Flexible knee joint exoskeleton for maintaining dynamic stability of knee joint

Abstract

The invention discloses a flexible knee joint exoskeleton for maintaining dynamic stability of knee joints in the technical field of lower limb rehabilitation, and the flexible knee joint exoskeleton comprises a control box, a control module and a power supply unit are integrated in the control box, and one side of the control box is fixed to the waist of a user through a bandage. The device further comprises a sensing unit, a tibia anterior movement limiting assembly and a support fixing assembly, the output end of the sensing unit is electrically connected with the input end of the control box, and the sensing unit is used for obtaining first movement information of a user and sending the first movement information to the control box; the control box identifies the walking trend and the knee joint state of the user based on the first motion information and an artificial intelligence algorithm, wherein the knee joint state is a six-degree-of-freedom kinematics state. According to the flexible knee joint exoskeleton, the Bowden cable and the tibia forward movement limiting assembly are cooperatively driven, human body lower limb kinematics information is collected in real time through the sensing unit, and the forward movement amplitude of the tibia of a user can be limited.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lower limb rehabilitation, in particular to a flexible knee exoskeleton for maintaining dynamic stability of knee joint. BACKGROUND

[0002] In the rehabilitation process of the population after anterior cruciate ligament reconstruction, the ACLR cuts off the sensory afferent when reconstructing the physical ligament, destroys the neuromuscular control loop, and makes the central system unable to make correct decisions due to lack of joint information, resulting in muscle coordination disorder and knee joint dynamic instability, which affects the internal stress distribution. In clinical practice, patients commonly face secondary mechanical problems such as osteoarthritis, graft re-rupture, and bone tunnel wear and collapse. By constructing an external motion control loop with a flexible exoskeleton, simulating sensory input with a multi-modal sensing module, simulating central decision-making with an intelligent algorithm, and providing dynamic correction with a flexible actuator, the patient's neural control mechanism is reconstructed, and the incidence of postoperative adverse consequences is reduced.

[0003] The existing knee exoskeleton device has the following technical bottlenecks:

[0004] 1. The rigid exoskeleton is heavy and has poor flexibility, making it difficult to adapt to the complex movement of the knee joint in daily activities; 2. The traditional flexible exoskeleton mainly uses a single driving mode to compensate for the lack of muscle strength, lacks targeted restriction of tibial advancement, and cannot effectively maintain joint stability; 3. The control system has insufficient recognition accuracy for human motion intention, making it difficult to achieve real-time dynamic assistance. SUMMARY

[0005] The purpose of the present application is to provide a flexible knee exoskeleton for maintaining dynamic stability of knee joint, which can meet the rehabilitation assistance needs of people with knee joint injury by setting up a flexible knee exoskeleton that is lightweight, highly adaptable, and can accurately control tibial advancement.

[0006] To achieve the above purpose, the present application provides the following technical scheme:

[0007] A flexible knee exoskeleton for maintaining dynamic stability of knee joint, comprising a control box, wherein a control module and a power supply unit are integrated inside the control box, and the control box is fixed on the waist of a user through a belt on one side, characterized in that it further comprises a sensing unit, a tibial advancement limiting assembly, and a bracket fixing assembly, the output end of the sensing unit is electrically connected to the input end of the control box, the sensing unit is used to obtain first motion information of the user and send the first motion information to the control box, and the control box identifies the walking trend and the knee joint state of the user based on the first motion information, wherein the first motion information is the kinematics information of the lower limbs of the user.

[0008] The tibial advancement limiting component includes a flexible fixing ring, the inner side of the flexible fixing ring is uniformly provided with a plurality of pressure sensors, the pressure sensors are used for collecting actual limiting force data borne by the tibia of the user, and the actual limiting force data is sent to the control box, the output end of the pressure sensor is electrically connected with the signal conditioning module of the control box through shielding wire, one end of the flexible fixing ring is fixedly connected with the Bowden cable component, the other end of the Bowden cable component is fixedly connected with the outer wall of the control box, and the output end of the control box sends a control signal to the input end of the tibial advancement limiting component, so as to control the tibial advancement limiting component to provide an auxiliary torque for limiting the excessive advancement of the tibia.

[0009] The support fixing component includes a first fixing support and a second fixing support, the first fixing support is used for pressing and fixing the sensing unit on the thigh of the user, the second fixing support is detachably connected with one side of the flexible fixing ring, and the second fixing support is used for fixing the tibial advancement limiting component at the tibial platform of the user, so that the sensing unit closely adheres to the skin in dynamic motion, signal interference is avoided, the second fixing support is detachably connected with one side of the flexible fixing ring through a magnetic quick-release interface, quick wearing and adjustment are supported, and the second fixing support is used for firmly fixing the tibial advancement limiting component at the tibial platform of the user, so that uniform support force distribution is provided, local pressure points are reduced, the support fixing component as a whole adopts a lightweight carbon fiber composite material, an ergonomic curve design is combined, user comfort is enhanced, and dynamic stability of the knee joint is maintained.

[0010] As a further scheme of the application, the sensing unit includes five inertial measurement units (IMUs), the inertial measurement units are inertial sensors, and the five inertial measurement units are respectively used for collecting the knee joint angle and the lower limb motion state of the user, the pressure sensor is used for collecting actual resistance force signals applied by the exoskeleton in real time, and the actual resistance force signals are sent to the control box, the control box calculates difference data between the actual resistance force signals and theoretical resistance force through the control module, specifically, the five inertial measurement units include a first inertial measurement unit, a second inertial measurement unit, a third inertial measurement unit, a fourth inertial measurement unit and a fifth inertial measurement unit, wherein the first inertial measurement unit is used for collecting motion state data of the middle segment of the left thigh of the user, the second inertial measurement unit is used for collecting motion state data of the middle segment of the right thigh of the user, the third inertial measurement unit is used for collecting motion state data of the left lower leg of the user, the fourth inertial measurement unit is used for collecting motion state data of the right lower leg of the user, and the fifth inertial measurement unit is used for collecting motion state data of the center of the waist and back of the user, each inertial measurement unit sends the collected motion data to the control box, the control box estimates the knee joint state of the user based on an artificial intelligence algorithm, the artificial intelligence algorithm is an LSTM algorithm, and the knee joint state is the 6-DOF kinematics performance of the tibia relative to the femur.

[0011] As a further scheme of the present application: the flexible fixing ring is integrally injection molded by silicone rubber, the flexible fixing ring is a hollow annular structure, the inner wall of the annular structure is integrally formed with corrugated protrusions, and the inner diameter of the flexible fixing ring is 60-70 mm. By setting the hollow annular flexible fixing ring, on the one hand, the convenience of fixing the flexible fixing ring on the tibial plateau contact area of the user can be improved, and on the other hand, it can be adapted to users with different leg circumferences, and has wide adaptability.

[0012] As a further scheme of the present application: the inner wall of the flexible fixing ring is provided with a plurality of grooves, the inner part of the grooves is respectively adapted to fix the pressure sensor, the pressure sensor is a flexible thin film pressure sensor with a thickness of 0.5 mm, the pressure sensor is used to collect pressure distribution data of the tibial plateau area of the user in real time, and the pressure distribution data is transmitted to the control box, so as to realize monitoring of the load state of the knee joint. By using the flexible thin film pressure sensor, the sensor can be closely attached to the skin surface, avoiding displacement or discomfort during movement, ensuring the stability and reliability of the measurement. The groove and the sensor thickness are matched, so that the sensor can be stably embedded in the groove, while maintaining the overall flexibility of the flexible fixing ring to maintain the dynamic stability of the knee joint.

[0013] As a further scheme of the present application: a sandwich layer is arranged inside the flexible fixing ring, and a spiral memory alloy wire is fixed inside the sandwich layer. The diameter of the spiral memory alloy wire is not more than 1.2 mm.

[0014] As a further scheme of the present application: the Bowden wire assembly includes a metal terminal, a Bowden wire and a steel wire rope clamp head. The metal terminal is fixed at one end of the flexible fixing ring, and the steel wire rope clamp head is movably connected to the inner wall of the flexible fixing ring. One end of the metal terminal is connected to one end of the spiral memory alloy wire, and the other end of the spiral memory alloy wire is connected to one end of the steel wire rope clamp head. The other end of the steel wire rope clamp head is fixedly connected to one end of the Bowden wire. An electric motor is fixed inside the control box, and the other end of the Bowden wire is fixedly connected to the output end of the electric motor. When the electric motor winds the Bowden wire to be tight, the steel wire rope clamp head will pull the spiral memory alloy wire to deform and elongate, which is used to provide passive elastic resistance to limit the forward movement of the tibia of the user. By setting the spiral memory alloy wire and cooperating with the traction of the Bowden wire, the tibial forward movement limiting assembly can be conveniently adjusted for installation and use on the knee joints of users with different body types, and can adapt to the knee joint sizes of different users.

[0015] As a further scheme of the present application: the first fixing support is made of hard carbon fiber material, and nylon bases are symmetrically sewn at both ends of the first fixing support. One end of the nylon base is riveted with a first Velcro strap, which is used to fix the first fixing support on the thigh of the user.

[0016] As a further scheme of the present application: the inner side of the first fixed support and the second fixed support is curved, a grub screw is threadedly connected at the connecting position between the second fixed support and the upper end of the flexible fixing ring, the second fixed support is made of elastic memory alloy, two ends of the second fixed support are provided with reserved holes, the second magic tape bandage is sutured in the reserved holes, a wire hole is formed in the upper end of the second fixed support, the wire hole is used for penetrating the connecting wire of the pressure sensor and the sensing unit, the flexible fixing ring is used for providing comfortable fixing around the user's leg, the pressure sensor is used for monitoring the human-machine interaction force, i.e. the auxiliary torque applied by the exoskeleton, the helical memory alloy wire is contracted or expanded through the shape memory effect under the traction of the Bowden cable, so as to dynamically limit the tibia forward movement, the first fixed support and the second fixed support are fixed on the user's thigh and calf respectively through the grub screw, rigid support and stability of the overall exoskeleton are provided, the Bowden cable assembly is connected with the control box and the tibia forward movement limiting assembly, the metal terminal is used for electrical signal interface, the Bowden cable is used for transmitting control instructions and mechanical force, and the steel wire rope clamp head ensures firm connection, the control box obtains the first motion information based on the sensing unit, analyzes the user's walking trend and the knee joint state, and adjusts the response of the helical memory alloy wire through the Bowden cable assembly, so as to realize adaptive stability of the knee joint in dynamic activity.

[0017] As a further scheme of the present application: when the user walks and the tibia has an abnormal forward movement trend, the pressure sensor is used as a part of the closed loop feedback, is used for feeding back the torque that should be applied by the exoskeleton, and transmits the mechanical signal to the signal conditioning module in the control box through the shielding wire, the signal conditioning module filters, amplifies and digitizes the original signal, and then transmits the signal to the control module for analysis, the control module compares the actual limiting force data with the preset safety threshold range, if it is detected that the tibia forward movement force exceeds the biomechanical safety limit, a dynamic adjustment instruction is generated through the control algorithm immediately, the dynamic adjustment instruction drives the servo motor integrated in the control box, the tension of the Bowden cable is accurately wound and unwound, so that the tibia forward movement limiter generates a constraint torque in the opposite direction of the abnormal forward movement, and the closed loop control system can continuously adapt to the mechanical demand in different gait phases, provides high strength constraint at the end of the gait support phase to prevent anterior cruciate ligament injury, and automatically reduces the constraint degree to ensure the freedom degree of joint activity in the swing phase, so as to realize dynamic stability protection in line with physiological characteristics.

[0018] As a further scheme of the present application: the control module comprises a high-level controller, a middle-level controller and a bottom-level controller, the high-level controller is electrically connected with the middle-level controller, the middle-level controller is electrically connected with the bottom-level controller, the high-level controller identifies the motion trend of the user based on an artificial intelligence algorithm, the middle-level controller is used for identifying the current tibia posture based on the lower limb kinematics information of the human body and the motion trend, and the bottom-level controller is used for outputting the driving instruction of the motor based on the motion trend and the current tibia posture, so as to provide dynamic stability performance maintenance for the knee joint, wherein the artificial intelligence algorithm is an LSTM neural network algorithm.

[0019] As a further scheme of the present application: the lower limb kinematics information of the user comprises the knee joint angle, the middle segment of the thigh, the upper segment of the lower leg and the forward direction acceleration of the center of the human back.

[0020] Compared with the prior art, the present application has the following advantages:

[0021] 1. The flexible knee exoskeleton of the present application cooperates with the tibia forward movement limiting assembly driven by the Bowden cable, and cooperates with the sensing unit and the control module integrated in the control box, so as to realize real-time acquisition of the lower limb kinematics information of the human body by the sensing unit, control of the motor and the sensing unit by the portable control box, limitation of the forward movement amplitude of the tibia of the user, reduction of the weight of the exoskeleton, maintenance of the tibia posture conducive to rehabilitation, adaptation to the complex motion of the knee joint in the daily activities of the user, and solution to the problems of poor flexibility and low wearing comfort of the existing rigid exoskeleton by setting the flexible fixing ring in contact with the leg.

[0022] 2. The present application can improve the response speed and control accuracy of the system through layered control, wherein the high-level controller identifies the walking and stair climbing modes based on a deep learning algorithm, the middle-level controller monitors the tibia posture in real time and triggers the dynamic limiting mechanism, and the bottom-level controller is used for adjusting the tension of the Bowden cable and the damping force of the limiter, so as to improve the control accuracy, ensure the stability of the knee joint, reduce the energy consumption, realize the convenience of all-day wearing, achieve the purpose of regulating the dynamic stability of the knee joint of the knee joint injury population, effectively reduce the load of the knee joint, speed up the rehabilitation process, and can be widely applied in the fields of sports rehabilitation, postoperative care, old-age assistance and the like, and has a wide market prospect.

[0023] 3. The inner sides of the first fixed support and the second fixed support are curved, so that the first fixed support and the second fixed support can better fit the leg curve of the user, so that the motion and torque generated by the tibia forward movement limiting assembly can be transmitted to the knee joint of the user, the Bowden cable can be driven by the motor, the fixing degree of the tibia forward movement limiting assembly and the leg can be adjusted, the deficiency of muscle strength can be compensated, and the joint stability can be effectively maintained. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a schematic diagram of the overall structure of the present application;

[0025] Figure 2 is a schematic diagram of the connection structure of the Bowden wire assembly of the present application;

[0026] Figure 3 is a schematic diagram of the connection structure of the second fixing support of the present application;

[0027] Figure 4 is a schematic diagram of the overall control of the present application.

[0028] In the figure: 1, control box; 2, sensing unit; 3, tibia forward movement limiting assembly; 301, flexible fixing ring; 302, pressure sensor; 303, helical memory alloy wire; 4, Bowden wire assembly; 401, metal terminal; 402, Bowden wire; 403, steel wire rope clamp; 5, support fixing assembly; 501, first fixing support; 502, second fixing support; 503, countersunk bolt; 6, bandage. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0030] Embodiment:

[0031] Please refer to Figures 1-4 The embodiment of the present application provides a flexible knee exoskeleton for maintaining dynamic stability of a knee joint, comprising a control box 1, a control module and a power supply unit are integrated in the control box 1, one side of the control box 1 is fixed on the waist of a user through a bandage 6, characterized in that further comprising a sensing unit 2, a tibia forward movement limiting assembly 3 and a support fixing assembly 5, the output end of the sensing unit 2 is electrically connected to the input end of the control box 1, the sensing unit 2 is used for acquiring first motion information of a user and sending the first motion information to the control box 1, the control box 1 identifies a walking trend and a knee joint state of the user based on the first motion information, wherein the first motion information is kinematics information of a lower limb of the user

[0032] When the tibia forward movement limiting component 3 is worn at the knee joint of the user, an auxiliary moment for limiting excessive forward movement of the tibia is provided by adjusting the tension of the Bowden cable 402 and the mechanical constraint of the tibia forward movement limiter, the tibia forward movement limiting component 3 comprises a flexible fixing ring 301, a plurality of pressure sensors 302 are uniformly mounted on the inner side of the flexible fixing ring 301, the pressure sensors 302 are used to collect actual limiting force data borne by the tibia of the user, and send the actual limiting force data to the control box 1, the output end of the pressure sensor 302 is electrically connected to the signal conditioning module of the control box 1 through a shielded wire, one end of the flexible fixing ring 301 is fixedly connected with the Bowden cable assembly 4, the other end of the Bowden cable assembly 4 is fixedly connected with the outer wall of the control box 1, and the output end of the control box 1 sends a control signal to the input end of the tibia forward movement limiting component 3, for controlling the tibia forward movement limiting component 3 to provide an auxiliary moment for limiting excessive forward movement of the tibia

[0033] The support fixing assembly 5 comprises a first fixing support 501 and a second fixing support 502, the first fixing support 501 is used to press and fix the perception unit 2 on the thigh of the user, the second fixing support 502 is detachably connected with one side of the flexible fixing ring 301, and the second fixing support 502 is used to fix the tibia forward movement limiting component 3 at the tibial platform of the user, so that the perception unit 2 closely adheres to the skin in dynamic movement and signal interference is avoided; the second fixing support 502 is detachably connected with one side of the flexible fixing ring 301 through a magnetic quick-release interface, quick wearing and adjustment are supported, and the second fixing support 502 is used to firmly fix the tibia forward movement limiting component 3 at the tibial platform of the user, so that uniform support force distribution is provided and local pressure points are reduced. The overall support is made of lightweight carbon fiber composite material, combined with ergonomic curve design, user comfort is enhanced, and dynamic stability of the knee joint is maintained.

[0034] In the embodiment, the motion trend of the human body includes walking on flat ground, going upstairs and going downstairs.

[0035] In this embodiment, the flexible fixing ring 301 is used to provide comfortable fixation around the user's legs, the pressure sensor 302 is used to monitor the human-computer interaction force in real time, that is, the auxiliary torque applied by the exoskeleton, the helical shape memory alloy wire 303 is contracted or expanded through the shape memory effect under the traction of the Bowden cable 402 to dynamically limit the forward movement of the tibia, the first fixing support 501 and the second fixing support 502 are fixed to the user's thighs and shanks respectively through the countersunk head bolt 503 to provide rigid support and stability of the overall exoskeleton, the Bowden cable assembly 4 is connected to the control box 1 and the tibia forward movement limiting assembly 3, the metal terminal 401 is used for electrical signal interface, the Bowden cable 402 is used for transmitting control instructions and mechanical force, and the steel wire rope clamp 403 ensures firm connection. The control box 1 analyzes the user's walking trend and the state of the knee joint based on the first motion information obtained by the sensing unit 2, and adjusts the response of the helical shape memory alloy wire 303 through the Bowden cable assembly 4, so as to realize the adaptive stability of the knee joint in dynamic activity.

[0036] In this embodiment, the control module includes a high-level controller, a middle-level controller and a bottom-level controller, the high-level controller is electrically connected with the middle-level controller, and the middle-level controller is electrically connected with the bottom-level controller. The high-level controller identifies the motion trend of the user based on an artificial intelligence algorithm, the middle-level controller is used to identify the current posture of the tibia based on the kinematics information and the motion trend of the lower limbs, and the bottom-level controller is used to output the driving instructions of the motor based on the motion trend and the current posture of the tibia, so as to maintain the dynamic stability performance of the knee joint. The artificial intelligence algorithm is an LSTM neural network algorithm.

[0037] Preferably, the sensing unit 2 includes five inertial measurement units, which are respectively used to collect the knee joint angle and the lower limb motion state of the user. The pressure sensor 302 is used to collect the actual resistance signal applied by the exoskeleton in real time and send the actual resistance signal to the control box 1. The control box 1 calculates the difference data between the actual resistance signal and the theoretical resistance signal through the control module. Specifically, the five inertial measurement units include a first inertial measurement unit, a second inertial measurement unit, a third inertial measurement unit, a fourth inertial measurement unit and a fifth inertial measurement unit. The first inertial measurement unit is used to collect the motion state data of the middle segment of the user's left thigh, the second inertial measurement unit is used to collect the motion state data of the middle segment of the user's right thigh, the third inertial measurement unit is used to collect the motion state data of the user's left shank, the fourth inertial measurement unit is used to collect the motion state data of the user's right shank, and the fifth inertial measurement unit is used to collect the motion state data of the center of the user's back. Each inertial measurement unit sends the collected motion data to the control box 1. In this embodiment, the power supply unit is used to supply power to the structure installed in the control box 1.

[0038] In this embodiment, when the user walks and the tibia appears to be abnormally forward, the pressure sensor 302 serves as a link in the closed-loop feedback, feeding back the torque that the exoskeleton should exert, and transmitting the mechanical signal to the signal conditioning module in the control box 1 through the shielding wire. After the signal conditioning module filters, amplifies and digitizes the original signal, it is transmitted to the control module for analysis. The control module compares the actual limiting force data with the preset safety threshold range. If it is detected that the tibia forward force exceeds the biomechanical safety limit, the dynamic adjustment instruction is generated immediately through the control algorithm, which drives the servo motor integrated in the control box 1, and by accurately adjusting the tension of the Bowden cable 402, the tibia forward limiting device generates a restraining torque in the opposite direction of the abnormal forward movement. This closed-loop control system can continuously adapt to the mechanical requirements of different gait phases, providing high-strength restraint at the end of the gait support phase to prevent anterior cruciate ligament injury, and automatically reducing the restraint force during the swing phase to ensure the freedom of joint movement, thereby achieving dynamic stability protection that meets the physiological characteristics.

[0039] Preferably, the flexible fixing ring 301 is integrally injection molded by silicone rubber, and the flexible fixing ring 301 has a hollow annular structure, and the inner wall of the annular structure is integrally formed with corrugated protrusions. The inner diameter of the flexible fixing ring 301 is 60-70 mm. By providing a hollow annular flexible fixing ring 301, on the one hand, the convenience of fixing the flexible fixing ring 301 on the tibial plateau contact area of the user can be improved, and on the other hand, it can be adapted for users with different leg circumferences, and has wide adaptability.

[0040] Preferably, the inner wall of the flexible fixing ring 301 is provided with a plurality of grooves, and the inner part of each groove corresponds to a fixed pressure sensor 302. The pressure sensor 302 is a flexible thin film pressure sensor with a thickness of 0.5 mm. The pressure sensor is used to collect pressure distribution data of the tibial plateau area of the user in real time and transmit the pressure distribution data to the control box, thereby monitoring the load state of the knee joint. By using a flexible thin film pressure sensor, the sensor can closely adhere to the skin surface, avoiding displacement or discomfort during exercise, ensuring the stability and reliability of the measurement. The groove is matched with the thickness of the sensor, so that the sensor can be stably embedded in the groove while maintaining the overall flexibility of the flexible fixing ring to maintain the dynamic stability of the knee joint.

[0041] Preferably, the inside of the flexible fixing ring 301 is provided with a sandwich layer, and the sandwich layer is fixed with a spiral memory alloy wire 303. The diameter of the spiral memory alloy wire 303 is 1.2 mm.

[0042] Preferably, the bowden wire assembly 4 comprises a metal terminal 401, a bowden wire 402 and a steel wire rope clamp 403, the metal terminal 401 is fixed at one end of the flexible fixing ring 301, the steel wire rope clamp 403 is movably connected to the inner wall of the flexible fixing ring 301, one end of the metal terminal 401 is connected to one end of the spiral memory alloy wire 303, the other end of the spiral memory alloy wire 303 is connected to one end of the steel wire rope clamp 403, the other end of the steel wire rope clamp 403 is fixedly connected to one end of the bowden wire 402, and the inside of the control box 1 is fixed with a motor.

[0043] In this embodiment, the motor is a micro servo motor, which comprises a reduction gear set, a drive circuit module and an encoder. The output shaft of the micro servo motor is connected to the reel of the bowden wire 402, which is used to wind or release the bowden wire 402 to adjust the tension. The reduction gear set is arranged between the servo motor and the reel, which is used to reduce the speed and increase the output torque. The control module is integrated with a PWM speed regulation module and a current detection circuit. The PWM speed regulation module and the current detection circuit adjust the motor output power according to the instruction signal of the bottom controller. The encoder is coaxially connected with the servo motor, which is used to monitor the motor angle position in real time and feedback to the controller.

[0044] Preferably, the output end of the motor is fixedly connected to the other end of the bowden wire 402. When the motor winds the bowden wire 402 to tension, the steel wire rope clamp 403 will pull the spiral memory alloy wire 303 to deform and elongate, which is used to provide passive elastic resistance to limit the forward movement of the tibia. By setting the spiral memory alloy wire 303 and cooperating with the traction of the bowden wire 402, it is convenient to adjust the installation and use of the tibia forward movement limiting assembly 3 on the knee joint of users with different body types, and adapt to the knee joint size of different users.

[0045] Preferably, the first fixed support 501 is made of hard carbon fiber material, and nylon bases are symmetrically sewn at both ends of the first fixed support 501. One end of the nylon base is riveted with a first magic tape bandage, which is used to fix the first fixed support 501 on the thigh of the user.

[0046] Preferably, the inner sides of the first fixing support 501 and the second fixing support 502 are curved, a countersunk bolt 503 is threadedly connected at the connection position between the second fixing support 502 and the upper end of the flexible fixing ring 301, the second fixing support 502 is made of elastic memory alloy, two ends of the second fixing support 502 are provided with reserved holes, the inside of each reserved hole is back-folded and sutured with a second magic tape bandage, and the upper end of the second fixing support 502 is provided with a wire hole for the connecting wires of the pressure sensor 302 and the sensing unit 2. By setting the inner sides of the first fixing support 501 and the second fixing support 502 to be curved, the first fixing support 501 and the second fixing support 502 can better fit the leg curve of the user, so as to transmit the movement and torque generated by the tibial advancement limiting assembly 3 to the knee joint of the user.

[0047] Preferably, the lower limb kinematics information of the user includes the knee joint angle, the middle segment of the thigh, the upper segment of the lower leg, and the forward direction acceleration of the center of the back of the human body.

[0048] The working principle of the control module is as follows: the high-level controller uses an LSTM neural network to extract features of the five IMU data in the sensing unit 2, and identifies multiple modes such as walking on flat ground, going upstairs, and going downstairs; the middle-level controller fuses the knee joint kinematics data based on an artificial intelligence algorithm, divides the gait cycle into a support phase and a swing phase, and calculates the tibial advancement difference of the current healthy side and the affected side in real time, fuses the current actual contact force collected by the thin film pressure sensor, and obtains the output torque at the next moment through a machine learning model; the bottom-level controller generates a PWM signal according to the torque demand, adjusts the speed of the servo motor, and simultaneously monitors the motor current through a current detection resistor, and triggers current limiting protection when overloaded.

[0049] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A flexible knee exoskeleton for maintaining dynamic stability of a knee joint, comprising a control box, an inside of the control box is integrated with a control module and a power supply unit, a side of the control box is fixed on a waist of a user through a bandage, characterized in that, The still includes: A perception unit, an output end of the perception unit being electrically connected to an input end of the control box, the perception unit being used to acquire first motion information of a user and send the first motion information to the control box, the control box identifying a walking trend and a knee joint state of the user based on the first motion information, wherein the first motion information is kinematics information of lower limbs of the user; A tibia forward movement limiting assembly, the tibia forward movement limiting assembly including a flexible fixing ring, a plurality of pressure sensors being uniformly mounted on an inner side of the flexible fixing ring, the pressure sensors being used to collect actual limiting force data borne by a tibia of the user and send the actual limiting force data to the control box, output ends of the pressure sensors being electrically connected to a signal conditioning module of the control box through shielding wires, one end of the flexible fixing ring being fixedly connected to a Bowden cable assembly, the other end of the Bowden cable assembly being fixedly connected to an outer wall of the control box, an output end of the control box sending a control signal to an input end of the tibia forward movement limiting assembly for controlling the tibia forward movement limiting assembly to provide an auxiliary torque for limiting excessive forward movement of the tibia; A support fixing assembly, the support fixing assembly including a first fixing support and a second fixing support, the first fixing support being used to press and fix the perception unit on a thigh of the user, the second fixing support being detachably connected to one side of the flexible fixing ring, the second fixing support being used to fix the tibia forward movement limiting assembly at a tibial plateau of the user.

2. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 1, characterized in that: The perception unit includes five inertial measurement units, the five inertial measurement units each being an inertial sensor, the five inertial measurement units including a first inertial measurement unit, a second inertial measurement unit, a third inertial measurement unit, a fourth inertial measurement unit and a fifth inertial measurement unit, wherein the first inertial measurement unit is used to collect motion state data of a middle segment of a left thigh of the user, the second inertial measurement unit is used to collect motion state data of a middle segment of a right thigh of the user, the third inertial measurement unit is used to collect motion state data of a left lower leg of the user, the fourth inertial measurement unit is used to collect motion state data of a right lower leg of the user, and the fifth inertial measurement unit is used to collect motion state data of a center of a back of the user, each of the inertial measurement units sends collected motion data to the control box, the control box estimates a knee joint state of the user based on an artificial intelligence algorithm, the pressure sensors are used to collect actual resistance force signals applied by the exoskeleton in real time and send the actual resistance force signals to the control box, the control box calculates difference data between the actual resistance force signals and theoretical resistance force through a control module, wherein the artificial intelligence algorithm is an LSTM algorithm, and the knee joint state is a 6-DOF kinematics performance of the tibia relative to the femur.

3. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 1, characterized in that: The flexible fixing ring is integrally injection molded by silicon rubber, the flexible fixing ring is a hollow annular structure, the inner wall of the annular structure is integrally formed with corrugated protrusions, and the inner diameter of the flexible fixing ring is 60-70 mm.

4. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 3, characterized in that: The inner wall of the flexible fixing ring is provided with a plurality of grooves, the inner part of each of the grooves is respectively adapted to fix the pressure sensor, and the pressure sensor is a flexible thin film pressure sensor with a thickness of 0.5 mm.

5. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 4, characterized in that: The flexible fixing ring is internally provided with a sandwich layer, and a spiral memory alloy wire is fixed in the sandwich layer.

6. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 1, wherein: The Bowden wire assembly comprises a metal terminal, a Bowden wire and a steel wire rope clamp head, the metal terminal is fixed at one end of the flexible fixing ring, the steel wire rope clamp head is movably connected to the inner wall of the flexible fixing ring, one end of the metal terminal is connected to one end of the spiral memory alloy wire, the other end of the spiral memory alloy wire is connected to one end of the steel wire rope clamp head, the other end of the steel wire rope clamp head is fixedly connected to one end of the Bowden wire, the control box is internally provided with a motor, the output end of the motor is fixedly connected to the other end of the Bowden wire, when the motor winds the Bowden wire to be tight, the steel wire rope clamp head will pull the spiral memory alloy wire to be deformed and elongated, so as to provide passive elastic resistance for limiting the forward movement of the tibia of the user.

7. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 6, characterized in that: The first fixing support is made of hard carbon fiber material, nylon bases are symmetrically sewn at both ends of the first fixing support, a first magic tape bandage is riveted at one end of the nylon base, and the first magic tape bandage is used to fix the first fixing support on the thigh of the user.

8. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 7, characterized in that: The inner sides of the first fixing support and the second fixing support are curved, a countersunk bolt is threadedly connected to the connection position between the second fixing support and the upper end of the flexible fixing ring, the second fixing support is made of elastic memory alloy, the second fixing support is provided with a reserved hole at both ends, a second magic tape bandage is back-folded and sewn in the reserved hole, and a wire hole is formed in the upper end of the second fixing support, the wire hole is used for passing the connecting wires of the pressure sensor and the sensing unit.

9. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 8, characterized in that: The control module comprises a high-level controller, a middle-level controller and a bottom-level controller, the high-level controller is electrically connected with the middle-level controller, and the middle-level controller is electrically connected with the bottom-level controller, the high-level controller identifies the movement trend of the user based on an artificial intelligence algorithm, the middle-level controller identifies the current posture of the tibia based on the kinematics information of the lower limbs of the human body and the movement trend, and the bottom-level controller outputs the driving instruction of the motor based on the movement trend and the current posture of the tibia.

10. The flexible knee exoskeleton that maintains dynamic stability of the knee joint according to claim 9, characterized in that: The kinematics information of the lower limbs of the user comprises the knee joint angle, the middle segment of the thigh, the upper segment of the lower leg and the forward direction acceleration of the center of the back of the human body.

Images