Self-balancing walking control method for rehabilitation exoskeleton under dynamic deformation

Abstract

The invention provides a self-balancing walking control method for rehabilitation exoskeleton under dynamic deformation. The self-balancing walking control method comprises the following steps: acquiring sole force and moment data when an exoskeleton robot executes walking actions; inputting the sole force and moment data into a feed-forward deformation compensator to obtain feed-forward compensation; the mass center active compliance controller carries out calculation based on sole force and moment data and an expected trajectory to obtain mass center compensation; determining a new expected posture of the robot according to feed-forward compensation and centroid compensation; the inverse kinematics solver calculates an instruction joint angle of each joint of the robot according to the new expected attitude; the exoskeleton robot performs walking and standing tasks based on the commanded joint angle. According to the method, by improving the deformation problem, the walking stability of the rehabilitation exoskeleton robot is improved.

Description

Technical Field

[0001] The invention relates to the technical field of robots, and in particular to a self-balancing walking control method of a rehabilitation exoskeleton under dynamic deformation. Background Art

[0002] In recent years, with the aging of the population and the aggravation of related diseases, the number of patients with lower limb dysfunction or loss, such as hemiplegia and quadriplegia, caused by factors such as stroke, accidents, and spinal cord injury has increased rapidly. The number of new stroke cases worldwide exceeds 13.7 million cases per year, becoming the third leading cause of disability in the world. Previous studies have shown that early intervention and goal-oriented training that gradually adapts to the patient's degree of injury and recovery stage can improve the functional prognosis of patients with lower limb weakness or loss of function. Therefore, people with motor function loss are increasingly dependent on various medical devices in their daily lives. As an auxiliary device, the lower limb exoskeleton robot can provide support and balance for patients with motor dysfunction, so that they can restore their normal walking ability. According to whether the lower limb exoskeleton robot has self-balancing ability, it can be divided into non-self-balancing exoskeleton robots and self-balancing exoskeleton robots. This patent focuses on the self-balancing lower limb exoskeleton robot (SBLLE). Since there is no external support device, the core control goal of SBLLEs is to maintain the balance of the human-machine hybrid system while carrying different patients.

[0003] The design limitations inevitably lead to insufficient rigidity of the exoskeleton robot, introducing unnecessary elasticity into the robot structure, which results in non-ideal deformation. The presence of such non-ideal deformation will prevent the self-balancing exoskeleton robot from accurately performing movements, making it prone to falling and possibly causing secondary injuries to the patient. Therefore, the existing exoskeleton robots have poor stability during walking. Summary of the invention

[0004] In view of this, the present invention provides a self-balancing walking control method of a rehabilitation exoskeleton under dynamic deformation to solve the above problems.

[0005] The present invention provides a self-balancing walking control method for a rehabilitation exoskeleton under dynamic deformation, comprising: collecting plantar force and torque data of an exoskeleton robot when performing a walking action; inputting the plantar force and torque data into a feedforward deformation compensator to obtain feedforward compensation; a center of mass active compliance controller performs calculations based on the plantar force and torque data and an expected trajectory to obtain center of mass compensation; determining a new expected posture of the robot according to the feedforward compensation and the center of mass compensation; an inverse kinematics solver calculates a command joint angle of each joint of the robot according to the new expected posture; and the exoskeleton robot performs walking and standing tasks based on the command joint angles.

[0006] In another implementation of the present invention, the feedforward compensation is expressed as:

[0007] Δ p T W = d T W - r T W

[0008] in, d T W Indicates the robot in the reference coordinate system Σ calculated offline in advance W Expecting attitude, r T W Represents the robot in the reference coordinate system Σ W The true attitude in.

[0009] In another implementation of the present invention, the feedforward deformation compensator is composed of an encoder and a decoder, which is used to establish a mapping relationship between the plantar force and torque data and the feedforward compensation, and the mapping relationship is expressed as:

[0010] Δ p T=f Δ (F)

[0011] Among them, Δ p T represents feedforward compensation, and F represents plantar force and torque data.

[0012] In another implementation of the present invention, it further includes: calculating according to the actual torque and the expected torque of the robot in the reference coordinate system to obtain a deviation formula between the actual position of the robot's center of mass and the expected position based on the torque data:

[0013] r τ B - d τ B =F ext ·z c

[0014] in, r τ B In the reference coordinate system Σ B The actual torque at the origin, In the reference coordinate system Σ B Desired torque at the origin, z c Represents the coordinate system Σ B The height of the centroid point.

[0015] In another implementation of the present invention, the center of mass active compliance controller calculates based on the plantar force and torque data and the expected trajectory to obtain center of mass compensation, including: the center of mass active compliance controller calculates based on the plantar force and torque data and the expected trajectory to obtain a deviation formula between the actual position of the robot's center of mass and the expected position based on the position data:

[0016]

[0017] in, r T represents the robot in the reference coordinate system Σ B The actual location in d T represents the robot in the reference coordinate system Σ B The desired position in the; combined with the deviation formula based on the torque data and the deviation formula based on the position data to calculate, obtain the center of mass compensation.

[0018] In another implementation of the present invention, the new desired posture is expressed as:

[0019]

[0020] in, d T W is the robot's expected posture calculated offline in advance, Δ p T represents feedforward compensation, ΔT W Denotes center of mass compensation.

[0021] The self-balancing walking control method of the rehabilitation exoskeleton under dynamic deformation of the present invention significantly improves the deformation problem through a feedforward deformation compensator; solves the strong parameter disturbance problem caused by different wearers in the exoskeleton robot hybrid system through a center of mass active compliance controller; enables the exoskeleton robot to achieve stable walking under dynamic deformation and carrying different subjects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. By reading the detailed description of the following implementations, the advantages and benefits of the solutions become clear to those skilled in the art. The drawings are only used to illustrate the preferred implementations and are not to be considered as limiting the present invention.

[0023] In the attached picture:

[0024] Figure 1 The figure is a flow chart of a method for controlling self-balancing walking of a rehabilitation exoskeleton under dynamic deformation according to an embodiment of the present invention.

[0025] Figure 2Schematic diagram of an exoskeleton robot system according to an embodiment of the present invention.

[0026] Figure 3 Schematic diagram of a robot compliance model according to an embodiment of the present invention.

[0027] Figure 4 FIG. 4 is a schematic diagram of a deformation mapping data acquisition platform according to an embodiment of the present invention.

[0028] Figure 5 Schematic diagram of a feedforward deformation compensator according to an embodiment of the present invention.

[0029] Figure 6 A schematic diagram of a mapping data set establishment process according to an embodiment of the present invention.

[0030] Figure 7 Schematic diagram of a walking experiment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and in detail 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, not all of the embodiments. All other embodiments obtained by ordinary technicians in the field based on the embodiments in the embodiments of the present invention should fall within the scope of protection of the embodiments of the present invention.

[0032] Figure 1 A schematic flow chart of a self-balancing walking control method for a rehabilitation exoskeleton under dynamic deformation provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, this embodiment mainly includes:

[0033] S101. Collect plantar force and torque data of the exoskeleton robot when it performs walking movements.

[0034] S102, inputting the plantar force and torque data into a feedforward deformation compensator to obtain feedforward compensation.

[0035] For example, Figure 2 As shown, the sensing system provides the plantar force and torque data obtained by the preceding F / T sensor to the feedforward deformation compensator (IFDC), and the IFDC predicts the next feedforward compensation.

[0036] S103, the center of mass active compliance controller calculates based on the plantar force and torque data and the expected trajectory to obtain center of mass compensation.

[0037] Exemplarily, the sensing system provides sensor data of real-time feedback of the state of the exoskeleton robot system to a center of mass active compliance controller (CACC), and the CACC calculates the center of mass compensation based on the desired trajectory generated by the trajectory generator and the current system state.

[0038] S104: Determine a new desired posture of the robot according to feedforward compensation and center of mass compensation.

[0039] S105. The inverse kinematics solver calculates the command joint angle of each joint of the robot according to the new desired posture.

[0040] S106. The exoskeleton robot performs walking and standing tasks based on the command joint angles.

[0041] The self-balancing walking control method of the rehabilitation exoskeleton under dynamic deformation of the present invention significantly improves the deformation problem through a feedforward deformation compensator; solves the strong parameter disturbance problem caused by different wearers in the exoskeleton robot hybrid system through a center of mass active compliance controller; enables the exoskeleton robot to achieve stable walking under dynamic deformation and carrying different subjects.

[0042] In another implementation of the present invention, the feedforward compensation is expressed as:

[0043] Δ p T W = d T W - r T W

[0044] in, d T W Indicates the robot in the reference coordinate system ∑ W Expecting attitude, r T W Represents the robot in the reference coordinate system ∑ W The true attitude in.

[0045] For example, Figure 4 As shown, due to deformation, ∑ O In the world coordinate system ∑ W Moved to ∑ O′ , the IFDC algorithm aims to convert ∑ O′ Attitude compensation return point ∑ O posture.

[0046] This deformation parameter can be directly used as the input of inverse kinematics, which can not only compensate for the problem of premature contact of the swing leg (deformation in the z direction), but also solve the problem of the robot tilting to one side of the swing leg (deformation in the y direction). For the self-balancing exoskeleton robot, the deformation parameter caused by the non-ideal characteristics changes with the change of the six-dimensional force / torque under the foot. The posture of the robot r T W Reflects the overall deformation parameters of the robot.

[0047] In another implementation of the present invention, the feedforward deformation compensator is composed of an encoder and a decoder, which is used to establish a mapping relationship between the plantar force and torque data and the feedforward compensation, and the mapping relationship is expressed as:

[0048] Δ p T=f Δ (F)

[0049] Among them, Δ p T represents feedforward compensation, and F represents plantar force and torque data.

[0050] For example, Figure 5 As shown in Figure 1, the network model of the feedforward deformation compensator consists of an encoder and a decoder. The network uses an efficient attention mechanism to focus on more prominent features in the data and capture long-term dependencies in the six-dimensional force dataset. The input of IFDC is F, which is the six-dimensional force / torque data under the feet, and the output of IFDC is Δ p T, the purpose is to establish a p The mapping relationship of T describes the conversion from the change of robot force to the change of robot posture.

[0051] In order to obtain the training data set and robot F r T W The real pose of the image is captured using a motion capture system. Figure 4 As shown in b, the motion capture system measures the posture of the reflective ball fixed in front of the exoskeleton. By tracking the posture changes of the reflective ball, the posture changes of the robot can be obtained, which will help collect the deformation dataset.

[0052] The Informer-based feedforward deformation compensator (IFDC) is used to effectively solve the dynamic deformation problem of the self-balancing exoskeleton robot human-machine hybrid system. The feedforward deformation compensator reduces the robot's position error by an average of 79.04% and the robot's posture error by an average of 84.95%, significantly improving the deformation problem.

[0053] In another implementation of the present invention, it further includes: calculating according to the actual torque and the expected torque of the robot in the reference coordinate system to obtain a deviation formula between the actual position of the robot's center of mass and the expected position based on the torque data:

[0054] r τ B - d τ B =F ext ·z c

[0055] in, r τ B In the reference coordinate system ∑ B The actual torque at the origin, In the reference coordinate system ∑ B Desired torque at the origin, z c Represents the coordinate system Σ B The height of the centroid point.

[0056] For example, Figure 3 As shown in Figure 2, based on the linear inverted pendulum model, the relationship between the robot's zero moment point (ZMP) and center of mass (COM) can be obtained:

[0057]

[0058] in, and T W Represents ZMP and CoM in Σ W The position in z w Represents the center point in the coordinate system Σ W The height of the medium.

[0059] Taking deformation compensation into account for interference, and set To control the variables, the system model can be described by the following state equation:

[0060]

[0061] y=Cx

[0062] Substitute the control variables into the equation and pass the sampling step t s By discretizing it, the state space equation can be rewritten as:

[0063] x(i+1)=Gx(i)+H(u(i)+u ff (i))

[0064] y(i)=Cx(i)

[0065] in, y represents the ZMP of the robot.

[0066] like Figure 3 As shown, the robot is in the reference coordinate system Σ B During the walking process, the actual position of the robot (expressed as r T) may deviate from the expected position d T. This deviation can be introduced by introducing a fictitious force F ext To express, therefore:

[0067] r τ B - d τ B =F ext ·z c

[0068] in, r τ B In the reference coordinate system Σ B The actual torque at the origin, In the reference coordinate system Σ B Desired torque at the origin, z c Represents the coordinate system Σ B The height of the center point.

[0069] In another implementation of the present invention, the center of mass active compliance controller calculates based on the plantar force and torque data and the expected trajectory to obtain center of mass compensation, including: the center of mass active compliance controller calculates based on the plantar force and torque data and the expected trajectory to obtain a deviation formula between the actual position of the robot's center of mass and the expected position based on the position data:

[0070]

[0071] in, r T represents the robot in the reference coordinate system Σ B The actual location in d T represents the robot in the reference coordinate system Σ B The desired position in the; combined with the deviation formula based on the torque data and the deviation formula based on the position data to calculate, obtain the center of mass compensation.

[0072] Exemplarily, CACC can be used to describe the relationship between the actual position and the desired position of the robot center point:

[0073]

[0074] Rearranging the equation, we can get:

[0075]

[0076] Continue to organize:

[0077]

[0078] Rearrange the equation:

[0079]

[0080] According to the above equation, if we know r τ B and The position deviation of the center of mass in the world coordinate system for each cycle can be obtained. The two variables can be calculated according to the following equations:

[0081]

[0082] therefore:

[0083]

[0084] The control law can be updated as:

[0085] T W (i+1)=T W (i)+ΔT W (i)

[0086] The center of mass active compliance controller is used to solve the problem of strong parameter disturbance caused by different wearers in the exoskeleton robot hybrid system.

[0087] In another implementation of the present invention, the new desired posture is expressed as:

[0088]

[0089] in, d T W is the robot's expected posture calculated offline in advance, Δ p T represents feedforward compensation, ΔT W Denotes center of mass compensation.

[0090] In another implementation of the present invention, since deformation hinders the collection of normal gait data, the deformation data collection of the self-balancing exoskeleton robot adopts a standing experiment. Figure 6 As shown, the experiment involves static actions such as squatting and standing, waist rotation, swinging the left leg and swinging the right foot. The squatting and standing and waist rotation actions simulate the double-leg support period, and the left leg swinging and right leg swinging actions simulate the single-leg support period. Therefore, these four actions can simulate the entire walking cycle, thereby establishing a mapping data set.

[0091] like Figure 7As shown, a total of 7 groups of self-balancing exoskeleton robot walking experiments were carried out. The results showed that the method achieved stable walking of the self-balancing exoskeleton robot in the load range of 0kg to 73kg, indicating that the method improved the stability of the exoskeleton robot during walking.

[0092] The present invention analyzes the manifestations and influencing factors of exoskeleton deformation, establishes an end-to-end mapping relationship between exoskeleton center of mass deformation and foot force / torque, and directly predicts and compensates for exoskeleton center of mass deformation through exoskeleton foot force / torque data, thereby ensuring that the lower limb self-balancing exoskeleton walks according to the expected trajectory.

[0093] Thus far, specific embodiments of the present invention have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve the desired results. Additionally, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing may be advantageous.

[0094] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0095] In the description of the present invention, the terms "first" and "second" are only used to facilitate the description of different components or names, and cannot be understood as indicating or implying a sequential relationship, relative importance, or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0096] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0097] It should be noted that although the specific embodiments of the present invention are described in detail in conjunction with the accompanying drawings, it should not be understood as limiting the scope of protection of the present invention. Within the scope described in the claims, various modifications and variations that can be made by those skilled in the art without creative work still belong to the scope of protection of the present invention.

[0098] The examples of the embodiments of the present invention are intended to concisely illustrate the technical features of the embodiments of the present invention so that those skilled in the art can intuitively understand the technical features of the embodiments of the present invention, and are not intended to be improper limitations of the embodiments of the present invention.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A self-balancing walking control method for a rehabilitation exoskeleton under dynamic deformation, characterized in that: include: Collect plantar force and torque data when the exoskeleton robot performs walking movements; Inputting the plantar force and moment data into a feedforward deformation compensator to obtain feedforward compensation; The center of mass active compliance controller calculates the center of mass compensation based on the plantar force and torque data and the desired trajectory; determining a new desired posture of the robot based on the feedforward compensation and the center of mass compensation; The inverse kinematics solver calculates the command joint angle of each joint of the robot according to the new desired posture; The exoskeleton robot performs walking and standing tasks based on the command joint angles.

2. The method according to claim 1, characterized in that: The feedforward compensation is expressed as: Δ p T W = d T W - r T W in, d T W Indicates the robot in the reference coordinate system Σ calculated offline in advance W Medium expected posture, T W Represents the robot in the reference coordinate system Σ W The true attitude in.

3. The method according to claim 2, characterized in that The feedforward deformation compensator is composed of an encoder and a decoder, which is used to establish a mapping relationship between the plantar force and torque data and the feedforward compensation. The mapping relationship is expressed as: Δ p T=f Δ (F) Among them, Δ p T represents feedforward compensation, and F represents plantar force and torque data.

4. The method according to claim 1, characterized in that: Also includes: According to the actual torque and expected torque of the robot in the reference coordinate system, the deviation formula between the actual position and the expected position of the robot's center of mass is obtained based on the torque data: r t B - d t B =F ext ·z c in, r τ B In the reference coordinate system Σ B The actual torque at the origin, d τ B In the reference coordinate system Σ B Desired torque at the origin, z c Represents the coordinate system Σ B The height of the centroid point.

5. The method according to claim 4, characterized in that The center of mass active compliance controller calculates based on the plantar force and torque data and the desired trajectory to obtain center of mass compensation, including: The center of mass active compliance controller calculates based on the plantar force and torque data and the desired trajectory to obtain the deviation formula between the actual position of the robot's center of mass and the desired position based on the position data: in, r T represents the robot in the reference coordinate system Σ B The actual location in d T represents the robot in the reference coordinate system Σ B The desired position in The center of mass compensation is obtained by combining the deviation formula obtained based on the torque data and the deviation formula obtained based on the position data for calculation.

6. The method according to claim 1, characterized in that The new desired posture is expressed as: in, d T W is the robot's expected posture calculated offline in advance, Δ p T represents feedforward compensation, ΔT W Denotes center of mass compensation.

Images