An adaptive gait generation and trajectory tracking method for lower limb load exoskeleton robot

Abstract

The present invention discloses a method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot, the method comprising the following steps: step 1, human body information estimation: after the human body wears the exoskeleton, the human body information estimation and solution are performed according to the length information of the exoskeleton-related rods and the encoder information of the driving motor; step 2, exoskeleton gait generation and iterative optimization: according to the information related to the human body and the exoskeleton, a human-machine geometric kinematic model is constructed in combination with the human-exoskeleton coupling relationship; step 3, a compliant trajectory tracking control method: an exoskeleton dynamics model and a PID controller model are constructed, and the output weight of the model is dynamically adjusted according to the current motion state. The present invention can realize the functions of providing support and power assistance to the human body, and enhance the load and exercise capacity of the human body.

Description

Technical Field

[0001] The present invention belongs to the technical field of mechanical exoskeleton control, and in particular relates to an adaptive gait generation and trajectory tracking method for a lower limb load exoskeleton robot. Background Art

[0002] In the field of industrial manufacturing, a large number of workers move a large number of objects under long-term, high-frequency and high-load environments. Most of them have certain symptoms of strain and severe wear of the lower limb joints, which has a great impact on physical health. Wearing a lower limb exoskeleton robot can bear most of the load pressure for the human body, and continuously provide assistance during the carrying process, reducing the use of lower limb muscles, thereby reducing the stress on joints such as the hip and knee, effectively reducing human energy consumption and reducing the degree of joint wear.

[0003] Lower limb load exoskeleton robots can provide auxiliary power and certain support for the wearer, and are widely used in daily life, rescue and disaster relief, patrol exploration, industrial handling and other fields. At present, lower limb load exoskeletons have widespread problems such as poor support and assistance efficiency, stiff human-machine interaction, and strong interference with the human body. This paper proposes a control strategy based on human-heterogeneous exoskeleton end alignment, which can automatically adapt to people with different body shapes and generate matching trajectories according to different human gait characteristics. It is mainly analyzed from two aspects: motion gait generation and human information estimation. This gait generation method only requires the trajectory information of the human lower limb joints and the exoskeleton joints, does not rely on force sensors, has a wide range of applications and low development costs, can be widely used in a variety of heterogeneous exoskeleton robots, enhance the actual application efficiency of exoskeletons, and has certain engineering application value.

[0004] In summary, the prior art has the following problems:

[0005] (1) Mismatch and incoordination between human and exoskeleton motion: During the motion process, the lower limb exoskeleton robot has problems such as insufficient capture of motion feature information, long system control cycle, and imperfect control algorithm. It is difficult to match human motion in real time and is very likely to interfere with the human body during the motion process, hindering human motion and even causing potential harm to the human body.

[0006] (2) Existing algorithms are difficult to apply to a variety of exoskeletons: Most of the current control algorithms are designed based on specific lower limb exoskeletons. If the same model of exoskeleton is replaced, a large number of parameters need to be readjusted. Exoskeletons with different structures may even fail directly. The generalization and practicality are poor, and it is difficult to meet the actual application needs.

[0007] (3) Poor load-supporting effect: The efficiency of lower limb load-supporting exoskeletons is generally low, making it difficult to effectively support large mass loads, and the assisting effect on the human body is limited. Summary of the invention

[0008] The purpose of the present invention is to provide a method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot, which improves the matching and coordination of human-exoskeleton movements and enhances the supporting force effect.

[0009] In order to achieve the purpose of the present invention, the present invention provides a method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot, comprising the following steps:

[0010] Step 1: Human body information estimation: After wearing the exoskeleton, the human body stands still, and the human body information is estimated and solved based on the length information of the exoskeleton-related rods and the encoder information of the drive motor;

[0011] Step 2: Exoskeleton gait generation and iterative optimization: Based on the information about the human body and exoskeleton, the human-machine geometric kinematic model is constructed in combination with the human-exoskeleton coupling relationship;

[0012] Step 3: Compliant trajectory tracking control method: construct an exoskeleton dynamics model and a PID controller model, and dynamically adjust the output weight of the model according to the current motion state.

[0013] Compared with the prior art, the significant progress of the present invention lies in: (1) By modifying the length parameters of the thigh rod and the calf rod of the exoskeleton, the present invention can be applied to exoskeletons with different structures such as homogeneous or heterogeneous, and has a certain degree of intelligence and can identify human body related parameters; (2) Through the iterative update of trajectory adaptive dynamic, the present invention can optimize trajectory parameters during walking in the exoskeleton, generate a trajectory that better fits the motion characteristics of the human body, express the tracking error in the form of virtual interactive force, and dynamically update the given expected trajectory through multiple Gaussian iterations; (3) Through flexible trajectory tracking control, the present invention modifies the output torque in real time according to the motion characteristics of the human body, realizes flexible assistance, improves the assistance effect while reducing power consumption, and dynamically regulates the output torque according to the relative position of the lower limbs of the human body.

[0014] In order to more clearly illustrate the functional characteristics and structural parameters of the present invention, further description is given below in conjunction with the accompanying drawings and specific implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0016] Figure 1 is a flow chart of the steps of the present invention;

[0017] Figure 2 Schematic diagram of human-machine alignment in different postures of the lower limb heterogeneous load exoskeleton robot of the present invention;

[0018] FIG3 (a) is a schematic diagram of the initial trajectory of the exoskeleton hip joint of the present invention after trigonometric calculation;

[0019] FIG3( b ) is a schematic diagram of the initial trajectory of the exoskeleton knee joint of the present invention after trigonometric calculation;

[0020] Figure 4 It is a schematic diagram of the three-dimensional Gaussian function distribution of the present invention;

[0021] FIG5( a ) is a schematic diagram of the exoskeleton hip joint trajectory of the present invention after 1000 iterations;

[0022] FIG5( b ) is a schematic diagram of the exoskeleton knee joint trajectory of the present invention after 1000 iterations. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] The present invention provides a method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot, which can automatically adapt to wearers of different body shapes and dynamically adjust the output torque according to the walking process. Figure 1 , including the following steps:

[0025] Step 1: Human body information estimation: After wearing the exoskeleton, the human body stands still, and the human body information is estimated and solved based on the length information of the exoskeleton-related rods and the encoder information of the drive motor;

[0026] 1-1. Wear the exoskeleton, place the human hip joints at the same level, and the exoskeleton robot is tied to the human shoulders, waist, and ankles. Record the length information of the exoskeleton-related rods: exoskeleton thigh length , exoskeleton calf length , exoskeleton hip angle , exoskeleton knee angle ;

[0027] 1-2. According to the length information of the exoskeleton-related rods and the encoder information of the drive motor, the human body information is estimated and solved. The formula is shown as follows:

[0028] ;

[0029] in, is the total length of the lower limbs, which is the sum of the thigh length and the calf length. is the length of the human thigh, is the length of the human calf, It is the distance between the human hip joint and the exoskeleton hip joint in the horizontal plane.

[0030] Step 2: Exoskeleton gait generation and iterative optimization: Based on the information about the human body and exoskeleton, the human-machine geometric kinematic model is constructed in combination with the human-exoskeleton coupling relationship;

[0031] 2-1. Using the end-aligned trajectory generation method, the lower limb posture of the exoskeleton is quickly mapped according to the lower limb posture of the human body, and the exoskeleton joint trajectory is calculated based on the triangular relationship:

[0032] ;

[0033] in, and They are all intermediate variables in the calculation process and have no specific meaning. Here, two variables are needed as intermediate variables, so they are sorted in the order of Arabic numerals starting from 1 and 2. The order has no practical meaning and is only used to reduce the complexity of the formula expression. and are the human body hip joint angle and knee joint angle obtained by measurement, and They are the exoskeleton hip joint angle and knee joint angle obtained through calculation respectively;

[0034] 2-2. Construct an exoskeleton-human interaction model based on the impedance model and calculate the results For subsequent three-dimensional Gaussian distribution model calculation:

[0035] ;

[0036] in, is the difference between the expected trajectory and the actual trajectory, that is, the difference between the exoskeleton joint angle fed back in real time and the exoskeleton trajectory generated by the trajectory generation method using the end alignment in step 2-1, is the rate of change of the difference between the expected trajectory and the actual trajectory. The specific calculation formula is: ,in, and They are Moment and The difference between the exoskeleton joint angle fed back in real time and the exoskeleton trajectory generated by the end-aligned trajectory generation method in step 2-1, for Moment and The time interval between moments; and are the difference between the expected trajectory and the actual trajectory and the weight of the difference change rate;

[0037] 2-3. Construct a three-dimensional Gaussian distribution model:

[0038] ;

[0039] in, is the variance of the three-dimensional Gaussian distribution function, which is 10 here; and It is the joint angle set of the exoskeleton hip and knee joints.

[0040] Take the iterative process of the exoskeleton knee joint as an example:

[0041] The total number of design iterations is :for (i=1, i<=n, i++);

[0042] Calculate in Output results of the exoskeleton virtual interaction model of the first iteration :

[0043] ;

[0044] in, For the The difference between the expected trajectory and the actual trajectory of the exoskeleton knee joint in the iteration, For the The rate of change of the difference between the expected trajectory and the actual trajectory of the exoskeleton knee joint in the iteration;

[0045] Will Introduce the three-dimensional Gaussian model:

[0046] ;

[0047] in, For the The correction trajectory generated by the iterative process.

[0048] Update exoskeleton knee angle:

[0049] ;

[0050] in, Indicates The expected angle of the exoskeleton knee joint in the iteration, Indicates The desired angle of the exoskeleton knee joint for iteration .

[0051] Step 3, compliant trajectory tracking control method: construct an exoskeleton dynamics model and a PID controller model, and dynamically adjust the output weight of the model according to the current motion state;

[0052] 3-1. Use Lagrange equation method to establish the lower limb exoskeleton dynamics model;

[0053] The lower limb exoskeleton dynamic model:

[0054] ;

[0055] in, is the output torque after model calculation, and are the torque outputs of the exoskeleton knee and hip joints, is the dimensional mass matrix of the overall moment of inertia of the system, is the inertia matrix including the Coriolis and centripetal moments, is the system gravity compensation gravity matrix, is the friction compensation matrix of the friction force of the hip and knee joints of the system, and are the moments of inertia of the exoskeleton thigh and shank rods, For the exoskeleton thigh bar, is the mass of the exoskeleton shank bar, and are the lengths of the center of mass of the exoskeleton thigh and calf rods (the radius of each mass point on the rod relative to the rotation center), usually half the length of the corresponding rod. and are the angular velocities of the exoskeleton hip and knee joints, and are the angular accelerations of the exoskeleton hip and knee joints, is the intermediate variable, represents the set of hip and knee joints of the exoskeleton. represents the set of angular velocities of the exoskeleton hip and knee joints, represents the set of angular accelerations of the hip and knee joints of the exoskeleton;

[0056] Calculate intermediate variables and the mass matrix :

[0057] ;

[0058] in, is the acceleration due to gravity, and its value is 9.8;

[0059] Calculate the gravity matrix :

[0060] ;

[0061] Calculate the inertia matrix :

[0062] ;

[0063] in, The friction compensation matrix for the friction force of the hip and knee joints of the system;

[0064] Approximate the static friction and dynamic friction respectively, and adjust the corresponding compensation system according to the actual debugging experience:

[0065] ;

[0066] in, and They represent the static friction and dynamic friction compensation coefficients respectively, and need to be adjusted through repeated experiments based on the actual model of the exoskeleton.

[0067] 3-2. Design dynamic compliant control strategy and build PID controller model;

[0068] The dynamic compliance control strategy is:

[0069] ;

[0070] in, is the output torque calculated by the dynamic model, is the output torque calculated by the PID controller, and They are The current time and The error in time, , and are the parameters of the PID controller respectively.

[0071] 3-3. According to the motion state of the human body and the exoskeleton, the weight change model of the lower limb exoskeleton dynamics model and the PID controller model is designed, and the final output torque obtained by multiplying the lower limb exoskeleton dynamics model and the PID controller model by the corresponding weight is finally sent to the exoskeleton driver to complete the power assistance; wherein, the weight is calculated based on the height of the lower limb of the exoskeleton from the ground;

[0072] By analyzing the change in the height of the current exoskeleton from the ground, the weights of the dynamic model and the PID controller are dynamically allocated. The weights are calculated as:

[0073] ;

[0074] in, and They are the weights of the lower limb exoskeleton dynamics model and the PID controller model respectively, and the sum of the two weights is always 1.

[0075] The detailed operation process of this application system is as follows.

[0076] S1: Initialize the system, calibrate the sensors, perform power-on detection, calibrate all joint angle sensors to zero degrees, and check whether the motors feed back status information to the control board. After initialization is completed, estimate the human body information based on the exoskeleton joint information.

[0077] S2: According to the human-exoskeleton coupling relationship, a human-machine joint mapping geometric model is constructed. The human joint movement is obtained through the geometric model to obtain the corresponding joint angle of the exoskeleton; the data of the three coordinate axes in Figure 3 (a) are: human hip joint angle, human knee joint angle, and exoskeleton mapping hip joint angle, expressed as , , ; The data of the three coordinate axes in Figure 3 (b) are: human hip joint angle, human knee joint angle, exoskeleton mapping knee joint angle, expressed as , , ; The range b of the coordinate axis represents the range of motion of the human joints.

[0078] S3: Design a three-dimensional Gaussian distribution model, combined with Figure 4 , in the figure and The data on the coordinate axis has no practical meaning. The coordinate axis data is used to represent the final output data of the three-dimensional Gaussian distribution model. Next, the three-dimensional Gaussian distribution model is introduced into the gait iteration generation method. After 1000 iterations, the final generated trajectory is obtained. The data of the three coordinate axes in Figure 5 (a) are: human hip joint angle, human knee joint angle, and exoskeleton mapping hip joint angle, expressed as , , ; The data of the three coordinate axes in Figure 5 (b) are: human hip joint angle, human knee joint angle, exoskeleton mapping knee joint angle, expressed as , , ; The range b of the coordinate axis represents the range of motion of the human joints.

[0079] S4: Based on the given exoskeleton joint angles, trajectory tracking control is performed and the output torque is adjusted according to the current height of the lower limbs off the ground.

[0080] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0081] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot, characterized in that: The following steps are involved: Step 1, human body information estimation: after wearing the exoskeleton, the human body information is estimated and solved according to the length information of the exoskeleton-related rods and the encoder information of the drive motor; the human body information includes the total length of the human lower limbs, the length of the human thigh, the length of the human calf, and the distance between the human hip joint and the exoskeleton hip joint in the horizontal plane; Step 2: Exoskeleton gait generation and iterative optimization: Based on the geometric and kinematic parameters of the human body and exoskeleton, a human-exoskeleton coupling kinematic model is constructed; The model construction process includes: (1) using the end-aligned trajectory generation method to map the lower limb posture of the human body to the lower limb posture of the exoskeleton, and calculating the motion trajectory of the exoskeleton joint through geometric triangulation; (2) Based on the impedance control theory, a dynamic interaction model between the exoskeleton and the human body is established to describe the transmission and compensation of human-machine interaction forces; (3) A three-dimensional Gaussian distribution model is constructed to characterize the motion uncertainty of the human-exoskeleton system, and the accuracy and adaptability of gait generation are improved through iterative optimization methods; Step 3: Compliant trajectory tracking control method: construct an exoskeleton dynamics model and a PID controller model, and dynamically adjust the output weight of the model according to the current motion state.

2. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 1, characterized in that: The step 1 is specifically as follows: 1-1. Wear the exoskeleton, place the human hip joints at the same level, and the exoskeleton robot is tied to the human shoulders, waist, and ankles. Record the length information of the exoskeleton-related rods: exoskeleton thigh length , exoskeleton calf length , exoskeleton hip angle , exoskeleton knee angle ; 1-2. Estimation and calculation of human body information based on the length information of exoskeleton-related rods and the encoder information of the drive motor: ； in, is the total length of the lower limbs of the human body, is the length of the human thigh, is the length of the human calf, It is the distance between the human hip joint and the exoskeleton hip joint in the horizontal plane.

3. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 2, characterized in that: The step 2 is specifically as follows: 2-1. Using the end-aligned trajectory generation method, the lower limb posture of the exoskeleton is quickly mapped according to the lower limb posture of the human body, and the exoskeleton joint trajectory is calculated based on the triangular relationship: ； in, and are intermediate variables in the calculation process. and are the human body hip joint angle and knee joint angle obtained by measurement, and They are the exoskeleton hip joint angle and knee joint angle obtained through calculation respectively; 2-2. Constructing an exoskeleton-human interaction model based on an impedance model : ； in, is the difference between the expected trajectory and the actual trajectory, that is, the difference between the exoskeleton joint angle fed back in real time and the exoskeleton trajectory generated by the trajectory generation method using the end alignment in step 2-1, is the rate of change of the difference between the expected trajectory and the actual trajectory. The specific calculation formula is: ,in, and They are Moment and The difference between the exoskeleton joint angle fed back in real time and the exoskeleton trajectory generated by the end-aligned trajectory generation method in step 2-1, for Moment and The time interval between moments; and are the difference between the expected trajectory and the actual trajectory and the weight of the difference change rate; 2-3. Construct a three-dimensional Gaussian distribution model: ； in, is the variance of the three-dimensional Gaussian distribution function; and It is the joint angle set of the exoskeleton hip and knee joints.

4. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 3, characterized in that: The iterative process of the exoskeleton knee joint is: The total number of design iterations is :for (i=1, i<=n, i++); Calculate in Output results of the exoskeleton virtual interaction model of the first iteration : ； in, For the The difference between the expected trajectory and the actual trajectory of the exoskeleton knee joint in the iteration, For the The rate of change of the difference between the expected trajectory and the actual trajectory of the exoskeleton knee joint in the iteration; Will Introduce the three-dimensional Gaussian model: ； in, For the The correction trajectory generated by the iterative process; Update exoskeleton knee angle: ； in, Indicates The expected angle of the exoskeleton knee joint in the iteration, Indicates The desired angle of the exoskeleton knee joint for iteration .

5. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 4, characterized in that: The step 3 comprises the following steps: 3-1. Use Lagrange equation method to establish the lower limb exoskeleton dynamics model; 3-2. Design dynamic compliant control strategy and build PID controller model; 3-3. According to the movement state of the human body and the exoskeleton, the weight change model of the lower limb exoskeleton dynamics model and the PID controller model is designed, and the final output torque obtained by multiplying the lower limb exoskeleton dynamics model and the PID controller model by the corresponding weights is finally sent to the exoskeleton driver to complete the assistance.

6. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 5 can dynamically adjust the exoskeleton joint torque output change according to the height of the lower limbs of the human body from the ground, characterized in that: The step 3-1 is specifically as follows: The lower limb exoskeleton dynamic model: ； in, is the output torque after model calculation, and are the torque outputs of the exoskeleton knee and hip joints, is the dimensionless mass matrix of the overall moment of inertia of the system, is the inertia matrix including the Coriolis and centripetal moments, is the system gravity compensation gravity matrix, is the friction compensation matrix of the friction force of the hip and knee joints of the system, and are the moments of inertia of the exoskeleton thigh and shank rods, For the exoskeleton thigh bar, is the mass of the exoskeleton shank bar, and are the center of mass lengths of the exoskeleton thigh and shank rods, and are the angular velocities of the exoskeleton hip and knee joints, and are the angular accelerations of the exoskeleton hip and knee joints, represents the set of hip and knee joints of the exoskeleton. represents the set of angular velocities of the exoskeleton hip and knee joints, represents the set of angular accelerations of the hip and knee joints of the exoskeleton; Calculate intermediate variables and the mass matrix : ； Calculate the gravity matrix : ； in, is the acceleration due to gravity, and its value is 9.8; Calculate the inertia matrix : ； Adjust the corresponding compensation system: ； in, and Represent the static friction and dynamic friction compensation coefficients respectively.

7. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 6, characterized in that: The dynamic compliance control strategy of step 3-2 is: ； in, is the output torque calculated by the dynamic model, is the output torque calculated by the PID controller, and They are The current time and The error in time, , , are the parameters of the PID controller respectively.

8. The method for adaptive gait generation and trajectory tracking of a lower limb load exoskeleton robot according to claim 7, characterized in that: The step 3-3 analyzes the change in the height of the current exoskeleton from the ground, and dynamically allocates the weights of the dynamic model and the PID controller. The weights are calculated as: ； in, and They are the weights of the lower limb exoskeleton dynamics model and the PID controller model respectively, and the sum of the two weights is always 1.

Images