Knee joint exoskeleton motion recognition method and knee joint assisting device

Abstract

The present application relates to the field of exoskeleton assisting device, in particular to a knee joint exoskeleton motion recognition method and a knee joint assisting device, comprising a lower limb exoskeleton device, the lower limb exoskeleton device comprises: a joint driving motor, the joint driving motor comprises a pair of rotating components, the joint driving motor is used for applying torque to the pair of relatively rotating rotating components, and the center of relative rotation of the rotating components corresponds to the knee joint;A thigh extension frame and a calf extension frame connected with the pair of rotating components respectively;A thigh binding lock device mounted on the thigh extension frame, the thigh binding lock device is bound to the thigh;A first calf binding lock device and a second calf binding lock device mounted on the calf extension frame, the first calf binding lock device and the second calf binding lock device are bound between the calf belly and the knee joint.The application has the advantages of good use adaptability and the effect of blocking and falling.In the process of downhill movement, the knee can be protected by simulating the control of damping.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of exoskeleton assisting device, in particular to a knee joint exoskeleton motion recognition method and a knee joint assisting device. BACKGROUND

[0002] The knee joint exoskeleton device binds the thigh and the lower leg of the user and drives the lower limb to perform assisting motion by using the motor located at the knee.

[0003] The existing lower limb knee joint exoskeleton is difficult to accurately recognize the motion state of the exoskeleton only by relying on the sensor located on the lower limb. This requires an additional position detection sensor to be provided outside the lower limb exoskeleton, which causes wearing trouble, or the lower limb exoskeleton is extended above the waist and hips, and the position detection sensor is installed at the position corresponding to the waist and hips of the lower limb exoskeleton, which causes the lower limb knee joint exoskeleton to be too bulky. SUMMARY

[0004] In order to overcome the shortcomings of the prior art, improve the recognition accuracy of the motion state, and ensure the light weight of the device.

[0005] In order to achieve the above-mentioned purpose, the present application is realized by the following technical scheme:

[0006] A knee joint exoskeleton motion recognition method based on a lower limb exoskeleton device, the lower limb exoskeleton device includes a thigh extension frame and a lower leg extension frame respectively bound to the thigh and the lower leg, the thigh extension frame and the lower leg extension frame are relatively rotatably arranged around the knee joint, and a motion parameter sensor is arranged on the lower limb exoskeleton device; the method comprises the following processes:

[0007] The motion parameter sensor is used to collect the motion parameters of the lower limb exoskeleton device to determine the motion characteristics, and then the motion state is confirmed according to the motion characteristics.

[0008] The motion characteristics include the knee joint angle, the knee joint angular velocity, the inclination angle of the lower leg and / or the thigh, the angular velocity of the lower leg and / or the thigh, the acceleration value of the thigh and / or the lower leg, and the hip projection difference, the hip projection difference being the change amount of the hip position in the current state relative to the lower limb in the preset initial posture.

[0009] A knee joint assisting device includes a lower limb exoskeleton device, the lower limb exoskeleton device includes:

[0010] A joint driving motor, the joint driving motor includes a pair of rotating components, the joint driving motor is used to apply torque to the pair of rotating components which are relatively rotatable, and the center of relative rotation of the rotating components corresponds to the knee joint;

[0011] A thigh extension frame and a calf extension frame are respectively connected to a pair of rotating components, and the thigh extension frame and the calf extension frame are respectively tied to the thigh and the calf.

[0012] As can be seen from the above technical solution, the present invention has the following beneficial effects:

[0013] By directly collecting knee joint angle, angular velocity, thigh / lower leg tilt angle and acceleration values ​​through motion parameter sensors (such as IMU and encoder) on the lower limb exoskeleton, and combining them with hip projection difference (the change in lower limb spatial displacement based on geometric calculation), a comprehensive identification of motion status can be achieved.

[0014] The hip projection difference is calculated by using the trigonometric relationship between lower limb length and joint angle, replacing the function of traditional waist and hip position sensors, avoiding the need for the device to extend upwards to the waist and hips, and significantly simplifying the structure.

[0015] The sensors are all integrated into the thigh and calf extension frames of the lower limb exoskeleton, eliminating the need for additional waist and hip straps, thus reducing the wearing steps and the weight of the device. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the use of a knee joint assist device according to one embodiment of this application;

[0017] Figure 2 This is a structural view of the front of a knee joint assist device according to an embodiment of this application (with the binding component hidden).

[0018] Figure 3 This is a structural diagram of the back of a knee joint assist device according to an embodiment of this application (with hidden fasteners);

[0019] Figure 4 This is a schematic diagram illustrating the principle of the hip projection difference in one embodiment of this application.

[0020] In the diagram: 1-Joint drive motor; 2-Thigh extension frame; 21-Thigh binding device; 211-Upper connecting seat; 212-First binding component; 3-Lower leg extension frame; 31-First lower leg binding device; 311-Lower connecting seat; 312-Second binding component; 32-Second lower leg binding device; 321-Knee connecting seat; 322-Third binding component. Detailed Implementation

[0021] Example 1

[0022] Combination Figures 1 to 3 As shown, this embodiment provides a knee joint assist device, including a lower limb exoskeleton device, which includes:

[0023] The joint driving motor 1 comprises a pair of rotating assemblies, and is used for applying torque to the pair of rotating assemblies which are relatively rotated at the center corresponding to the knee joint; a thigh extension frame 2 and a shank extension frame 3 which are respectively connected with the pair of rotating assemblies; a thigh binding device 21 which is installed on the thigh extension frame 2 and is bound on the thigh; and a shank binding device 22 which is installed on the shank extension frame 3 and is bound on the shank. In the embodiment, the pair of rotating assemblies are the stator and the rotor of the joint driving motor 1. The joint driving motor 1 further comprises a motion parameter sensor which is installed on the lower extremity exoskeleton device and is used for collecting the motion parameters of the lower extremity exoskeleton device to determine the motion characteristics.

[0024] The existing knee exoskeleton is prone to falling when being bound on the lower extremity, and has poor use adaptability because the bound position needs to be adjusted according to different users with different leg lengths. To this end, the knee assisting device further comprises a first shank binding device 31 and a second shank binding device 32 which are installed on the shank extension frame 3 and are bound between the shank belly and the knee joint.

[0025] Based on the above structure, the principle of the knee assisting device is that the joint driving motor 1 is used for applying torque to the rotating assemblies to transmit the force to the thigh and the shank through the thigh extension frame 2 and the shank extension frame 3 to assist the knee joint movement. In the application, the first shank binding device 31 and the second shank binding device 32 are arranged above the shank belly and below the knee joint, and the thigh binding device 21, the first shank binding device 31 and the second shank binding device 32 form three anchor points of the lower extremity exoskeleton device connected with the lower extremity. In the embodiment, on the one hand, the protruding characteristics of the shank belly have the effect of preventing falling, and on the other hand, the distance between the shank belly and the knee joint is short in the lower extremity part, so that the two anchor points of the shank are arranged at this position to adapt to users with different leg lengths, and the knee assisting device has the advantage of good use adaptability. In an embodiment, in order to further prevent the lower extremity exoskeleton device from falling, the lower extremity exoskeleton device can be bound on the waist or above the waist.

[0026] Further, the thigh binding device 21 comprises an upper connecting seat 211 and a first binding member 212. The upper connecting seat 211 is rotatably installed on the thigh extension frame 2, and the rotation axis of the upper connecting seat 211 is transverse. The upper connecting seat 211 is attached to the back side of the thigh, and the first binding member 212 binds the thigh on the upper connecting seat 211.

[0027] Further, in the embodiment, the first lower leg binding device 31 comprises a lower connecting seat 311 and a second binding member 312, the lower connecting seat 311 is rotatably installed on the lower leg extension frame 3, the rotation axis of the lower connecting seat 311 is transverse, the lower connecting seat 311 is attached to the back side of the lower leg, and the second binding member 312 binds the lower leg to the lower connecting seat 311. The rotatably arranged lower connecting seat 311 and upper connecting seat 211 have better fit with the lower limb.

[0028] Further, in the embodiment, the second lower leg binding device 32 comprises a knee connecting seat 321 and a third binding member 322, the knee connecting seat 321 is installed on the lower leg extension frame 3, the knee connecting seat 321 is attached to the side of the lower leg, the third binding member 322 binds the lower leg to the knee connecting seat 321, and the knee connecting seat 321 is above the lower connecting seat 311. Specifically, the binding member is a strap, and the connecting seat is connected by a magnetic buckle.

[0029] Further, in the embodiment, the joint driving motor 1 is arranged beside the knee joint, the thigh extension frame 2 and the lower leg extension frame 3 are integrally curved plate members, the ends of the thigh extension frame 2 and the lower leg extension frame 3 away from the joint driving motor 1 extend to the back side of the leg, and the thigh binding device 21 and the first lower leg binding device 31 are respectively installed at the ends of the thigh extension frame 2 and the lower leg extension frame 3 away from the joint driving motor 1. The structure is reasonable and the quality is light.

[0030] Further, in the embodiment, a power module is further included, the power module is electrically connected with the joint driving motor 1, and the power module is arranged on the waist or above the waist.

[0031] Further, in the embodiment, a motion parameter sensor is further included, the number of the motion parameter sensors is 2, the motion parameter sensors are used to collect motion parameters of the lower limb exoskeleton device to determine at least one motion feature of the knee joint angle, the inclination angle of the lower leg and the thigh, the angular velocity, and the hip projection difference. The use of the motion parameter sensors is reduced to the greatest extent, and the cost is low.

[0032] In an embodiment, the motion parameter sensors are a pair of IMUs respectively installed on the thigh extension frame 2 and the lower leg extension frame 3. The knee joint angle can be calculated from the motion data such as inclination angle and angular velocity obtained by the pair of IMUs.

[0033] In another embodiment, the motion parameter sensors are an IMU mounted on the thigh extension 2 or the shank extension 3, and a knee angle sensor. The motion state (e.g. angular velocity and acceleration) of the extension without the IMU can be calculated from the motion data obtained by the IMU and the change of the knee angle obtained by the knee angle sensor. In an embodiment, the knee angle sensor is an encoder mounted on the joint drive motor 1. The encoder is used to monitor the change of the angle of the stator and the rotor of the joint drive motor 1, so as to obtain the change signals of the knee angle and the angular velocity.

[0034] Embodiment 2

[0035] Based on the knee joint assisting device provided in Embodiment 1, the embodiment provides a control method,

[0036] comprising the following steps:

[0037] Motion parameter sensor data processing, obtaining the detection data input by the motion parameter sensors arranged on the lower extremity exoskeleton, and processing the detection data to extract motion features;

[0038] According to the motion features, the motion state is confirmed.

[0039] Specifically, the motion state includes gait phase and activity type.

[0040] Further, in an embodiment, the control method further comprises:

[0041] Gait phase detection, determining the current gait phase according to the motion features, the gait phase including a support phase and a swing phase, wherein the support phase corresponds to the state of the lower extremity touching the ground, and the swing phase corresponds to the state of the lower extremity leaving the ground;

[0042] Activity recognition, extracting the motion features in the support phase to determine the activity type, the activity type including at least one of walking, ascending, descending, running, squatting, reversing, and unknown;

[0043] Control strategy, selecting a corresponding control strategy according to the gait phase and the activity type to control the joint drive motor 1.

[0044] The standing support phase (hereinafter referred to as "support phase") is defined as the gait phase in which the lower extremity is in contact with the ground, and generally marks the beginning of a complete gait cycle. This stage occurs universally, regardless of the activity type (e.g. walking, ascending, squatting or descending) of the wearer, and the gait always starts from the contact of the foot with the ground. Therefore, the support phase provides a natural time window that can be used as the starting point of the gait cycle for capturing and analyzing the initial characteristics of the gait.

[0045] And, in the stance phase, motion parameter sensor data (e.g., acceleration from thigh inertial measurement unit, IMU, and knee angular velocity) exhibit relatively stable trends. For example, acceleration approaches the gravitational acceleration (about 9.8 m / s²), and angular velocity amplitude is small (typically below a certain threshold, e.g., 0.5 rad / s). This stability provides a reliable reference signal for feature extraction, in contrast to the fluctuations in subsequent dynamic phases (e.g., swing phase).

[0046] In one aspect, the extraction of motion features relies on key characteristics of the motion pattern, and the stance phase, with its specific biomechanical and dynamic properties, becomes the best point in time to identify the activity type. This is manifested in:

[0047] I. In the stance phase, the wearer's body posture (e.g., knee angle, calf tilt angle, hip position, etc.) directly reflects the type of activity. For example, the knee angle significantly decreases (e.g., less than 120 degrees) when squatting, and the calf tilt angle is larger (e.g., more than 30 degrees) when ascending. These posture information is most evident and stable when the foot is in contact with the ground, facilitating the capture through motion parameter sensor data. Therefore, in contrast, the swing phase has the foot off the ground, and the posture change is more driven by inertia, making it difficult to directly associate with a specific activity type.

[0048] II. The motion parameter sensor signals in the stance phase have a lower noise level, as the foot's contact with the ground reduces the uncertainty brought by free motion. This stability allows algorithms to more accurately extract features, such as the smoothness of acceleration, the low amplitude of knee angular velocity, or the static value of knee angle. In the swing phase, the signal fluctuates more (e.g., acceleration may vary by 10%-50% due to swing speed), and features are easily disturbed by environmental factors (e.g., uneven ground) or individual differences (e.g., leg swing habits).

[0049] III. As the starting point of a gait cycle, the stance phase provides the earliest opportunity to identify the activity type. This early identification is crucial for the detection and control of subsequent gait phases. For example, after identifying an "ascending" activity, the algorithm can adjust the switching conditions of the swing phase (e.g., increase the knee angular velocity threshold), thereby optimizing the phase determination of the entire gait cycle.

[0050] Further, in one embodiment, the motion features include knee angle, knee angular velocity, tilt angle of calf and / or thigh, angular velocity of calf and / or thigh, acceleration of thigh and / or calf.

[0051] Further, in an embodiment, the motion feature includes a hip projection difference, which is a variation of the hip position of the lower limbs in a preset initial posture relative to a current state. Specifically, the difference in position change of the hip when moving in the horizontal direction (forward and backward) and the vertical direction (up and down) is used to capture the overall trend of leg movement. In simple terms, it "projects" the movement of the hip onto the ground and the sagittal plane, including the forward and backward movement (X direction) and the up and down movement (Y direction), and then calculates the difference between the two. This difference can represent whether the wearer's legs tend to step forward, step backward, or more raise or lower. For example, when walking forward, the horizontal projection usually changes more than the vertical direction; when walking backward, the horizontal direction may decrease or even become negative, while the vertical direction may increase slightly. By measuring the hip projection difference, different motion states can be more clearly distinguished, such as the difference between normal walking and walking backward. Specifically, in combination with the Figure 3 As shown in FIG. 11, the lower limbs in the preset initial posture are in an upright state, i.e., the knee joint angle is 180°. The hip projection difference includes an X-direction hip projection difference and a Y-direction hip projection difference.

[0052] The X-direction hip projection difference can be represented as the vector sum of the horizontal displacement, and the X-direction is positive in the horizontal forward direction: ΔX = -[l_shk · cos(θ_shk) + l_thg · cos(θ_thg)] where:

[0053] ΔX represents the hip projection difference of the hip in the X-direction (unit: meters);

[0054] l_shk, l_thg are the lengths of the shank and the thigh, respectively (typical values are both 0.42 meters);

[0055] θ_shk, θ_thg are the angles of the shank and the thigh relative to the vertical direction, respectively (unit: radians);

[0056] The Y-direction hip projection difference, in combination with normalization processing, can be represented as: ΔY = l_shk · sin(θ_shk) + l_thg · sin(π - θ_thg) - L_total where:

[0057] ΔY represents the hip projection difference of the hip in the Y-direction (unit: meters);

[0058] l_shk, l_thg are the lengths of the shank and the thigh, respectively (typical values are both 0.42 meters);

[0059] θ_shk, θ_thg are the angles of the shank and the thigh relative to the vertical direction, respectively (unit: radians);

[0060] sin(π - θ_thg);

[0061] L_total = l_shk + l_thg is the total length of the leg, used for normalization. After normalization, it is suitable for users of different heights or leg lengths.

[0062] The motion features based on the hip projection difference also include the rate of change of the hip projection difference, i.e. the time derivative of the hip projection difference, including the hip joint X direction hip projection difference differential and the hip joint Y direction hip projection difference differential, and then the rate of change of the hip projection difference is captured through a smoothing differentiator to further enhance the description of the motion trend.

[0063] In an embodiment, the hip projection difference plays a role in gait phase detection in the following way:

[0064] 1. Switching from support phase to swing phase:

[0065] When switching from the support phase to the swing phase, the hip projection difference provides the spatial context of the body's center of gravity and the position of the leg. For example, when the X direction hip projection difference becomes negative (e.g. less than -0.12 meters) and the Y direction hip projection difference shows an upward shift (e.g. greater than -0.08 meters), combined with the knee joint angular velocity exceeding 1.5 rad / s, the system determines that the leg has started to lift off the ground and enters the swing phase.

[0066] The dynamic change of the hip projection difference (their first order differential with respect to time) further assists in determining the switching time. For example, when the first order differential of the hip x-axis hip projection difference is > 0 and the first order differential of the hip y-axis hip projection difference is <-0.1, it indicates that the hip moves forward and the leg lifts up, strengthening the recognition of the swing phase.

[0067] 2. Switching from swing phase to support phase:

[0068] When switching from the swing phase to the support phase, the hip projection difference is used to detect the spatial motion features of the foot landing. For example, when the hip y-axis hip projection difference reaches a positive peak (e.g. greater than 0.15 meters) and its first order differential value (with respect to time) is > 0.25, it indicates that the leg has completed the swing and re-contacted the ground, triggering the determination of the support phase.

[0069] The extreme value of the hip projection difference combined with a time window (e.g. within 150 milliseconds after the peak occurs) ensures the robustness of the switching condition.

[0070] 3. Enhancement of activity type context:

[0071] Hip projection difference is not only used for phase switching, but also provides support for activity type recognition through its trends and amplitudes. For example, during upstroke, the X-direction hip projection difference is usually small (e.g., less than 0.05 meters) while the Y-direction hip projection difference shows a significant negative value (e.g., less than -0.07 meters), reflecting the spatial characteristics of the leg lifting motion. These motion features are used to dynamically adjust the switching threshold, making the phase detection adaptive to specific activities.

[0072] Compared with other motion features, hip projection difference has the following unique advantages:

[0073] 1. Comprehensive spatial context:

[0074] Hip projection difference directly reflects the two-dimensional displacement (x, y directions) of the leg relative to the body's center of gravity, providing more comprehensive spatial information than single joint angle or angular velocity. For example, knee joint angle only describes the local joint state, while hip projection difference captures the overall coordination of the hip and leg. This global perspective is particularly important in complex activities such as upstroke or down the stairs, as these activities involve significant body center of gravity movement.

[0075] 2. Sensitivity to center of gravity movement:

[0076] Hip projection difference is highly sensitive to changes in the wearer's center of gravity, effectively distinguishing between the stability of the support phase and the dynamics of the swing phase. For example, in walking, the process of hip x-axis projection changing from negative to positive reflects the trend of the center of gravity moving from the back to the front, while the negative change in hip y-axis projection indicates the height of leg lifting. This two-axis motion feature is superior to relying solely on knee joint angular velocity, which may lose its distinguishing power due to individual gait differences.

[0077] 3. Robustness and environmental adaptability:

[0078] Hip projection difference is based on geometric calculations (trigonometric functions), which are less sensitive to sensor noise (such as acceleration drift in IMU). In contrast, acceleration motion features are easily affected by external disturbances (such as uneven ground), while hip projection difference maintains high stability through the standardized calculation of leg length. This robustness makes it more advantageous in dynamic and variable scenarios such as mountain climbing.

[0079] 4. Low computational overhead:

[0080] The calculation of hip projection difference only involves basic trigonometric operations and addition and subtraction, with low time complexity, without the need for complex filtering or iterative optimization. Compared with combined motion features based on acceleration and angular velocity (such as deviation ratio or curvature change), hip projection difference maintains high efficiency while providing sufficient distinguishing power, meeting the design goal of lightweight algorithms in the present invention.

[0081] If the hip projection difference is removed and only other motion features are relied upon for gait phase detection, the following adverse consequences may occur:

[0082] 1. Phase switching accuracy decreases:

[0083] Without the spatial context of the hip projection difference, the system may struggle to accurately determine the position of the leg relative to the body. For example, in a stair descent activity, if only angular velocity is relied upon and the positive peak of the hip x-axis projection is ignored, a brief leg adjustment may be mistakenly identified as a swing phase, leading to incorrect application of control strategies (e.g., transparent control applied during the stance phase).

[0084] 2. Activity type differentiation ability weakens:

[0085] The hip projection difference provides a crucial spatial motion feature in activity recognition, such as the negative trend of Y-direction hip projection difference during ascending. If this motion feature is missing, the system may not effectively distinguish between walking and ascending (both angular velocity patterns may be similar), affecting the precision of dynamic threshold adjustment and ultimately leading to mismatched assistance output.

[0086] 3. Robustness decreases:

[0087] In complex terrains (e.g., uneven ground during mountain climbing), angular velocity and acceleration are susceptible to noise interference, while the hip projection difference maintains relative stability through geometric relationships. If the hip projection difference is not used, the system's adaptability to environmental changes will significantly decrease, potentially frequently entering an unknown phase during abnormal situations (e.g., temporary distortion of sensor signals), reducing the safety and comfort of the wearer.

[0088] 4. Transition smoothness is impaired:

[0089] The rate of change of the hip projection difference (first-order derivative of time) provides a smooth dynamic basis for gait phase switching. If this motion feature is missing, the switching condition may overly rely on instantaneous values (e.g., angular velocity threshold), leading to jitter or abrupt transitions in gait phases, affecting the continuity of exoskeleton control and user experience.

[0090] Therefore, the hip projection difference (x, y directions) significantly improves the accuracy and adaptability of gait phase detection by providing spatial context, center-of-gravity sensitivity, and robustness. Its unique advantage lies in efficiently representing complex motion patterns with low computational cost, enabling the system to maintain stable phase switching and control output in various activities (e.g., mountain climbing, ascending). If the hip projection difference is not used, the system will face issues such as decreased accuracy, weakened adaptability, and impaired transition smoothness, limiting its performance in dynamic scenarios.

[0091] In an embodiment, the hip projection difference plays a role in activity type judgment in the following ways:

[0092] The hip projection difference in the X direction and the Y direction is a spatial displacement index calculated based on the angles and lengths of the lower leg and the upper leg, and is used to represent the position change of the knee exoskeleton during movement. By combining the hip projection difference and its differential (i.e., the rate of change of the hip projection difference over time), the activity type of the user can be effectively determined, such as descending stairs, ascending stairs, or walking, etc. These hip projection differences reflect the relative movement trajectories of the hip in the horizontal (X direction) and vertical (Y direction), and are closely related to the leg posture and movement dynamics.

[0093] In activity type determination, the X direction hip projection difference is mainly used to detect the displacement movement characteristics in the horizontal direction. For example, in the activity of descending stairs, the X direction hip projection difference usually shows a small positive or negative offset, and its differential can further capture the dynamic changes of the movement, thereby distinguishing between fast and slow descending stairs. In the walking activity, the X direction hip projection difference usually shows a periodic negative offset, reflecting the movement characteristics of the legs stepping forward. In the activity of ascending stairs, the X direction hip projection difference may show a large positive offset, indicating the action of lifting the legs and advancing forward.

[0094] The Y direction hip projection difference mainly reflects the vertical displacement change and is often used to detect activities related to height changes. For example, when descending stairs, the Y direction hip projection difference is usually negative, indicating that the hip moves downward relative to the initial position; when ascending stairs, the Y direction hip projection difference may show a change from negative to positive, reflecting the process of lifting the leg and landing. By normalizing the Y direction hip projection difference (subtracting the total length of the leg), the influence of individual leg length differences can be eliminated, making the determination result more universal.

[0095] By integrating the X direction and Y direction hip projection differences and their differentials, the system can construct a multi-dimensional movement feature space, combined with other sensor data (such as knee joint angle, lower leg and upper leg angle, gyroscope data, etc.) to form the discrimination conditions of activity types. This method can capture the spatio-temporal characteristics of movement, thereby achieving accurate classification of different activity types.

[0096] In this regard, the hip projection difference has the following unique advantages:

[0097] 1. Spatial Intuitiveness: The hip projection difference directly reflects the displacement of the hip in three-dimensional space, providing more intuitive movement trajectory information than single angle or velocity, which helps to distinguish activities with similar angle movement characteristics but different spatial paths (such as walking and descending stairs).

[0098] 2. Dynamic Adaptability: By introducing the differential of the hip projection difference, the acceleration and trend changes of the movement can be captured, making the system more sensitive to dynamic activities (such as sudden leg lifting when quickly descending or ascending stairs).

[0099] 3. Robustness: The hip projection difference integrates the length and angle information of the lower leg and thigh, which can offset the influence of sensor noise or individual posture differences to some extent, making it more stable than directly using raw sensor data.

[0100] 4. Multi-dimensional synergy: The combination of X-direction and Y-direction hip projection differences provides a two-dimensional plane motion description, which is complementary to other motion features (such as knee angular velocity), enhancing the accuracy and specificity of activity classification.

[0101] If the X and Y direction hip projection differences of the hip are not used, the activity type judgment may face the following problems:

[0102] 1. Reduced resolution: Relying solely on a single motion feature such as angle or angular velocity may not effectively distinguish between activities with similar spatial trajectories but different dynamics. For example, going down stairs and walking may have similar knee angle ranges, but the horizontal and vertical displacement patterns of the hip are significantly different, and removing the hip projection difference will increase the misjudgment rate.

[0103] 2. Missing dynamic information: The differential of the hip projection difference provides real-time change information of the motion trend, and if not used, the system may have difficulty capturing fast actions or transition states (such as switching from standing to going down stairs), affecting the response speed and stability of the control algorithm.

[0104] 3. Reduced adaptability: The normalized design of the hip projection difference makes it suitable for users of different heights or leg lengths. If the hip projection difference is not used, the system may need additional calibration steps to adapt to individual differences, increasing the complexity of use.

[0105] 4. Limited control accuracy: In the control of the knee exoskeleton, the hip projection difference provides a key basis for gait phase switching and torque distribution. If this information is missing, the control algorithm may not accurately match the user's motion intention, resulting in insufficient or excessive exoskeleton assistance, affecting comfort and safety.

[0106] In summary, the hip projection difference plays an irreplaceable role in activity type judgment, which can significantly improve the control performance of the knee exoskeleton in complex terrain and dynamic activities.

[0107] Further, in an embodiment, the motion feature includes the product of thigh acceleration and knee angular velocity to represent the mechanical interaction in motion. Specifically, the product of thigh acceleration and knee angular velocity is designed as an intermediate variable, the core idea of which is to fuse two different dimensional motion parameters (linear acceleration and angular velocity) to extract more discriminative features, in order to enhance the accuracy of activity type recognition and gait phase detection. The specific applications in some embodiments include the following aspects:

[0108] 1. Characterization of mechanical interaction

[0109] This feature directly reflects the dynamic interaction between the linear motion of the thigh and the rotational motion of the knee joint. For example, in the swing phase, when the wearer lifts the leg, the thigh acceleration can exhibit a positive peak (e.g., 12 m / s²), while the knee joint angular velocity rapidly increases (e.g., 2.0 rad / s), resulting in a product of 24 m·rad / s³, indicating strong mechanical coupling. In the flat walk, due to the lower knee joint angular velocity (e.g., 0.8 rad / s), even if the acceleration is similar, the product value is smaller (e.g., 9.6 m·rad / s³). This difference provides a significant basis for distinguishing between activities.

[0110] 2. Enhancement of dynamic trends

[0111] Through multiplication, this feature amplifies the dynamic changes during motion. For example, when rapidly transitioning from the swing phase to the support phase (such as landing during running), the thigh acceleration can rapidly decrease from a high value (e.g., from 15 m / s² to 9.8 m / s²), while the knee joint angular velocity changes from a positive value to near zero (e.g., from 1.5 rad / s to 0.2 rad / s). The rapid change in the product (from 22.5 m·rad / s³ to 1.96 m·rad / s³) can sensitively capture this trend, thereby assisting in accurate determination of gait phase switching.

[0112] 3. Robustness of noise suppression

[0113] When using thigh acceleration or knee joint angular velocity alone, the signal can be affected by environmental noise (such as acceleration jitter caused by uneven ground) or sensor drift. By multiplying the two, this engineering feature to some extent smooths out transient abnormalities of a single variable. For example, if the acceleration exhibits an abnormal peak (e.g., 20 m / s²) due to a short-term disturbance, but the knee joint angular velocity remains low (e.g., 0.3 rad / s), the product value (6 m·rad / s³) remains within a reasonable range, avoiding false positives.

[0114] In specific applications in activity recognition, the product of thigh acceleration and knee joint angular velocity is used as a key discriminative indicator when extracting motion features in the support phase to determine activity type. In some embodiments, for example:

[0115] Flat walk: characterized by periodic fluctuations in the product value, typically ranging from 5 to 15 m·rad / s³, reflecting a stable gait rhythm.

[0116] Upward: characterized by a significant increase in the product value (e.g., 20 to 30 m·rad / s³) when lifting the leg, as both acceleration and angular velocity reach their peaks.

[0117] Squat: characterized by a small product value and a long duration (e.g., 2 to 5 m·rad / s³ for about 500 milliseconds), reflecting low-speed high-load characteristics.

[0118] Downward: characterized by a moderate product value and a gentle change (e.g., 10 to 20 m·rad / s³), with acceleration tending to stabilize due to low angular velocity.

[0119] By setting threshold rules combined with time window analysis (such as 300 milliseconds average), the system can efficiently distinguish different activity types.

[0120] Role in control strategy

[0121] The product of thigh acceleration and knee angular velocity also directly affects the selection of the actuator control strategy. For example:

[0122] When higher than a certain threshold (such as 25 m·rad / s³), it indicates that the wearer is in a high dynamic activity (such as running or ascending), and the system can choose a hybrid control that combines elastic and damping effects to provide strong assistance.

[0123] When it is lower and stable (such as 3 to 8 m·rad / s³), it indicates that the wearer is in a quasi-static state (such as squatting or standing), and the system can use elastic control to simulate spring effect.

[0124] Compared to using thigh acceleration or knee angular velocity alone, the product of thigh acceleration and knee angular velocity has the following advantages:

[0125] 1. Information fusion: By integrating linear and rotational motion information in the form of product, it provides a more comprehensive mechanical description than single variable.

[0126] 2. Dynamic sensitivity: The product operation amplifies the cooperative change of the two variables, making the feature response to motion intention more sensitive.

[0127] 3. Computational efficiency: Only one multiplication operation is needed to generate the feature, which meets the lightweight algorithm design goal and is suitable for real-time processing of embedded systems.

[0128] 4. Strong adaptability: The range and trend of the product value can adapt to different wearers' movement habits (such as stride or speed differences), without the need for additional calibration.

[0129] If the product of thigh acceleration and knee angular velocity is removed and only the original variables are relied on, the following problems may occur:

[0130] Downward differentiation: Single variable is difficult to capture the coupling effect of linear and rotational motion at the same time, which may confuse activities with similar dynamics (such as walking and slow ascending).

[0131] Dynamic detection deficiency: The amplification effect of product features on trend changes is lacking, and the system may miss the critical point of rapid transition (such as swing to support).

[0132] Inaccurate control: The control strategy may not accurately match the movement demand due to the lack of comprehensive mechanical indicators, resulting in insufficient or excessive assistance. Further, in an embodiment, the motion feature includes a normalized thigh acceleration-knee angular velocity ratio. By applying an upper threshold to limit the influence of outliers, the feature robustness is enhanced.

[0134] The core of the normalized thigh acceleration-knee angular velocity ratio lies in the positivity of the knee angular velocity, reflecting whether the knee is in flexion (e.g., the angular velocity is positive when lifting the leg) or extension (e.g., the angular velocity is negative when lowering the leg). By distinguishing this directionality, the system can better understand the wearer's leg movement intention. For example, in the swing phase, the knee usually flexes quickly, while in the support phase, the knee may slowly extend or remain stable.

[0135] On this basis, further combined with the relationship between thigh acceleration and knee angular velocity. Specifically, it generates a ratio by comparing the linear acceleration of the thigh with the rotational speed of the knee. This ratio is not simply divided by two, but is normalized to ensure that its value is within a controllable range, avoiding inconsistency due to the magnitude difference of the original data (such as acceleration is usually in meters per second squared, while angular velocity is in radians per second). The normalized ratio can more intuitively reflect the relative strength of the two, for example, in fast leg lifting, the thigh acceleration and knee angular velocity may be high at the same time, while in slow squatting, both may be low. In addition, an upper threshold is introduced to limit extreme values caused by sensor noise, external interference or abnormal wearer movements. For example, when the wearer is suddenly impacted by external force, the thigh acceleration may have a short-term abnormal peak, while the knee angular velocity remains normal, and the upper threshold can limit the ratio within a reasonable range, avoiding system misjudgment.

[0136] This feature performs well in various application scenarios of lower extremity exoskeletons, especially for activity type recognition and gait phase detection. For example:

[0137] Gait analysis during flat walking: In normal flat walking, the knee angular velocity is positive in the swing phase (knee flexion) and close to zero or negative in the support phase (knee extension). The normalized ratio can highlight the rapid increase of thigh acceleration in the swing phase and the synergistic effect of angular velocity, helping the system accurately distinguish between swing and support phases.

[0138] Dynamic capture during ascent: During ascent, the knee angular velocity exhibits a clear positive value during the swing phase, while the thigh acceleration increases significantly due to the leg lifting. The normalized ratio reflects this high dynamic state and is protected by an upper threshold to avoid abnormal fluctuations caused by uneven ground or excessive force from the wearer.

[0139] Stability judgment during squatting or standing: During squatting, the knee angular velocity is generally negative (knee slowly extends), and the thigh acceleration is also low and stable. The value of the normalized ratio is small and changes smoothly, indicating a low-speed motion state, and the system can be adjusted to a low-assistance mode accordingly.

[0140] Filtering of abnormal actions: When the wearer's thigh acceleration suddenly increases due to a fall or external impact, the knee angular velocity may not change synchronously, and the normalized ratio may have an abnormally high value. After the upper threshold intervenes, the ratio is limited to a reasonable range, preventing the system from mistakenly considering it as a high dynamic activity.

[0141] The reasons for its robustness include:

[0142] Positive and negative nature of knee angular velocity: By focusing on the directionality of angular velocity, the feature can naturally distinguish different phases of motion, avoiding ambiguity caused by a single numerical value. For example, even if the absolute value of angular velocity is the same, the positive and negative difference can clearly distinguish between flexion and extension intentions.

[0143] Normalization: Normalization ensures that the ratio is not directly affected by the dimension of sensor data or individual differences of the wearer (such as body weight, stride). For example, a heavier wearer may generate greater acceleration, but after normalization, the ratio remains consistent with the trend of a lighter wearer, enhancing the universality of the feature.

[0144] Protection of upper threshold: The upper threshold acts like a safety valve, preventing abnormal values from interfering with system judgment. For example, if the acceleration suddenly rises due to a sudden ground protrusion during running, the threshold mechanism can prevent the ratio from getting out of control, ensuring that the system still operates in the normal running mode.

[0145] The thigh acceleration-knee angular velocity ratio normalized from the outside can support the optimization of control strategies for lower extremity exoskeletons. In some embodiments, for example:

[0146] When the ratio is high and the knee angular velocity is positive, it indicates that the wearer is in a fast leg-lifting state (such as running or ascending), and the system can increase the assistance output of the actuator to provide greater thrust.

[0147] When the ratio is low and the angular velocity is negative, it indicates that the wearer may be slowly lowering the leg (such as squatting or descending stairs), and the system can switch to a damping mode to enhance stability and reduce impact.

[0148] When the ratio exceeds the upper threshold due to abnormal values, the system can temporarily maintain the current control strategy to avoid frequent mode switching due to temporary disturbances, thereby improving the continuity of the wearing experience.

[0149] Further, in an embodiment, the motion features include accumulating the positive knee joint angular velocity to quantify the sustained trend of knee joint motion; when the knee joint angular velocity is negative, the accumulated value is reset.

[0150] Further, in an embodiment, the motion features include the maximum / minimum values of the angles and angular velocities of the knee, thigh, and calf, representing the peak trend of motion.

[0151] Further, in an embodiment, the control strategy includes at least one of elastic control, damping control, and hybrid control; wherein:

[0152] Elastic control is to control the torque of the joint drive motor 1 acting on the swing arm 11 according to the knee joint angle to simulate a spring;

[0153] Damping control is to control the torque of the joint drive motor 1 acting on the swing arm 11 according to the angular velocity of the knee joint to simulate a damper;

[0154] Hybrid control is a combination of elastic and damping control to adapt to the current activity type.

[0155] The control strategy corresponding to the support phase includes at least one of elastic control, damping control, and hybrid control.

[0156] When executing the control strategy, the control parameters are adjusted in real time according to the current gait phase and / or activity type and / or live motion features. For example, adjusting the spring force or damping force of the simulated damper.

[0157] Regarding the control strategy corresponding to the standing support phase, it needs to be noted that:

[0158] In the standing support phase (hereinafter referred to as "support phase"), the system detects the motion state when the foot is in contact with the ground, and applies a specific control strategy to optimize the stability, comfort, and efficiency of the wearer. These strategies include elastic control, damping control, and hybrid control, each designed for different motion needs and scenarios.

[0159] For elastic control, the system applies a torque based on the joint angle, simulating the behavior of a virtual spring. For example, when the wearer is in a half-squatting position (knee joint angle of about 90 to 120 degrees, usually occurring when transitioning from standing to squatting), the system detects this angle change, indicating that the leg is in a static load or quasi-static transition state. At this time, the system applies an auxiliary torque proportional to the angle offset (e.g., torque size of 5 to 15 Nm, depending on the wearer's weight and joint stiffness requirements), to reduce the leg muscle burden and simulate the natural rebound effect of a spring. The benefit of this control is that it not only enhances the wearer's strength by providing assistance, but also reduces energy consumption when recovering from a posture (such as standing up from a squat), while avoiding discomfort caused by excessive joint flexion. Joint angle is detected by an inertial measurement unit (IMU), which combines accelerometer and gyroscope data to calculate the relative angle of the leg in real time, ensuring detection accuracy and response speed.

[0160] For damping control, the system applies a torque based on the knee joint angular velocity, simulating the buffering effect of a damper. For example, when the wearer is walking on uneven ground or quickly squatting, the knee joint angular velocity may reach 0.5 to 1.5 rad / s, indicating that the leg is experiencing dynamic adjustment or external disturbance (such as ground impact). In this scenario, the system applies a reverse torque proportional to the knee joint angular velocity (e.g., 3 to 10 Nm, depending on the size of the knee joint angular velocity), to slow down the joint movement speed and avoid instability or muscle fatigue caused by excessive flexion. The advantage of this control is that it can smooth the motion trajectory and suppress unnecessary oscillations (such as shaking during walking), thereby improving the safety and comfort of the wearer.

[0161] Hybrid control combines elastic and damping effects, dynamically adjusting according to the activity type. For example, in the scenario of "carrying heavy objects and squatting", the wearer may first enter a half-squat (triggering elastic control to provide a 10 Nm spring torque assist), and then experience rapid joint adjustment due to the shaking of the heavy objects (knee joint angular velocity of 1.0 rad / s, triggering damping control to apply a 5 Nm damping torque). This combined strategy provides assistance and stability simultaneously, ensuring that the system adapts to the wearer's needs in complex activities. Its advantage is that it combines the power enhancement of elasticity and the motion smoothing of damping, which can adapt to various support phase scenarios (such as ascending, standing up or walking with weights), thereby improving overall motion efficiency and wearing experience.

[0162] The fundamental reason for using these control strategies in the support phase is that this gait phase is a key point for activity type recognition and load support. The wearer's actions (such as walking, squatting or standing adjustment) usually start from the support phase, at which time the joint angle and knee joint angular velocity provide rich motion information, allowing the system to accurately determine the activity intention and optimize the torque output.

[0163] Moreover, the stability of the support phase and the assistance demand directly affect the fluency of the subsequent swing phase, so through the synergy of elastic, damping and hybrid control, the system not only improves the performance of the current gait, but also lays the foundation for the phase transition.

[0164] Further, in an embodiment, the control strategy comprises at least one of: transparent control, and reduced resistance control; wherein:

[0165] The control strategy corresponding to the swing phase comprises transparent control and / or reduced resistance control.

[0166] Regarding the control strategy corresponding to the swing phase, it needs to be explained that:

[0167] In the swing phase, the foot leaves the ground, and the leg is in a free motion state, and the system aims to minimize intervention or provide light enhancement to support natural movement and improve efficiency. The control strategy includes transparent control and reduced resistance control, which are respectively aimed at different dynamic needs.

[0168] Transparent control allows the wearer to move the leg in a natural way by setting the torque output to zero. For example, in normal walking, when the leg enters the swing phase (such as from the heel off the ground to the toe touching the ground), the knee joint swings freely at an angular velocity of about 0.8 to 1.2 rad / s. At this time, the system does not apply any active torque, but only relies on the low friction design of the device (such as through high-efficiency motors and transmission systems), to ensure that the wearer does not perceive additional resistance. The advantage of this control is that it maximizes the natural gait of the wearer, avoiding discomfort or energy waste due to unnecessary intervention, while maintaining high transparency of the device, making it suitable for daily activities (such as walking or jogging on flat ground).

[0169] Reduced resistance control, on the other hand, applies a small auxiliary torque to offset the effects of friction or gravity. For example, during long-distance walking or climbing, the knee joint in the swing phase may increase the burden on the wearer due to the weight of the device (usually 2 to 5 kg) or the resistance of the terrain. At this time, the system detects a decrease in knee joint angular velocity (such as below 0.5 rad / s, indicating that the swing is blocked), and applies a light auxiliary torque (for example, 1 to 3 Nm, in the same direction as the swing), to compensate for the resistance caused by device friction and gravity. The advantage of this control is that it reduces the muscle fatigue of the wearer without changing the natural motion trajectory, especially in long-term activities or high-intensity tasks (such as mountain climbing), significantly improving efficiency.

[0170] The reason for swing phase selection of these control strategies lies in the core requirements of freedom and efficiency. Transparent control meets the requirement of natural motion by zero torque output, while resistance reduction control optimizes the wearer's experience by fine-tuning torque to compensate for inherent physical limitations of the device (such as mass or friction). In addition, the control of the swing phase directly affects the continuity of the gait cycle, and a light intervention strategy can ensure seamless connection with the support phase and avoid gait disorders caused by excessive control.

[0171] Further, in an embodiment, the gait phase further includes an unknown phase, the unknown phase corresponding to at least one of the following situations: a transition state when the motion characteristics do not clearly conform to the support or swing phase, the algorithm cannot determine the current gait, and the algorithm initialization; the control strategy of the unknown phase includes transparent control and / or resistance reduction control to ensure safety and comfort.

[0172] Regarding the control strategy corresponding to the unknown phase, it needs to be explained that:

[0173] When the system cannot clearly determine the current gait phase (i.e., in the unknown phase), transparent control or resistance reduction control is preferred to ensure safety and comfort. This strategy is suitable for scenarios where the gait transition is ambiguous or the motion parameter sensor signal is uncertain.

[0174] In the unknown phase, the system defaults to transparent control, maintaining zero torque output. For example, when the wearer suddenly starts walking from standing, but the initial motion parameter sensor data (such as IMU signal) cannot accurately distinguish between support and swing due to noise or temporary obstruction, the system avoids applying any active torque and relies solely on the passive characteristics of the device to support motion. The advantage of this method is that it avoids the risk of misjudgment (such as stumbling due to the application of damping torque in the swing phase) by zero intervention, thereby ensuring the safety and smoothness of the wearer's action.

[0175] If a slight motion trend is detected (such as a knee joint angular velocity between 0.2 and 0.5 rad / s, indicating a potential entry into swing or fine-tuning posture), the system can switch to resistance reduction control and apply a small torque (for example, 0.5 to 2 Nm) to support the potential action intention. For example, when the wearer slowly adjusts the standing posture or prepares to take a step, this light assistance can offset the device friction to ensure the smoothness of the action start. The advantage of this strategy is that it can still provide moderate support in uncertainty while minimizing intervention and avoiding interference with the wearer.

[0176] The reason for adopting these control strategies in the unknown phase is safety and robustness. When the gait phase is not clear, any aggressive control (such as high torque output) may cause discomfort or even danger, so preferential transparent control can effectively avoid risks. And the drag reduction control is a conservative supplement to ensure that the system is still responsive under possible motion intention. This design not only improves the fault tolerance of the system, but also makes it reliable and comfortable to wear in complex or non-standard scenarios (such as irregular gait or sudden action).

[0177] Further, in an embodiment, further comprising corresponding adjustment of gait transition conditions and / or control parameters of control strategies according to activity type and / or motion characteristics. Ensure that the exoskeleton adapts to changing scenarios.

[0178] The motion characteristics extracted in conjunction with the support phase are not only used to classify the current activity, but also directly affect the logic of subsequent gait switching: Specifically:

[0179] I. Transition from support phase to swing phase. According to the activity type identified in the support phase, the algorithm dynamically adjusts the switching conditions. For example, if it is identified as "squatting", it requires a longer acceleration stabilization time (such as 500 milliseconds) to confirm the end of the support phase, to avoid misjudging the swing start due to a short rise; If identified as "up", it may shorten the stabilization time (such as 150 milliseconds) and increase the knee joint angular velocity threshold (such as 1.5 rad / s) to adapt to fast leg lifting. This adjustment ensures the accuracy of gait phase determination and the relevance of gait switching.

[0180] II. Realization of seamless control: By extracting motion characteristics in the support phase, the system can establish a motion context early in the gait cycle, thereby providing continuous instructions for the control of exoskeletons or mobility aids. For example, the characteristics identified in the "down" activity can trigger a more moderate actuator response to avoid sudden changes.

[0181] It should be noted that: phase transition refers to the process of switching the exoskeleton system from one gait phase (such as standing support phase) to another gait phase (such as swing phase or unknown phase) according to the real-time motion state of the wearer. This transition is achieved by detecting specific gait characteristics (such as joint angle, knee joint angular velocity) and comparing them with predefined thresholds or conditions. The core logic of gait phase transition relies on multi-motion parameter sensor data fusion and dynamic threshold judgment to ensure that the system can accurately identify the wearer's motion intention and adjust the control strategy in a timely manner.

[0182] In the stance phase, the system continuously monitors the knee joint angular velocity filtered value. When this parameter breaks through the preset dynamic threshold (typical value 0.8 rad / s) and is accompanied by a hip forward projection offset of 0.25 m, the transition to swing phase is triggered. This process uses a triple verification mechanism: first, the thigh inertia measurement unit captures the persistence of the motion trend, second, the joint angle extreme locking technology is used to suppress transient noise interference, and finally, the confidence check is combined with the phase prediction model of the historical gait data to ensure the reliability of the transition decision.

[0183] The adjustment of the activity type to the transition logic is reflected in the dynamic compensation of the threshold and the control domain remapping. Taking the up scenario as an example, the system raises the knee joint angular velocity decision threshold to 1.5 rad / s to compensate for the response delay caused by increased joint load, and introduces a kinematic compensation algorithm for the hip forward projection offset to dynamically correct the forward movement of the body center of gravity when climbing stairs.

[0184] Further, in an embodiment, the motion characteristics further include the frequency domain characteristics of the knee joint motion, so that in a complex terrain scenario, the system characterizes the ground roughness through the frequency domain characteristics of the knee joint motion (such as the tremor component of 2-4 Hz), dynamically couples the spring elastic coefficient in the elastic control and the damping coefficient of the damper in the damping control, such as, when the terrain complexity is too large, i.e., the ground roughness exceeds the critical value, the damping coefficient is proportionally raised to the maximum value, and the elastic coefficient is attenuated to 60% of the reference value, effectively suppressing joint oscillation caused by irregular impact.

[0185] For low-speed high-load activities such as squats, the system constructs a nonlinear elastic control strategy: when the knee joint angle enters the 60°-100° working interval, an auxiliary torque is applied that is exponentially related to the rate of change of the angle, and the torque produces a peak assist (typical value 12 Nm) when the angle reaches 90°. In this process, the Y-direction hip projection difference is used as the stance phase maintenance condition, and when it exceeds 0.1 m for 200 ms, the system will forcibly lock the current control parameters to prevent false switching caused by accidental center of gravity fluctuations.

[0186] The abnormal condition handling mechanism uses a simple safety strategy: when entering an unknown phase, the system automatically switches to a transparent control mode.

[0187] In some embodiments, the method of gait phase detection and gait phase switching condition control is specifically manifested as:

[0188] Before adjusting the switching conditions according to the detected activity type, the gait phase switching conditions are based on pre-defined rules and characteristic trends of sensor data, as follows:

[0189] The switching condition from swing phase to stance phase is: when the sensor detects that the foot is in contact with the ground, for example, the thigh acceleration value tends to be stable (e.g. close to the typical range of gravitational acceleration, such as near 9.8 m / s²), and the knee joint angular velocity amplitude significantly decreases (e.g. below a certain preset threshold, such as 0.5 rad / s), it is determined to enter the stance phase. Because, the stable acceleration and lower knee joint angular velocity reflect the static characteristics of the movement after the foot is in contact with the ground, which is in contrast to the dynamic changes of the swing phase.

[0190] The switching condition from stance phase to swing phase is: when the sensor detects that the foot is off the ground, for example, the thigh acceleration appears significant fluctuations (such as ±10% range), and the knee joint angular velocity amplitude increases (e.g. exceeds a certain preset threshold, such as 1.0 rad / s), it is determined to enter the swing phase. Because, the acceleration fluctuations and the increase of knee joint angular velocity indicate that the foot begins to leave the ground and enters a dynamic state of movement.

[0191] The condition for determining the unknown gait phase is: when the sensor data does not clearly meet the characteristic conditions of the stance phase or the swing phase, for example, the combination of the acceleration measured at the thigh and the knee joint angular velocity is in the ambiguous interval of the predefined threshold (e.g. the knee joint angular velocity fluctuates between 0.5-1.0 rad / s and the acceleration is not stable), or the data is disturbed by noise causing the characteristics to be unclear, it is determined to be in the unknown gait phase. As the default processing of the algorithm for the transition state or the uncertain state, it usually occurs at the boundary of the gait phase switching or when the algorithm is initialized.

[0192] These unadjusted switching conditions are often based on fixed thresholds and direct trend analysis of sensor data, suitable for general gait patterns (such as walking on flat ground).

[0193] Further, after the activity type changes, the gait phase switching conditions will be adjusted. That is, when a specific activity (such as going up, going down, squatting, etc.) is detected, the gait phase switching conditions will be dynamically adjusted according to the activity characteristics to improve the accuracy and adaptability of detection. In some embodiments, the specific adjustment method is as follows:

[0194] The activity characteristics (such as knee joint angle, calf inclination angle, hip projection difference, etc.) extracted by the activity recognition module are used to represent the context of the current movement.

[0195] Adjust the threshold or feature weight according to the activity type. For example, a higher knee joint angular velocity threshold may be required when going up, and the duration requirement of the stance phase may be extended when squatting.

[0196] When going up, for example, the switching condition in some embodiments is:

[0197] Switching condition from swing to support: require shorter acceleration settling time at thigh (e.g. shorten from 200ms to 150ms), while higher knee angular velocity peak (e.g. adjust from 1.0 rad / s to 1.5 rad / s) to accommodate faster foot landing.

[0198] Switching condition from support to swing: detect larger knee angle change (e.g. exceed 60 degrees) as additional condition, reflecting leg lifting motion.

[0199] Switching condition as follows in some embodiments when ascending:

[0200] Switching condition from swing to support: lower knee angular velocity threshold (e.g. adjust from 1.0 rad / s to 0.8 rad / s) to accommodate slower falling motion.

[0201] Switching condition from support to swing: extend acceleration fluctuation detection window (e.g. extend from 200ms to 300ms) to capture smooth transition.

[0202] Switching condition as follows in some embodiments when descending:

[0203] Support phase prolongation: require acceleration to settle and knee angle to significantly decrease (e.g. less than 120 degrees) for a time exceeding a certain threshold (e.g. 500ms) to avoid false positive as short pause.

[0204] By dynamically adjusting thresholds and time windows, switching conditions can better match the motion characteristics of specific activities, avoiding false positives. For example, prevent false positive as swing phase due to fast landing when ascending, or prevent false positive as swing start due to longer support time when descending.

[0205] Further to ensure seamless control of gait phase switching, the following control logic is adopted:

[0206] Smooth transition mechanism: use overlapping time window (e.g. 50-100ms) at gait phase switching boundary for smooth processing of motion features. For example, by applying sliding average (e.g. 5 or 10 point average) to thigh acceleration and knee angular velocity, reduce switching jitter due to noise or transient fluctuations.

[0207] When potential switching is detected (e.g. from support to swing), the algorithm will briefly maintain the current gait phase state (e.g. additional 50ms) before confirming the new gait phase, to verify the persistence of feature trends.

[0208] State continuity check: requires the feature to meet the switching condition for more than a minimum duration (e.g., 100 ms) to avoid false switching due to transient anomalies (e.g., sensor jitter). For example, when switching from stance to swing phase, only when the knee angular velocity continuously exceeds the threshold and the acceleration fluctuation consistently occurs, the switch is confirmed.

[0209] Dynamic threshold update: adjusts the threshold in real-time based on the activity type to ensure the switching condition is consistent with the current motion context. For example, increase the angular velocity threshold when running to match the faster swing frequency. If the activity type changes (e.g., switching from walking to ascending), the algorithm updates the threshold within the first complete gait cycle after detecting the new activity to avoid misjudgment of gait phase during the transition period.

[0210] Fallback mechanism: if an anomaly is detected after switching (e.g., acceleration suddenly stabilizes after switching to swing phase), the algorithm can fall back to the previous gait phase within a short time (e.g., 100 ms) and mark it as an unknown gait phase, waiting for the data to stabilize before re-determining. This mechanism is particularly suitable for transition states or unstable sensor signals.

[0211] Seamless control results: through the above logic, this method ensures the continuity and accuracy of gait phase switching, avoiding interruptions in control signals. For example, in the application of exoskeletons or walking aids, the smoothness of gait phase switching can directly translate into seamless response of motor control, improving the comfort and safety of the wearer.

[0212] Further, in an embodiment, the process of activity recognition includes:

[0213] Threshold rule-based feature comparison, using pre-defined threshold rules to compare motion features to determine if the conditions of a specific activity are met;

[0214] Extreme value detection within a time window, using a time window (e.g., 100 to 350 ms) to analyze the extreme values (such as maximum or minimum) of motion features to capture the dynamic changes of motion. For example, by detecting the maximum value of calf tilt angle and the peak value of hip projection difference, it is determined whether there is a feature pattern related to ascending or descending, and then filtered by time conditions (e.g., the duration after the peak value occurs) to ensure the robustness of detection;

[0215] Cumulative counting and state persistence check, cumulative counting of motion features that meet pre-set conditions, such as the number of times the calf angle peak value exceeds a certain threshold is continuously detected, or the duration of the knee angular velocity remaining within a certain range. Through state persistence check (e.g., requiring a certain combination of motion features to continuously meet the conditions for a certain time), to avoid false judgments caused by transient noise or abnormal data;

[0216] Dynamic update, within a support phase, dynamically update motion features, such as the product of knee joint angular velocity and hip projection difference, or feature integration based on time window to characterize the trend and intensity of motion. Through these dynamically updated motion features, distinguish activity types; for example, upstroke can exhibit greater knee joint angular change and negative hip projection difference, while downstroke can exhibit more stable knee joint, thigh angular velocity and specific shank angle range.

[0217] Conditional logic combination, use conditional logic combination (e.g. "and", "or" logic) to jointly analyze multiple motion features. For example, if the knee joint angle exceeds a certain threshold, while the knee joint angular velocity shows a positive trend, and the hip projection difference is less than a certain negative value, it can be determined as upstroke activity. This logical combination avoids complex calculation models and relies only on basic mathematical operations (such as addition, subtraction, multiplication, division, comparison), thereby achieving lightweight.

[0218] Therefore, in this application, the process of activity recognition is a lightweight analysis method based on the trend of motion parameter sensor data, which does not rely on resource-intensive techniques (such as machine learning or deep learning), but is based on direct feature extraction and threshold logic judgment of real-time motion parameter sensor data. Through pre-defined rules and time window analysis, the computational complexity is reduced to ensure efficient operation on embedded devices. All feature calculations use low-overhead mathematical operations (such as addition, multiplication, comparison), and combine cached data within a limited time window to avoid large-scale data storage and processing.

[0219] Through the above lightweight analysis method, various activity types can be efficiently distinguished. For example:

[0220] Flat walk: characterized by stable knee joint angular velocity fluctuations and knee joint angle changes.

[0221] Upstroke: characterized by large knee joint angle changes, negative Y-direction hip projection difference, and specific shank inclination angle.

[0222] Downstroke: characterized by stable knee joint angular velocity and small shank inclination angle.

[0223] Running: characterized by high-frequency fluctuations in knee joint angular velocity and significant enhancement of thigh acceleration. Specifically, during running, the knee joint angular velocity shows rapid and periodic positive and negative alternation, reflecting the rapid flexion of the leg during the swing phase (leg lifting) and the brief extension during the support phase (landing). At the same time, the thigh acceleration reaches a peak in the swing phase, especially showing a large positive change when the leg moves forward, while accompanied by a brief deceleration impact when landing. The knee joint angle changes significantly and stably, and the shank inclination angle frequently switches with the pace, showing a high-dynamic motion pattern as a whole.

[0224] Squatting: characterized by a significant decrease in knee joint angle and a longer duration of support phase.

[0225] Backward walking: characterized by a negative X-direction hip projection difference and a positive Y-direction hip projection difference, and a positive cumulative value of knee joint angular velocity.

[0226] Unknown: when the feature pattern does not meet any known activity condition, it is classified as unknown.

[0227] The technical principles of the present application are described above in combination with specific embodiments, and these descriptions are only for the purpose of explaining the principles of the present application, and cannot be interpreted in any way as a limitation on the scope of protection of the present application. Based on the explanations here, those skilled in the art can think of other specific embodiments of the present application without creative labor, and these embodiments will fall within the scope of protection of the present application.

Claims

1. A method for recognizing the motion of a knee exoskeleton, characterized in that: Based on a lower limb exoskeleton device, the lower limb exoskeleton device includes a thigh extension frame (2) and a lower limb extension frame (3) respectively strapped to the thigh and lower leg. The thigh extension frame (2) and the lower limb extension frame (3) are arranged to rotate relative to each other around the knee joint. The lower limb exoskeleton device is equipped with motion parameter sensors; including the following process: Motion parameter sensors are used to collect motion parameters of the lower limb exoskeleton device to determine motion characteristics, and then the motion state is confirmed based on the motion characteristics. The motion characteristics include: knee joint angle, knee joint angular velocity, tilt angle of the lower leg and / or thigh, angular velocity of the lower leg and / or thigh, acceleration value of the thigh and / or lower leg, and hip projection difference, where the hip projection difference is the change in hip position relative to the lower limb in a preset initial posture in the current state. The lower limbs are in a preset initial posture, including the thighs and lower legs being in an upright position; The hip projection difference includes the hip projection difference in the X direction and the hip projection difference in the Y direction; The difference in hip projection in the X direction is represented as the vector sum of horizontal displacements, with the X direction being positive forward: ΔX = -[l_shk · cos(θ_shk) + l_thg · cos(θ_thg)], where: ΔX represents the difference in hip projection in the X direction; l_shk and l_thg are the lengths of the lower leg and thigh, respectively; θ_shk and θ_thg are the angles of the lower leg and thigh relative to the vertical direction, respectively; The difference in hip projection in the Y direction, with the vertical upward direction as positive, can be expressed as: ΔY = l_shk · sin(θ_shk) + l_thg · sin(π - θ_thg) - L_total Where: ΔY represents the difference in hip projection in the Y direction; l_shk and l_thg are the lengths of the lower leg and thigh, respectively; θ_shk and θ_thg are the angles of the lower leg and thigh relative to the vertical direction, respectively; L_total = l_shk + l_thg represents the total length of the legs.

2. A knee joint assist device, characterized in that: Including a lower limb exoskeleton device, the lower limb exoskeleton device comprising: A joint drive motor (1) includes a pair of rotating components, the joint drive motor (1) being used to apply torque to the pair of relatively rotating rotating components, the center of the relatively rotating components corresponding to the knee joint; The thigh extension frame (2) and the calf extension frame (3) are respectively connected to a pair of rotating components, and the thigh extension frame (2) and the calf extension frame (3) are respectively tied to the thigh and the calf; It also includes motion parameter sensors installed on the lower limb exoskeleton device. The motion parameter sensors are used to collect motion parameters of the lower limb exoskeleton device to determine motion characteristics. The motion characteristics include: knee joint angle, knee joint angular velocity, tilt angle of the lower leg and / or thigh, angular velocity of the lower leg and / or thigh, acceleration value of the thigh and / or lower leg, and hip projection difference. The hip projection difference is the change in hip position relative to the lower limb in a preset initial posture in the current state. The lower limbs are in a preset initial posture, including the thighs and lower legs being in an upright position; The hip projection difference includes the hip projection difference in the X direction and the hip projection difference in the Y direction; The difference in hip projection in the X direction is represented as the vector sum of horizontal displacements, with the X direction being positive forward: ΔX = -[l_shk · cos(θ_shk) + l_thg · cos(θ_thg)], where: ΔX represents the difference in hip projection in the X direction; l_shk and l_thg are the lengths of the lower leg and thigh, respectively; θ_shk and θ_thg are the angles of the lower leg and thigh relative to the vertical direction, respectively; The difference in hip projection in the Y direction, with the vertical upward direction as positive, can be expressed as: ΔY = l_shk · sin(θ_shk) + l_thg · sin(π - θ_thg) - L_total Where: ΔY represents the difference in hip projection in the Y direction; l_shk and l_thg are the lengths of the lower leg and thigh, respectively; θ_shk and θ_thg are the angles of the lower leg and thigh relative to the vertical direction, respectively; L_total = l_shk + l_thg represents the total length of the legs.

3. A knee joint assist device according to claim 2, characterized in that: The motion parameter sensors are a pair of IMUs respectively mounted on the thigh extension frame (2) and the calf extension frame (3).

4. A knee joint assist device according to claim 2, characterized in that: The motion parameter sensors are an IMU mounted on the thigh extension frame (2) or the calf extension frame (3) and a knee joint angle sensor.

5. A knee joint assist device according to claim 2, characterized in that, Also includes: The thigh binding device (21) is installed on the thigh extension frame (2) and is bound to the thigh; The first calf binding device (31) and the second calf binding device (32) are installed on the calf extension frame (3) and are bound between the calf and the knee joint.

6. A knee joint assist device according to claim 5, characterized in that: The thigh binding device (21) includes an upper connecting seat (211) and a first binding member (212). The upper connecting seat (211) is rotatably mounted on the thigh extension frame (2). The rotation axis of the upper connecting seat (211) is horizontal. The upper connecting seat (211) fits against the back of the thigh. The first binding member (212) binds the thigh to the upper connecting seat (211).

7. A knee joint assist device according to claim 5, characterized in that: The first calf binding device (31) includes a lower connecting seat (311) and a second binding member (312). The lower connecting seat (311) is rotatably mounted on the calf extension frame (3). The rotation axis of the lower connecting seat (311) is horizontal. The lower connecting seat (311) is attached to the back of the calf. The second binding member (312) binds the calf to the lower connecting seat (311).

8. A knee joint assist device according to claim 5, characterized in that: The second lower leg binding device (32) includes a knee connector (321) and a third binding member (322). The knee connector (321) is mounted on the lower leg extension frame (3) and fits against the side of the lower leg. The third binding member (322) binds the lower leg to the knee connector (321). The knee connector (321) is located above the lower connector (311).

9. A knee joint assist device according to claim 5, characterized in that: The joint drive motor (1) is located on the side of the knee joint. The thigh extension frame (2) and the calf extension frame (3) are integrally bent plates. The ends of the thigh extension frame (2) and the calf extension frame (3) away from the joint drive motor (1) extend to the back of the leg. The thigh binding device (21) and the first calf binding device (31) are respectively installed on the ends of the thigh extension frame (2) and the calf extension frame (3) away from the joint drive motor (1).

10. A knee joint assist device according to claim 2, characterized in that: It also includes a power module, which is electrically connected to the joint drive motor (1), and the power module is located at or above the waist.

Images