A knee exoskeleton motion monitoring method based on inertial sensor

Abstract

A knee joint exoskeleton motion monitoring method based on inertial sensors. After the inertial sensors are installed, the gait cycle is designed. First, the motion trajectories of the wearer's joints and leg ends are obtained. At the same time, the acceleration data collected by the calf exoskeleton sensor is integrated to obtain the calf exoskeleton motion trajectory. The foot motion trajectory is inferred based on the obtained motion trajectory. The obtained trajectories are combined to obtain the overall motion trajectory of the lower limbs, thereby completing the exoskeleton motion data collection and analysis.

Description

Technical Field

[0001] The invention relates to a knee joint exoskeleton motion monitoring method based on an inertial sensor, and belongs to the field of novel exoskeleton design. Background Art

[0002] As a new type of exoskeleton, light-load power-assisted exoskeleton has gradually attracted attention and research from all parties with its higher mobility and lower energy consumption. Compared with traditional heavy-load exoskeletons that provide comprehensive assistance to human limbs, light-load power-assisted exoskeletons mostly provide assistance for one or several specific joints. In order to reduce the weight of the exoskeleton and improve the mobility of the system, light-load exoskeletons are mostly driven by motors to drive the movement of flexible or light rigid structures; at the same time, the torque of the motor is reduced, and the motor is no longer used as the main power to drive the movement of the human body and exoskeleton, but the user's movement is used as the main power. When the user's muscles exert force, the exoskeleton provides auxiliary power to specific muscle groups, thereby improving the user's exercise effect and reducing the user's exercise consumption. Compared with traditional heavy-load exoskeletons, light-load exoskeletons pay more attention to the control system's perception of human movement. The control system needs to provide corresponding auxiliary torque to the joint-related muscle groups at a specific time. The perception of the power-assistance range and the size of the power-assistance has become an important factor in optimizing the power-assistance effect of light-load exoskeletons, and detecting the user's movement state has become a key technology.

[0003] As a light-load exoskeleton for a specific joint, the knee exoskeleton can provide assistance to the wearer during walking and climbing stairs. Walking and climbing stairs are the main situations faced by human legs, and can be used as a reference to approximately simulate most human leg activities. In the future, the knee exoskeleton can be used to assist the elderly and the disabled, field marches, mountaineering expeditions, and fire rescue.

[0004] The current sensing method of the knee exoskeleton mainly adopts the sensing method of the lower limb exoskeleton, that is, installing inertial sensors on the thigh, calf and foot. This method has been proven to be effective in the laboratory, but the knee exoskeleton needs to install an additional foot sensing module, which poses certain challenges to the productization of the knee joint. This design will make the product structure more complicated. However, if the foot sensor module is not used, the sensing accuracy will decrease, which will affect the decision-making of the control algorithm in terms of data accuracy. Summary of the invention

[0005] The technical problem solved by the present invention is: in view of the defects of the knee joint exoskeleton motion monitoring method in the current prior art, a knee joint exoskeleton motion monitoring method based on inertial sensors is proposed.

[0006] The present invention solves the above technical problems by the following technical solutions:

[0007] A knee exoskeleton motion monitoring method based on inertial sensors, the steps are as follows:

[0008] (1) Install the inertial sensor on the outside of the leg exoskeleton in the specified installation direction;

[0009] (2) Using an inclinometer, measure the angle between the thigh and calf of the leg exoskeleton, measure the size of the wearer's lower limbs, calculate the real-time motion position of the wearer's lower limbs on the sagittal plane, generate the motion trajectory of the wearer's joints and leg ends, that is, the heel excluding the sole of the foot, and generate a connecting rod structure based on the obtained motion trajectory;

[0010] (3) integrating the acceleration data collected by the exoskeleton sensor of the calf part to obtain the motion trajectory of the exoskeleton of the calf part, and calculating the motion trajectory of the sole of the foot according to the obtained motion trajectory;

[0011] (4) The motion trajectories obtained in step (2) and step (3) are combined to obtain the overall motion trajectory of the lower limbs.

[0012] The three-axis installation directions of the inertial sensor are: forward direction, direction perpendicular to the sagittal plane, and direction perpendicular to the ground.

[0013] The gait cycles of the joints, the end motion trajectories of the legs, and the sole motion trajectories are the same. In a single step gait cycle:

[0014] The segmentation time point of the gait cycle of the leg exoskeleton is determined based on the angular velocity of the axis of the inertial sensor perpendicular to the sagittal plane. During the single-leg movement of the exoskeleton, the midpoint of the support state is taken as the starting point of the gait cycle. At the beginning of the gait cycle, the acceleration data integration data and the lower limb end position data are recorded simultaneously. After the gait cycle ends, the final stride length is obtained based on the maximum distance of the lower limb end and the calculated integration result.

[0015] The integral data is calculated and recorded from the start of the gait cycle until the sole of the foot leaves the ground safely; before the end of the gait cycle, the integral result is calculated based on the recorded integral data.

[0016] The integral data is obtained based on acceleration integral calculation. Through the zero correction method, the motion trajectory of a specific time period is selected in each gait cycle, and an approximate conversion is performed based on the angular velocity of the sensor perpendicular to the sagittal plane, and then recalculated in the next gait cycle to avoid the accumulation of gait cycle errors.

[0017] The advantages of the present invention compared with the prior art are:

[0018] The present invention provides a knee joint exoskeleton motion monitoring method based on inertial sensors. The method analyzes the motion state of the lower limbs by combining a geometric method with a kinematic integration method, synthesizes gait data with higher accuracy, and utilizes the combined use of different functions of multi-axis inertial sensors to judge each time point in the human walking gait cycle. The method can approximately infer the foot motion through the leg sensors without adding foot sensors. At the same time, the method is installed on the exoskeleton structure without adding additional structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of the installation of the inertial sensor for the knee exoskeleton provided for the invention;

[0020] Figure 2 A human lower limb motion solution diagram provided for the invention;

[0021] Figure 3 The foot motion solution provided by the invention is related to the IMU motion trajectory diagram;

[0022] Figure 4 A schematic diagram of the integral interval selection of the angular velocity curve during walking provided by the invention;

[0023] Figure 5 A schematic diagram of the integral interval selection of the angular velocity curve when climbing stairs provided by the invention; DETAILED DESCRIPTION

[0024] A knee exoskeleton motion monitoring method based on inertial sensors is specifically used to monitor the user's motion data collection and analysis during the use of the exoskeleton. It can measure data such as step length, step frequency, step speed and slope, and then feed back to the control system. The specific monitoring process is as follows:

[0025] (1) Install the inertial sensor on the outside of the leg exoskeleton in the specified installation direction;

[0026] The three-axis installation directions of the inertial sensor are: forward direction, perpendicular to the sagittal plane, and perpendicular to the ground;

[0027] (2) Using an inclinometer, measure the angle between the thigh and calf of the leg exoskeleton, measure the size of the wearer's lower limbs, calculate the real-time motion position of the wearer's lower limbs on the sagittal plane, generate the motion trajectory of the wearer's joints and leg ends, and generate a connecting rod structure based on the obtained motion trajectory;

[0028] (3) integrating the acceleration data collected by the exoskeleton sensor of the calf part to obtain the motion trajectory of the exoskeleton of the calf part, and calculating the motion trajectory of the sole of the foot according to the obtained motion trajectory;

[0029] Among them, the gait cycle of the joint and leg end motion trajectory and the foot sole motion trajectory is the same. In a single foot gait cycle:

[0030] The segmentation time point of the gait cycle of the leg exoskeleton is determined based on the angular velocity of the axis of the inertial sensor perpendicular to the sagittal plane. During the single-leg movement of the exoskeleton, the midpoint of the support state is taken as the starting point of the gait cycle. At the beginning of the gait cycle, the acceleration data integration data and the lower limb end position data are recorded simultaneously; after the gait cycle ends, the final stride length is obtained based on the maximum distance of the lower limb end and the calculated integration result;

[0031] The integral data is calculated and recorded from the beginning of the gait cycle until the sole of the foot leaves the ground safely; before the end of the gait cycle, the integral result is calculated based on the recorded integral data;

[0032] The integral data is obtained by calculating the integral of acceleration. Through the zero correction method, the motion trajectory of a specific time period is selected in each gait cycle, and the gold wire conversion is performed according to the angular velocity of the sensor perpendicular to the sagittal plane, and then recalculated in the next gait cycle to avoid the accumulation of gait cycle errors.

[0033] (4) The motion trajectories obtained in step (2) and step (3) are combined to obtain the overall motion trajectory of the lower limbs.

[0034] The following is further described based on specific embodiments:

[0035] In the current embodiment, the knee exoskeleton is composed of four 9-axis inertial sensors, which are three-axis acceleration, three-axis angular velocity, and three-axis angle. When walking with the exoskeleton, the inertial sensor is installed on the outside of the leg. Through the real-time acquisition of 9-axis data, a sagittal angle meter can be used to build a real-time lower limb bar graph of human walking. The gait state is judged by the angular velocity, and the displacement caused by the foot movement is calculated by the acceleration integration.

[0036] like Figure 1 As shown, the sensors are installed on the exoskeleton structure on the outside of the leg, and the installation method is shown in the module indicated by the arrow in the figure. The three axes of the sensor have fixed installation directions, one axis points to the forward direction, one axis is perpendicular to the sagittal plane, and the other axis is perpendicular to the ground.

[0037] Angle errors are inevitable during installation, so this method corrects the sensor coordinate axis through an initialization algorithm, with the goal of reaching the predetermined installation direction.

[0038] like Figure 2 As shown, based on the angles of the thigh and calf measured by the goniometer and the size of the wearer's lower limbs, the real-time movement position of the human lower limbs is calculated on the sagittal plane, and the movement trajectory of the joints and leg ends is generated, and the connecting rod structure is generated based on the experimental results.

[0039] By monitoring a specific time point, this method can determine the segmentation point between two gait cycles. Figure 4 , 5 As shown, certain specific time points in the user's walking process can be determined based on the angular velocity of the axis of the inertial sensor perpendicular to the sagittal plane, and one of the time points is used to segment the gait cycle.

[0040] Gait cycle setting, for the single-foot cycle, the real-time monitoring system uses the midpoint of the support state as the starting point, and the acceleration is zero at this point in time. At the beginning of a gait cycle, the integral data and the lower limb end position data are recorded simultaneously. The integral is calculated from the starting point of the cycle until the sole of the foot is completely off the ground, and then the gait cycle has not yet ended. During this period of time, the integral result is calculated; when the gait cycle ends, the maximum distance of the lower limb end plus the converted integral result can be considered as the final stride length. For the double-foot cycle, the cycle starting point of both feet can be set as the starting point of the double-foot cycle, and the end point of the other foot cycle can be set as the end point.

[0041] Determination of integral period: according to Figure 4 The angular velocity image can be used to find the integration period used to infer the foot movement, starting from the black point (midpoint of the support state) and ending at the red asterisk (the sole of the foot leaves the ground). By integrating, the inertial sensor motion trajectory can be obtained. Figure 5 This is the integration period set for the process of climbing stairs, which is the same as the walking process.

[0042] Integration error removal: The motion trajectory generated by the acceleration integration has the problem of error accumulation. As the integration length increases, the error continues to accumulate. To avoid the error accumulation problem, the zero correction method is used to select a specific time period in each gait cycle for approximate conversion according to the angular velocity, and recalculate in the same way in the next gait cycle to avoid the accumulation of errors in previous gait cycles. In the selection of the integral of the gait cycle, in order to avoid the cumulative error caused by the initial velocity calculation, the initial point of the total integration time period is advanced, and the integration starts from the time point when the velocity approaches zero, thereby avoiding dependence on the initial velocity and thus avoiding cumulative errors.

[0043] like Figure 3 As shown, the inertial sensor motion trajectory can be approximately inferred from the ankle joint motion trajectory (i.e., foot motion) based on the ankle geometry. Conversion relationship between sensor displacement and lower limb end joint displacement: Assuming that the ankle joint rotation angle is negligible, the position of the sensor and the toe joint (rotation axis) can be obtained based on the body size and exoskeleton size, so that the sensor displacement and lower limb end displacement become a proportional arc length relationship with the same angle but different radius, so the approximate displacement of the lower limb end can be inferred based on the body size.

[0044] Finally, by combining the leg end motion trajectory inferred by the connecting rod structure with the approximate foot motion trajectory, the movement distance of the lower limbs in a gait cycle, i.e. the stride, can be fully estimated. The speed of this gait cycle can be inferred based on the duration of the gait cycle. The same principle can be applied to the process of climbing stairs, and data such as step height and stair slope can be obtained.

[0045] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

[0046] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. A knee exoskeleton motion monitoring method based on inertial sensors, Features Here are the steps: (1) Install the inertial sensor on the outside of the leg exoskeleton in the specified installation direction; (2) Using an inclinometer, measure the angle between the thigh and calf of the leg exoskeleton, measure the size of the wearer's lower limbs, calculate the real-time motion position of the wearer's lower limbs on the sagittal plane, generate the motion trajectory of the wearer's joints and leg ends, that is, the heel excluding the sole of the foot, and generate a connecting rod structure based on the obtained motion trajectory; (3) integrating the acceleration data collected by the exoskeleton sensor of the calf part to obtain the motion trajectory of the exoskeleton of the calf part, and calculating the motion trajectory of the sole of the foot according to the obtained motion trajectory; (4) merging the motion trajectories obtained in step (2) and step (3) to obtain the overall motion trajectory of the lower limb; The three-axis installation directions of the inertial sensor are: forward direction, direction perpendicular to the sagittal plane, and direction perpendicular to the ground; The gait cycles of the joints, the end motion trajectories of the legs, and the sole motion trajectories are the same. In a single step gait cycle: According to the angular velocity of the axis of the inertial sensor perpendicular to the sagittal plane, the segmentation time point of the gait cycle of the leg exoskeleton is determined. During the single-leg movement of the exoskeleton, the midpoint of the support state is taken as the starting point of the gait cycle. At the beginning of the gait cycle, the acceleration data integration data and the lower limb end position data are recorded simultaneously; After the gait cycle is completed, the final stride length is obtained based on the maximum distance of the lower limb end and the calculated integral result; The integral data is calculated and recorded from the start of the gait cycle until the sole of the foot leaves the ground safely; before the end of the gait cycle, the integral result is calculated based on the recorded integral data; The integral data is obtained by calculating the integral of acceleration, and the motion trajectory of a specific time period is selected in each gait cycle through a zero correction method, and an approximate conversion is performed according to the angular velocity of the sensor perpendicular to the sagittal plane, and then recalculated in the next gait cycle to avoid the accumulation of gait cycle errors; The knee exoskeleton is composed of four 9-axis inertial sensors, namely three-axis acceleration, three-axis angular velocity, and three-axis angle. When walking with the exoskeleton, the inertial sensor is installed on the outside of the leg, and a real-time lower limb bar graph of human walking is constructed using a sagittal plane angle meter through real-time collection of 9-axis data. The gait state is determined by the angular velocity, and the displacement caused by the foot movement is calculated by the acceleration integral. For the two-foot cycle, by setting, either of the two-foot cycle starting points is used as the two-foot cycle starting point, and the other foot cycle end point is used as the end point; Integration period determination: The integration period of the foot movement is inferred based on the angular velocity image, starting from the midpoint of the support state and ending when the sole of the foot leaves the ground. The motion trajectory of the inertial sensor can be obtained through integration; According to the geometric relationship of the ankle, the motion trajectory of the ankle joint is approximately calculated from the motion trajectory of the inertial sensor. The motion trajectory of the leg end inferred by the connecting rod structure is combined with the approximate motion trajectory of the foot to fully estimate the movement distance of the lower limb in a gait cycle, that is, the stride length. The pace of this gait cycle is inferred based on the duration of the gait cycle. At the same time, it is used to obtain the step height and stair slope data during the process of climbing stairs.

Images