamic model is constructed using the Lagrange method. After first-order state space transformation and mass matrix transformation, the rigid-flexible coupling state matrix and state input matrix are obtained. S2. The vehicle side measurement signals during the vehicle's driving process are collected using the vehicle's own sensors. The injection reaction force increment is used as a known disturbance input to construct a state observer and calculate the rigid-flexible coupling state estimation vector. S3. Based on the vehicle structure and operating parameters consistent with the rigid-flexible coupling dynamic model, the segmented flexible working arm is rigidly equivalent. The basic attitude of the whole vehicle and the suspension stability requirements are used as the benchmark control targets to construct the extended-dimensional rigid benchmark gain. S4. Based on the rigid-flexible coupling state matrix and state input matrix, establish the jet elevation and elevation pointing error according to the spatial projection relationship between the vehicle pitch angle, vehicle roll angle, fire monitor nominal spray yaw direction and arm end flexible rotation angle. Based on the extended rigid reference gain, combined with the high and low pointing errors, a bounded pointing increment compensation is constructed by utilizing the dynamic coupling relationship between the rigid body and the segmented flexible working arm. S5. Based on the rigid-flexible coupling state estimation vector, the extended-dimensional rigid reference gain and the bounded pointing increment compensation are fused together, and a six-channel actual driving force is formed within the active suspension execution constraint range and fed back to the state observer, so that the active suspension can correct the injection direction while maintaining the vehicle attitude stability, thus forming a closed-loop control of the active suspension.

2. The rigid-flexible coupling active suspension control method for a mobile aerial spray fire truck according to claim 1, characterized in that, S1 includes: S11. Establish a three-dimensional Cartesian inertial coordinate system, a three-dimensional rectangular coordinate system fixed to the vehicle body, and a local axial coordinate system for the working arm; divide the working arm in the extended state into four segments along the axial direction, and obtain the natural circular frequency and normalized mode function through numerical eigenvalue analysis. S12. Based on the normalized modal function, suspension dynamic travel vector, suspension dynamic travel rate of change and tire dynamic deformation vector, calculate the vehicle system kinetic energy, suspension dissipated energy and system potential energy one by one, and construct the whole vehicle parameter matrix. S13. Based on the three-dimensional rectangular coordinate system of the vehicle body, select the static equilibrium point under the combined action of vehicle gravity, working arm gravity and average injection reaction force as the static equilibrium state, and calculate the injection reaction force increment near the static equilibrium state. S14. Based on the principle of virtual work, combined with the suspension geometric mapping relationship, tire dynamic deformation vector and the position and direction of action of the arm end injection reaction force, construct the corresponding generalized input matrix respectively; S15. Based on the vehicle parameter matrix obtained in S12 and the generalized input matrix obtained in S14, and combined with the injection reaction force increment obtained in S13, a rigid-flexible coupled dynamic model is constructed using the Lagrange method. After first-order state space transformation and mass matrix transformation, the rigid-flexible coupled state matrix and state input matrix are obtained.

3. The rigid-flexible coupling active suspension control method for a mobile aerial spray fire truck according to claim 2, characterized in that, The vehicle parameter matrix of S12 includes: the vehicle mass parameter matrix, the vehicle damping parameter matrix, and the vehicle stiffness parameter matrix; its steps include: S121. Based on the normalized mode function, the system kinetic energy is constructed by combining the rigid body mass and moment of inertia, the mass at the ends of the six unsprung wheels, the distributed mass of the working arm, and the equivalent concentrated mass at the arm end. The system kinetic energy is then rearranged according to the square and cross terms of each generalized velocity to obtain the vehicle mass parameter matrix including inertial coupling terms. : ； in, Here are the mass and moment of inertia matrices. The coupling matrix obtained from the cross-kinetic energy term. The mass matrix consists of the masses at the ends of six unsprung wheels. Let T be the modal mass matrix of the flexible working arm, and T be the transpose vector. S122. The suspension energy dissipation is constructed using the suspension dynamic travel change rate and the preset suspension damping coefficient. Modal damping is constructed by combining the natural circular frequency and the preset modal damping ratio. The two are then rearranged according to the generalized velocity to obtain the vehicle damping parameter matrix. : ； in, For geometric mapping matrix, This is the suspension damping matrix composed of six suspension damping coefficients. The modal damping matrix of the flexible working arm; S123. Based on the suspension dynamic travel vector and the tire dynamic deformation vector, construct the suspension elastic potential energy and the tire elastic potential energy respectively. Combine the natural circular frequency to construct the modal elastic potential energy. Construct the gravitational potential energy according to the mass parameters of the working arm and arm end. Linearize the corresponding generalized gravitational force near the static equilibrium posture to obtain the gravitational tangential stiffness correction. Organize the system potential energy according to the generalized displacement to obtain the vehicle stiffness parameter matrix including stiffness coupling terms. : ； in, This is the suspension stiffness matrix composed of six suspension stiffness coefficients. Correction for the gravitational tangential stiffness corresponding to the rigid body motion of the vehicle body. This is the gravitational stiffness coupling term between the rigid body motion and the flexible modal motion of the vehicle body. This is a tire stiffness matrix consisting of six tire stiffness coefficients. The modal stiffness matrix of the flexible working arm. This is a correction for the gravitational tangential stiffness corresponding to flexible modal motion.

4. The rigid-flexible coupling active suspension control method for a mobile aerial spray fire truck according to claim 2, characterized in that, S13 includes: S131. Based on the three-dimensional rectangular coordinate system fixed to the vehicle body, and combined with the quantitative relationship between the effective cross-sectional area of ​​the fire monitor nozzle and the spray volume flow rate, the total spray reaction force of the fire monitor is obtained. S132. Select a unit time interval, take the static equilibrium point under the combined action of vehicle weight, working arm weight and average injection reaction force as the static equilibrium state, and calculate the average injection reaction force. S133. Calculate the difference between the total jet reaction force and the average jet reaction force of the fire monitor to obtain the jet reaction force increment relative to the static equilibrium state.

5. The rigid-flexible coupling active suspension control method for a mobile aerial spray fire truck according to claim 2, characterized in that, The generalized input matrix of S14 includes: the generalized input matrix of the active suspension, the generalized input matrix of the road surface, and the generalized input matrix of the injection reaction force; its steps include: S141. Based on the position and direction of action of the six active suspension actuators, the six-channel active suspension action force is mapped to the rigid body channel of the vehicle body and the six wheel end channels through the virtual work relationship to obtain the generalized input matrix of the active suspension. S142. Based on the relative relationship between wheel end displacement and road surface roughness in tire dynamic deformation, the tire elastic effect is expressed as the influence of wheel end displacement on the vehicle stiffness and the external excitation effect of road surface roughness on the system. The effect corresponding to wheel end displacement is included in the vehicle stiffness parameter matrix, and the effect corresponding to road surface roughness forms the road surface generalized input matrix. S143. Based on the three-dimensional rectangular coordinate system fixed to the vehicle body, the position of the arm end relative to the center of mass of the vehicle body is quantified. According to the virtual work done by the arm end jet reaction force on the vertical, pitch, roll and flexible modes of the vehicle body, the generalized input matrix of the jet reaction force is obtained.

6. The rigid-flexible coupling active suspension control method for a mobile aerial spray fire truck according to claim 1, characterized in that, The relationship between the vehicle-side measurement signals in S2 is as follows: ; in, This is the vehicle-side measurement vector, and it is a 10-dimensional real vector. The vehicle body pitch angle, The body roll angle, , Angular velocity, This is the suspension dynamic travel vector. For vehicle-side measurement matrix, For rigid-flexible coupling state vectors, This is a 10-dimensional measurement noise vector.

7. The rigid-flexible coupling active suspension control method for a fire truck operating while in motion according to claim 1, characterized in that, S4 include: S41. Based on the three-dimensional Cartesian inertial coordinate system and the vehicle body fixed three-dimensional rectangular coordinate system, according to the projection relationship of the vehicle body pitch angle, vehicle body roll angle, fire monitor yaw direction and arm end flexible rotation angle in the fire monitor spray height direction, the spray height pointing error of the fire monitor relative to the nominal spray direction is obtained. S42. Based on the linear relationship between normalized energy and rigid-flexible coupling state vector, and combined with the projection relationship of injection height and low pointing error, construct the pointing performance output matrix; based on the normalized modal function and rigid-flexible coupling state vector, obtain the first mode arm end deflection and the first mode arm end flexible rotation angle; combined with the projection relationship of injection height and low pointing error, obtain the first mode pointing proxy quantity; and construct the pointing performance output matrix. S43. Set the basic state weight matrix based on the basic control requirements of vehicle posture, wheel ends, suspension travel, and tire dynamic deformation; set the output pointing performance weight matrix based on the importance of the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing proxy quantity; map the pointing performance weights to the rigid-flexible coupling state space using the pointing performance output matrix, and add it to the basic state weight matrix to obtain the pointing perception state weight matrix. ; ； in, Based on the basic state weight matrix, This refers to the performance output matrix. To point to the performance weight matrix, This refers to the state weight increment formed after mapping the performance weights to the rigid-flexible coupling state space; S44. Input the rigid-flexible coupling state matrix, the state input matrix, and the pointing sensing state weight matrix into the linear quadratic regulator, and obtain the pointing sensing feedback gain offline. S45. Generate bounded pointing increment compensation based on the difference between the pointing perception feedback gain and the extended dimension rigid reference gain.

8. The rigid-flexible coupling active suspension control method for a mobile aerial spray fire truck according to claim 7, characterized in that, S42 includes: S421. Based on the normalized mode function, extract the first flexible mode coordinates from the rigid-flexible coupling state vector, combine the two to obtain the first mode arm end deflection and the first mode arm end flexible rotation angle. S422. Based on the projection relationship of the jet elevation and elevation pointing errors, the vehicle pitch angle, vehicle roll angle, fire monitor yaw direction, and first mode arm end flexible rotation angle are combined to obtain the target first mode pointing proxy quantity. ; ； in, The vehicle body pitch angle, The body roll angle, To fix the nominal yaw angle of the fire monitor; S423. Normalize the first modal arm end deflection, the first modal arm end flexible rotation angle, and the first modal pointing proxy quantity using non-zero reference quantities to obtain normalized energy. Based on the linear relationship between normalized energy and rigid-flexible coupling state vector, construct the pointing performance output matrix.

9. The rigid-flexible coupling active suspension control method for a fire truck operating while in motion, as described in claim 1, is characterized in that, S5 include: Based on the extended-dimensional rigid benchmark gain and estimated state, the fundamental dynamics are generated. : ; in, For extended-dimensional rigid reference gain, This is the state estimation vector for rigid-flexible coupling. The bounded pointing increment compensation is multiplied by the rigid-flexible coupling state estimation vector, and after adjusting the pointing increment compensation coefficients and limiting the amplitude of each channel, a six-channel pointing increment compensation vector is obtained. : ; in, The upper limit of the directional increment compensation amplitude for a single active suspension actuator; To restrict the input of each of the six channels to a range Inside, This refers to the incremental compensation coefficient. This refers to the gain of the sensor feedback. The basic working force and the six-channel directional incremental compensation vector are added together, and constraints on the rate of change of working force and the final amplitude are applied to obtain the actual working force of the six channels. : ; in, This represents the upper limit of the final actuation force amplitude of a single active suspension actuator. To ultimately perform channel-by-channel saturation processing, This represents the upper limit of the rate of change of the actuating force of a single active suspension actuator. To limit the rate of change of dynamics for each channel.

10. The rigid-flexible coupling active suspension control method for a fire truck operating while in motion, as described in claim 7, is characterized in that... The pointing performance output matrix of S42 is: ; in, For standard partial derivatives, The first mode arm end deflection, For the flexible rotation angle at the end of the first modal arm, The target first modality points to the proxy quantity. This serves as the non-zero normalized reference value for the deflection at the end of the first mode arm. This is a non-zero normalized reference value for the flexible rotation angle at the end of the first mode arm. It is the non-zero normalized reference value for the first mode pointing to the surrogate quantity.

Images