Hydraulic limb leg unit flexible landing control method based on magnetorheological damper
By adopting a dual closed-loop control method based on magnetorheological dampers, the problems of insufficient damping adjustment and multi-terrain adaptability of hydraulic leg units in landing control are solved, achieving high compliance and high precision landing control, and optimizing energy consumption and terrain adaptability.
Patent Information
- Application Number
- CN202511456849.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-13
- Publication Date
- 2026-02-10
AI Technical Summary
Existing hydraulic leg units suffer from problems such as insufficient damping adjustment, inadequate dynamic modeling accuracy, poor coordination of control strategies, and weak adaptability to various terrains in landing control, resulting in difficulty in effectively buffering impact forces and achieving both trajectory accuracy and stability.
A dual closed-loop control method based on magnetorheological dampers is adopted. By establishing a rigid-flexible coupled dynamic model, optimizing parameters, and combining the coordinated control of the position outer loop and the force inner loop, the damping force is dynamically distributed, thereby realizing intelligent decision-making and energy consumption optimization of the hydraulic cylinder and the magnetorheological damper.
It achieves highly compliant landing, balances control precision and energy consumption optimization, has multi-terrain adaptive capability, and ensures smooth landing and motion coordination of mechanical structures.
Smart Images
Figure CN121500745A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of robot control technology, specifically to a compliant landing control method for a hydraulic leg unit based on a magnetorheological damper. Background Technology
[0002] In the field of robot control technology, hydraulic leg units are widely used in heavy-duty handling robots, outdoor mobile robots, and other equipment due to their advantages such as high load-bearing capacity and large output torque. During the operation of these robots, the impact during the landing phase of the leg unit can easily damage the mechanical structure and may also affect the operation accuracy. Therefore, compliant landing has become a core requirement for the control of hydraulic leg units.
[0003] However, existing hydraulic leg unit landing control technology has significant limitations: First, the damping adjustment methods are insufficient, mostly using passive damping elements or active damping control with fixed parameters, which cannot dynamically adapt to the impact requirements according to the real-time landing status. This results in either insufficient buffering capacity leading to excessive impact or excessive damping force causing motion lag. Second, the accuracy of dynamic modeling is insufficient, often ignoring the rigid-flexible coupling characteristics such as the elastic deformation of the hydraulic cylinder and the flexible buffering of the foot rubber pad. The parameters optimized based on such models deviate greatly from the actual working conditions, making it difficult to minimize the impact force from the source. Third, the coordination of control strategies is poor, mostly using single closed-loop control. Position closed-loop alone cannot buffer the impact, and force closed-loop alone is prone to trajectory deviation, making it difficult to balance trajectory accuracy and impact smoothness. Fourth, the adaptability to various terrains is weak, with control parameters often designed for specific terrains. When switching operating terrains, manual readjustment is required, which is cumbersome and inefficient.
[0004] Therefore, a compliant landing control method for hydraulic leg units based on magnetorheological dampers is proposed to address the above problems. Summary of the Invention
[0005] The purpose of this invention is to provide a compliant landing control method for a hydraulic leg unit based on a magnetorheological damper, so as to solve the problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] A compliant landing control method for a hydraulic leg unit based on a magnetorheological damper, wherein the hydraulic leg unit includes a hydraulic damping actuator, a servo valve, a current driver, and sensing elements, and the hydraulic damping actuator comprises a hydraulic cylinder and a magnetorheological damper connected in series and sharing a piston rod; the method includes the following steps:
[0008] S1. Establish a rigid-flexible coupled dynamic model and optimize parameters: simplify the landing process of the hydraulic leg unit into a mass-spring-damping system, where the hydraulic cylinder is equivalent to a spring module and the magnetorheological damper is equivalent to a variable damping module; optimize the model with the goal of minimizing the impact force at the foot end to obtain the optimal landing trajectory and the corresponding optimal stiffness and damping parameters.
[0009] S2. Trajectory Decomposition and Displacement Command Generation: The optimal landing trajectory is decomposed into the target displacement of the joints using the inverse kinematics model of the limb units;
[0010] S3, Dual Closed-Loop Coordinated Control: Based on the target displacement and optimal stiffness and damping parameters, the hydraulic damping actuator is subjected to dual closed-loop control of position and force.
[0011] Among these methods, the hydraulic cylinder is controlled by the outer position ring to track the target displacement;
[0012] The target damping force is output from the magnetorheological damper by controlling the inner force loop;
[0013] Specifically, the timing of the intervention or withdrawal of the magnetorheological damper is determined based on the piston rod motion state and oil pressure state of the hydraulic cylinder; and the target damping force is dynamically allocated based on the piston rod motion state information.
[0014] Specifically, the detected actual force signal is negatively fed back to the outer position loop, and a displacement compensation amount is generated through an impedance control strategy to correct the target displacement.
[0015] As a preferred option, in step S3, the conditions for deciding when to intervene or withdraw the magnetorheological damper are as follows:
[0016] When the direction of the piston rod's movement speed is opposite to the direction of the hydraulic force formed by the pressure difference between the oil at both ends of the piston, the magnetorheological damper is activated.
[0017] When the direction of the piston rod's movement speed is the same as the direction of the hydraulic force, the magnetorheological damper is controlled to stop working.
[0018] As a preferred option, the conditions for determining when to engage or disengage the magnetorheological damper are specifically determined by the following formula:
[0019] When satisfied At that time, the magnetorheological damper is engaged;
[0020] When satisfied At that time, the magnetorheological damper is deactivated;
[0021] in, The piston's speed. and These are the pressures at the inlet and outlet of the hydraulic cylinder, respectively. This represents the effective area of the piston.
[0022] As a preferred embodiment, in step S3, the dynamic allocation of the target damping force is specifically performed by calculating the target desired force of the magnetorheological damper using the following formula. : ,in, , , These are preset weighting coefficients; , , These are the displacement influence factor, velocity influence factor, and acceleration influence factor, respectively, which are calculated using the following formulas: , , ,in, The difference between the expected displacement and the actual displacement. and The preset maximum and minimum thresholds for displacement error; The speed of the piston rod. and Preset maximum and minimum thresholds for speed; The acceleration of the piston rod. and The preset maximum and minimum thresholds for acceleration.
[0023] As a preferred option, in step S3, the inner force loop controls the magnetorheological damper using a feedforward-feedback composite control method, specifically as follows:
[0024] The feedforward channel maps the base control current based on the target damping force;
[0025] The feedback channel collects the actual foot force and calculates the force error between it and the target damping force.
[0026] After the force error is processed by the controller, the basic control current is compensated and corrected to generate the final control current output to the magnetorheological damper.
[0027] As a preferred option, the rigid-flexible coupling dynamic model in step S1 is an equivalent collision model of a single mechanical leg, and its system dynamics are described by the following state equations: ,in, , , These are the equivalent compression amounts of the fuselage, foot rubber pads, and steel plates, respectively. , , The equivalent mass corresponding to the fuselage, foot rubber pads, and steel plate; The equivalent stiffness of the actuator; This is the equivalent damping of a magnetorheological damper; , These are the equivalent stiffness and damping of the rubber pad at the foot end, respectively. It is the impact force on the ground.
[0028] As a preferred option, in step S1, the terrain parameters of the rigid-flexible coupling dynamic model are adjusted to obtain the optimal stiffness and damping parameters and the optimal landing trajectory that are suitable for different landing surfaces, so that the hydraulic leg unit has the ability to adapt to multiple terrains and make compliant landing.
[0029] As can be seen from the technical solution provided by the present invention above, the beneficial effects of the compliant landing control method for a hydraulic leg unit based on a magnetorheological damper provided by the present invention are:
[0030] Achieving highly compliant landing: A rigid-flexible coupled dynamic model is constructed and optimized with the goal of minimizing the impact force on the feet. Combined with the dual closed-loop control of "force-position" coordination, the target displacement is corrected through feedback of actual force signals, which effectively buffers the landing impact, protects the mechanical structure of the limbs and the landing surface, and avoids rigid collision damage.
[0031] Balancing control precision and energy consumption optimization: The magnetorheological damper intelligently decides to intervene or withdraw based on the piston rod's motion state and the direction of hydraulic force, working only on demand to reduce energy consumption; at the same time, it dynamically distributes damping force based on displacement, velocity, and acceleration to achieve "resistance supply on demand," ensuring control precision while avoiding energy waste.
[0032] It has multi-terrain adaptive capability: by adjusting the terrain parameters of the rigid-flexible coupling dynamic model, the optimal stiffness and damping parameters and landing trajectory corresponding to different ground surfaces (such as hard ground and soft ground) can be obtained without manual adjustment, thus expanding the application scenarios of the robot in complex environments.
[0033] Ensuring high-precision motion coordination: The dual closed-loop control of "position outer loop + force inner loop" ensures that the hydraulic cylinder accurately tracks the target displacement of the joint, and also ensures that the magnetorheological damper stably outputs the target damping force. The two work together to avoid trajectory deviation and motion lag, and achieve smooth landing of limbs.
[0034] Strong dynamic response and robustness: The inner force loop adopts feedforward-feedback composite control. The feedforward provides the base current quickly, and the feedback compensates for the force error, shortening the damping force adjustment delay. It can effectively cope with the uncertainty in the landing process and ensure control stability. Attached Figure Description
[0035] Figure 1 This is a schematic diagram of the steps of a compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to the present invention. Detailed Implementation
[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0037] To better understand the above technical solutions, the following will provide a detailed description of the technical solutions in conjunction with the accompanying drawings and specific embodiments.
[0038] like Figure 1 As shown, this embodiment of the invention provides a compliant landing control method for a hydraulic leg unit based on a magnetorheological damper. The hydraulic leg unit includes a hydraulic damping actuator, a servo valve, a current driver, and a sensing element. The hydraulic damping actuator includes a hydraulic cylinder and a magnetorheological damper connected in series and sharing a piston rod. The method includes the following steps:
[0039] S1. Establish a rigid-flexible coupled dynamic model and optimize parameters: simplify the landing process of the hydraulic leg unit into a mass-spring-damping system, where the hydraulic cylinder is equivalent to a spring module and the magnetorheological damper is equivalent to a variable damping module; optimize the model with the goal of minimizing the impact force at the foot end to obtain the optimal landing trajectory and the corresponding optimal stiffness and damping parameters.
[0040] S2. Trajectory Decomposition and Displacement Command Generation: The optimal landing trajectory is decomposed into the target displacement of the joints using the inverse kinematics model of the limb units;
[0041] S3, Dual Closed-Loop Coordinated Control: Based on the target displacement and optimal stiffness and damping parameters, the hydraulic damping actuator is subjected to dual closed-loop control of position and force.
[0042] Among these methods, the hydraulic cylinder is controlled by the outer position ring to track the target displacement;
[0043] The target damping force is output from the magnetorheological damper by controlling the inner force loop;
[0044] Specifically, the timing of the intervention or withdrawal of the magnetorheological damper is determined based on the piston rod motion state and oil pressure state of the hydraulic cylinder; and the target damping force is dynamically allocated based on the piston rod motion state information.
[0045] Specifically, the detected actual force signal is negatively fed back to the outer position loop, and a displacement compensation amount is generated through an impedance control strategy to correct the target displacement.
[0046] In this embodiment, step S1 simplifies the landing process of the hydraulic leg unit into a calculable mass-spring-damping system by constructing an accurate rigid-flexible coupling dynamic model and optimizing core parameters. The optimal landing trajectory and corresponding stiffness and damping parameters are determined with the goal of minimizing the impact force at the foot end. Simultaneously, adaptive adaptation to different terrains is achieved, providing a reliable dynamic basis for subsequent trajectory decomposition and closed-loop control. The detailed steps are as follows:
[0047] Step S1-1: Simplification of the landing process and equivalence of system modules:
[0048] Based on the structural characteristics and landing mechanics of the hydraulic leg unit, its landing process is simplified into a mass-spring-damping system. The hydraulic cylinder primarily provides elastic support during landing, and its mechanical response conforms to the force characteristics of a spring module; therefore, the hydraulic cylinder is equivalent to a spring module. The magnetorheological damper can dynamically offset the impact energy during landing by adjusting the input current to change the damping coefficient. Its damping characteristics are controllable and variable; therefore, the magnetorheological damper is equivalent to a variable damping module. Through this equivalence, the complex landing process of the hydraulic leg unit is transformed into a standardized mechanical system that is easy to model and calculate, laying the foundation for subsequent dynamic analysis.
[0049] Step S1-2: Construction of the rigid-flexible coupling dynamic model:
[0050] The constructed rigid-flexible coupled dynamic model is an equivalent collision model of a single mechanical leg. The system dynamics of this model are described by the following state equations: ,in, , , These are the equivalent compression amounts of the fuselage, foot rubber pads, and steel plates, respectively. , , The equivalent mass corresponding to the fuselage, foot rubber pads, and steel plate; The equivalent stiffness of the actuator; This is the equivalent damping of a magnetorheological damper; , These are the equivalent stiffness and damping of the rubber pad at the foot end, respectively. The model represents the ground impact force; it can fully reflect the mechanical transmission relationship between the fuselage, foot rubber pads, steel plates and the ground during landing, demonstrating the characteristics of rigid-flexible coupling.
[0051] Step S1-3: Model parameter optimization with the goal of minimizing foot impact force:
[0052] The optimization objective of the model is determined to be minimizing the foot impact force, which is achieved by adjusting the stiffness and damping parameters in the model to minimize the ground impact force. Always kept at a minimum level; optimization targets include the equivalent stiffness of the actuator. Equivalent stiffness of foot rubber pads Equivalent damping of magnetorheological dampers Equivalent damping of foot rubber pads The landing process under different parameter combinations was simulated using dynamic simulation methods, and the ground impact force corresponding to each parameter combination was calculated. Numerical values, filtering out those that make The minimum parameter combination is the optimal stiffness and damping parameters; at the same time, the motion trajectory of the foot end of the hydraulic leg unit under the optimal parameter combination is recorded, and this trajectory is the optimal landing trajectory, ensuring that the impact force is controllable and minimized during landing.
[0053] Steps S1-4: Obtaining optimal parameters and trajectory for multi-terrain adaptive design:
[0054] To enable the hydraulic leg unit to adapt to different landing surfaces, the terrain parameters of the rigid-flexible coupling dynamics model are adjusted. These parameters include ground hardness coefficient, ground friction coefficient, and other parameters related to ground mechanical properties. For each target terrain, the corresponding terrain parameters are input into the rigid-flexible coupling dynamics model, and the parameter optimization process in steps S1-3 is repeated. The optimal stiffness and damping parameters that minimize the impact force at the foot end under that terrain are recalculated, and the optimal landing trajectory adapted to that terrain is determined. In this way, the correspondence between different terrains and the optimal stiffness and damping parameters and the optimal landing trajectory is established, enabling the hydraulic leg unit to have multi-terrain adaptive compliant landing capability, and to switch the corresponding optimal parameters and trajectory according to the actual landing surface.
[0055] In this embodiment, step S2 takes the optimal landing trajectory obtained in step S1 and, through the inverse kinematics model of the limb unit, transforms the overall optimal landing trajectory of the foot into the executable target displacement of each joint. This bridges the gap between "trajectory planning" and "execution control," providing precise joint-level control commands for the subsequent dual closed-loop control in step S3, ensuring coordinated movement of each joint of the hydraulic limb unit to achieve a compliant landing. The detailed steps are as follows:
[0056] Step S2-1: Extract key feature parameters of the optimal landing trajectory:
[0057] First, feature analysis is performed on the optimal landing trajectory output in step S1 to extract the core parameters of the trajectory. These parameters include the pose and motion state information of the foot throughout the landing process: the pose information refers to the three-dimensional position coordinates of the foot in the Cartesian coordinate system (denoted as ). , , , (For the landing process time variable) and the attitude angle of the foot relative to the fuselage (denoted as ) , , (These correspond to the rotation angles around the x, y, and z axes, respectively); motion state information refers to the foot's velocity at each moment. , , ) and acceleration , , The above parameters need to be extracted completely in time series to ensure coverage of the entire process from the start of the landing to the moment of stabilization, so as to provide a complete "foot target motion benchmark" for subsequent inverse kinematics calculations.
[0058] Step S2-2: Construct the inverse kinematic model of the hydraulic leg unit:
[0059] Based on the mechanical structure parameters of the hydraulic limb unit, an inverse kinematics model adapted to this unit is constructed. First, the mechanical structure characteristics of the hydraulic limb unit are clarified: this limb unit contains multiple motion joints (such as the hip joint, knee joint, etc., the specific number and type of joints are determined by the limb mechanical design), each joint is connected by linkages, and the motion of the hydraulic damping actuator (including hydraulic cylinders and magnetorheological dampers) directly drives the joint movements. The core of the inverse kinematics model is to establish a mathematical mapping relationship between "foot position" and "joint displacement"—that is, given the foot position parameters (…). , , The model is constructed by inversely calculating the target displacement of each joint; the model construction needs to be based on the geometric parameters of the limb links, including the length of each link (denoted as...). , … , The model includes the number of links, the installation coordinates of the joints, and the motion type of the joints (rotational or translating joints, with rotational joints using "rotation angle" as displacement and translating joints using "linear stroke" as displacement). Through homogeneous coordinate transformation (DH parameter method) or the kinematic equations of the linkage mechanism, the three-dimensional pose and attitude angle of the foot are transformed into the displacement equations of each joint, forming a complete inverse kinematic model of the limb unit, ensuring that the calculation accuracy of the model is consistent with the actual mechanical motion characteristics of the hydraulic limb unit.
[0060] Step S2-3: Perform trajectory decomposition calculation based on the inverse kinematics model:
[0061] The optimal landing trajectory extracted in step S2-1 is discretized according to the time series to obtain several discrete time nodes (denoted as ). , The number of discrete nodes is specified, and the node interval must meet the control accuracy requirements (usually on the order of milliseconds); for each time node... ), perform the following operations:
[0062] Read the foot pose parameters corresponding to this node. , , , , , ;
[0063] Substituting the above pose parameters into the inverse kinematics model constructed in step S2-2, and solving the displacement equations of the model, the displacement of each joint is calculated. The target displacement value at any given time—if it is a rotary joint, the target displacement is the joint rotation angle (denoted as ). , … , (Number of joints); if it is a prismatic joint, the target displacement is the linear travel of the joint (denoted as ). , … ;
[0064] Verify the rationality of the calculation results: Check whether the target displacement of each joint is within its mechanical motion range (e.g., the rotation angle does not exceed the maximum swing angle of the joint, and the stroke does not exceed the maximum extension of the actuator). If it exceeds the range, return to step S2-1 to fine-tune the foot position parameters of the node and recalculate until the mechanical constraints are met.
[0065] Step S2-4: Generate the joint target displacement command set:
[0066] The joint target displacements calculated in steps S2-3 for all discrete time nodes are organized into a target displacement command sequence for each joint in the format of "timestamp-joint number-target displacement value". For example, for the first rotary joint, its command sequence is as follows: For the first moving joint, the command sequence is as follows: ;
[0067] The instruction set must be identifiable and executable. In the subsequent step S3, the position outer loop control module can read the timestamp and target displacement value in the instruction set and drive each joint to move according to the instructions in real time, ultimately achieving accurate tracking of the foot to the optimal landing trajectory.
[0068] In this embodiment, the condition for deciding when to intervene or withdraw the magnetorheological damper in step S3 is as follows:
[0069] When the direction of the piston rod's movement speed is opposite to the direction of the hydraulic force formed by the pressure difference between the oil at both ends of the piston, the magnetorheological damper is activated.
[0070] When the direction of the piston rod's movement speed is the same as the direction of the hydraulic force, the magnetorheological damper is controlled to stop working.
[0071] The specific conditions for determining when to intervene or withdraw the magnetorheological damper are determined by the following formula:
[0072] When satisfied At that time, the magnetorheological damper is engaged;
[0073] When satisfied At that time, the magnetorheological damper is deactivated;
[0074] in, The piston's speed. and These are the pressures at the inlet and outlet of the hydraulic cylinder, respectively. This represents the effective area of the piston.
[0075] The dynamic allocation of the target damping force is specifically calculated using the following formula: the target desired force of the magnetorheological damper is calculated. : ,in, , , These are preset weighting coefficients; , , These are the displacement influence factor, velocity influence factor, and acceleration influence factor, respectively, which are calculated using the following formulas: , , ,in, The difference between the expected displacement and the actual displacement. and The preset maximum and minimum thresholds for displacement error; The speed of the piston rod. and Preset maximum and minimum thresholds for speed; The acceleration of the piston rod. and Preset maximum and minimum thresholds for acceleration;
[0076] In step S3, the inner force loop controls the magnetorheological damper using a feedforward-feedback composite control method, specifically as follows:
[0077] The feedforward channel maps the base control current based on the target damping force;
[0078] The feedback channel collects the actual foot force and calculates the force error between it and the target damping force.
[0079] After the force error is processed by the controller, the basic control current is compensated and corrected to generate the final control current output to the magnetorheological damper.
[0080] Furthermore, step S3, based on the joint target displacement generated in step S2 and the optimal stiffness and damping parameters obtained in step S1, achieves precise trajectory tracking of the hydraulic cylinder in the hydraulic damping actuator and dynamic output of the target damping force of the magnetorheological damper through dual closed-loop coordinated control of "position outer loop + force inner loop". Simultaneously, through damper intervention / exit decisions, target damping force distribution, and force feedback correction, it ensures a smooth landing process and minimal impact force for the hydraulic leg unit. The detailed steps are as follows:
[0081] Step S3-1: Building a dual closed-loop control framework:
[0082] First, a dual closed-loop control structure adapted to the hydraulic damping actuator is constructed, clarifying the control objects, inputs and outputs, and collaborative logic of the outer position loop and the inner force loop:
[0083] Control object division: The position outer loop uses the hydraulic cylinder as the control object. Its core task is to drive the hydraulic cylinder piston rod to move and drive the limb joints to track the target displacement generated in step S2; the force inner loop uses the magnetorheological damper as the control object. Its core task is to adjust the damping force of the magnetorheological damper so that its output matches the target damping force with the optimal stiffness.
[0084] Signal interaction relationship: The output of the outer position loop (the motion state of the hydraulic cylinder, such as piston rod displacement and speed) serves as one of the reference inputs of the inner force loop, used to determine the working timing of the magnetorheological damper; the actual damping force signal collected by the inner force loop is fed back to the outer position loop to correct the target displacement, forming a collaborative control closed loop of "trajectory tracking-damping adjustment-error correction";
[0085] Hardware adaptation: Connect the servo valve to the outer loop control circuit of the position, and control the oil flow and pressure of the hydraulic cylinder by adjusting the opening degree of the servo valve; connect the current driver to the inner loop control circuit of the force, and change the damping characteristics of the magnetorheological damper by adjusting the output current; at the same time, connect the sensing and detection elements (such as displacement sensor, speed sensor, pressure sensor, force sensor) to collect the status signals required for closed-loop control in real time.
[0086] Step S3-2: Position outer loop control hydraulic cylinder tracks target displacement:
[0087] The outer position ring takes the joint target displacement generated in step S2 as input, and drives the hydraulic cylinder to move by controlling the servo valve, so as to realize the piston rod tracking the target displacement. The specific process is as follows:
[0088] Displacement signal acquisition and error calculation: The actual displacement of the hydraulic cylinder piston rod (denoted as ) is acquired in real time by a displacement sensor. Read the joint target displacement (denoted as) in step S2 at the same time. ), calculate displacement error ;
[0089] Servo valve control signal generation: This includes generating displacement errors. The input position controller (usually a PID controller or model predictive controller) outputs a control signal (such as a voltage or current signal) to the servo valve based on the error characteristics. This signal is linearly related to the opening degree of the servo valve.
[0090] Hydraulic cylinder motion adjustment: The servo valve adjusts the opening degree according to the control signal, controls the flow rate and pressure of oil entering or leaving the rodless chamber and rod chamber of the hydraulic cylinder, changes the pressure difference on both sides of the piston in the hydraulic cylinder, and then drives the piston rod to extend and retract, causing the limb joints to move in the target displacement direction;
[0091] Real-time closed-loop correction: Continuously repeat the process of "displacement acquisition - error calculation - signal generation - motion adjustment" until the error between the actual displacement of the piston rod and the target displacement is corrected. The displacement is less than a preset threshold (usually in the millimeter range) to ensure that the hydraulic cylinder stably tracks the target displacement;
[0092] Step S3-3: Decision on the timing of magnetorheological damper intervention and withdrawal:
[0093] Based on the motion state of the hydraulic cylinder piston rod and the oil pressure state, it is determined whether the magnetorheological damper needs to be engaged to avoid increased impact due to conflict between the damping force and the direction of hydraulic cylinder movement. The specific process is as follows:
[0094] Status signal acquisition: The movement speed of the piston rod is acquired through a speed sensor (denoted as...). The velocity direction is positive when the piston rod extends and negative when it retracts; the pressure at the inlet and outlet of the hydraulic cylinder is collected by pressure sensors (denoted as...). The inlet pressure is... (This refers to the oil outlet pressure); read the effective area of the hydraulic cylinder piston (denoted as...). (Determined by the structural parameters of the hydraulic cylinder);
[0095] Judgment criteria calculation: Calculate the judgment index based on the collected parameters. ,in The hydraulic force generated by the pressure difference of the oil on both sides of the piston (the direction of the force is the same as the direction of the pressure difference, i.e.) The force of time pushes the piston rod to extend. (Timely retraction);
[0096] Work status switching: when the judgment indicator When the piston rod's movement speed is opposite to the direction of the hydraulic force (e.g., the hydraulic force pushes the piston rod out, but the piston rod actually retracts), the magnetorheological damper is activated to counteract the impact of the reverse movement by outputting damping force; when the judgment index... When the piston rod moves in the same direction as the hydraulic force, the hydraulic cylinder can autonomously track the trajectory and control the magnetorheological damper to stop working, thus avoiding additional damping force that increases energy consumption.
[0097] Step S3-4: Dynamically distribute the target damping force based on the piston rod's motion state:
[0098] When the magnetorheological damper is engaged, the target desired damping force is calculated based on the real-time motion state (displacement, velocity, acceleration) of the piston rod. The dynamic distribution of damping force is achieved through the following process:
[0099] Motion state parameter acquisition: The displacement error of the piston rod is calculated using a displacement sensor. , For the target displacement, (actual displacement); the piston rod speed is collected by a speed sensor. The acceleration of the piston rod is obtained by using an accelerometer or by differentiating the velocity signal. Call the preset displacement error threshold. , ), speed threshold , ), acceleration threshold , );
[0100] Influence factor calculation: Calculate the displacement influence factor separately. Speed Influence Factor Acceleration Influence Factor The formula is as follows: , , ,in, The values of are all in the range of [0, 1], which respectively reflect the degree of influence of displacement error, velocity, and acceleration on the target damping force;
[0101] Target expected damping force calculation: call preset weighting coefficients , , ( Adjustments can be made based on landing compliance requirements, such as increasing the landing clearance when the impact is significant. , Substitute the values into the formula to calculate the target desired damping force: ,Should This refers to the target damping force that the magnetorheological damper controlled by the inner loop needs to output.
[0102] Step S3-5: Feedforward-feedback composite control of the inner force loop:
[0103] The inner loop employs a feedforward-feedback composite control method to ensure that the magnetorheological damper accurately outputs the target damping force. The specific process is as follows:
[0104] The feedforward channel generates the basic control current: The "damping force-current" characteristic curve of the magnetorheological damper (i.e., the optimal control current relationship corresponding to different damping forces) is pre-established experimentally, based on the calculations in step 53-4. Look up the corresponding basic control current on the characteristic curve. The feedforward channel directly... Output to current driver;
[0105] Feedback channel force error acquisition: The actual output damping force of the magnetorheological damper is acquired through a force sensor. Calculation force error ;
[0106] Error compensation and final current output: force error Input force controller (usually a PID controller), the controller for... After proportional, integral, and derivative adjustments, the output compensation current is obtained. The final control current is obtained by superimposing the basic control current and the compensation current. and will The output is sent to a current driver, which drives a magnetorheological damper to adjust the damping force until... and The error is less than the preset threshold;
[0107] Step S3-6: Negative feedback of actual force signal and correction of target displacement:
[0108] To further improve landing compliance, the actual force signal is negatively fed back to the outer position loop, and the target displacement is corrected through an impedance control strategy. The specific process is as follows:
[0109] Actual force signal acquisition and feedback: The actual foot impact force during the landing process of the hydraulic leg unit is acquired through sensing and detection elements (such as foot force sensors). The signal is then used as a negative feedback signal and fed into the impedance control module of the outer position loop.
[0110] Impedance control generates displacement compensation: The impedance control module presets the desired impedance model (e.g., For desired stiffness For desired damping, (For the rate of change of displacement error), the actual impact force Compared with the desired impedance model, the displacement compensation amount that needs to be corrected is calculated. (like If the impact force is greater than the expected impact force, then A negative value indicates a reduction in target displacement to decrease impact; conversely, a positive value indicates a positive value.
[0111] Target displacement correction: adjust the displacement compensation amount Superimposed on the original target displacement generated in step S2 The corrected target displacement is obtained. and will As a new input position outer ring, it guides the hydraulic cylinder to adjust its movement trajectory, achieving "force-position" coordinated correction and further reducing landing impact.
[0112] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A compliant landing control method for a hydraulic leg unit based on a magnetorheological damper, wherein the hydraulic leg unit includes a hydraulic damping actuator, a servo valve, a current driver, and a sensing element, and the hydraulic damping actuator comprises a hydraulic cylinder and a magnetorheological damper connected in series and sharing a piston rod; characterized in that, The method includes the following steps: S1. Establish a rigid-flexible coupled dynamic model and optimize parameters: simplify the landing process of the hydraulic leg unit into a mass-spring-damping system, where the hydraulic cylinder is equivalent to a spring module and the magnetorheological damper is equivalent to a variable damping module; optimize the model with the goal of minimizing the impact force at the foot end to obtain the optimal landing trajectory and the corresponding optimal stiffness and damping parameters. S2. Trajectory Decomposition and Displacement Command Generation: The optimal landing trajectory is decomposed into the target displacement of the joints using the inverse kinematics model of the limb units; S3, Dual Closed-Loop Coordinated Control: Based on the target displacement and optimal stiffness and damping parameters, the hydraulic damping actuator is subjected to dual closed-loop control of position and force. Among these methods, the hydraulic cylinder is controlled by the outer position ring to track the target displacement; The target damping force is output from the magnetorheological damper by controlling the inner force loop; Specifically, the timing of the intervention or withdrawal of the magnetorheological damper is determined based on the piston rod motion state and oil pressure state of the hydraulic cylinder; and the target damping force is dynamically allocated based on the piston rod motion state information. Specifically, the detected actual force signal is negatively fed back to the outer position loop, and a displacement compensation amount is generated through an impedance control strategy to correct the target displacement.
2. The compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to claim 1, characterized in that: In step S3, the conditions for determining when to intervene or withdraw the magnetorheological damper are as follows: When the direction of the piston rod's movement speed is opposite to the direction of the hydraulic force formed by the pressure difference between the oil at both ends of the piston, the magnetorheological damper is activated. When the direction of the piston rod's movement speed is the same as the direction of the hydraulic force, the magnetorheological damper is controlled to stop working.
3. The compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to claim 2, characterized in that: The specific conditions for determining when to intervene or withdraw the magnetorheological damper are determined by the following formula: When satisfied At that time, the magnetorheological damper is engaged; When satisfied At that time, the magnetorheological damper is deactivated; in, The piston's speed. and These are the pressures at the inlet and outlet of the hydraulic cylinder, respectively. This represents the effective area of the piston.
4. The compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to claim 1, characterized in that: In step S3, the dynamic allocation of the target damping force specifically involves calculating the target desired force of the magnetorheological damper using the following formula. : ,in, , , These are preset weighting coefficients; , , These are the displacement influence factor, velocity influence factor, and acceleration influence factor, respectively, which are calculated using the following formulas: , , ,in, The difference between the expected displacement and the actual displacement. and The preset maximum and minimum thresholds for displacement error; The speed of the piston rod. and Preset maximum and minimum thresholds for speed; The acceleration of the piston rod. and The preset maximum and minimum thresholds for acceleration.
5. The compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to claim 1, characterized in that: In step S3, the inner force loop controls the magnetorheological damper using a feedforward-feedback composite control method, specifically: The feedforward channel maps the base control current based on the target damping force; The feedback channel collects the actual foot force and calculates the force error between it and the target damping force. After the force error is processed by the controller, the basic control current is compensated and corrected to generate the final control current output to the magnetorheological damper.
6. The compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to claim 1, characterized in that: The rigid-flexible coupling dynamic model in step S1 is an equivalent collision model of a single mechanical leg, and its system dynamics are described by the following state equations: ,in, , , These are the equivalent compression amounts of the fuselage, foot rubber pads, and steel plates, respectively. , , The equivalent mass corresponding to the fuselage, foot rubber pads, and steel plate; The equivalent stiffness of the actuator; This is the equivalent damping of a magnetorheological damper; , These are the equivalent stiffness and damping of the rubber pad at the foot end, respectively. It is the impact force on the ground.
7. The compliant landing control method for a hydraulic leg unit based on a magnetorheological damper according to claim 1, characterized in that: In step S1, by adjusting the terrain parameters of the rigid-flexible coupling dynamic model, the optimal stiffness and damping parameters and the optimal landing trajectory are obtained to adapt to different landing surfaces, so that the hydraulic leg unit has the ability to adapt to multiple terrains and make compliant landing.