Detection system, walking training system, detection method and storage medium

Abstract

The present disclosure provides a detection system, a walking practice system, a detection method, and a storage medium. The detection system includes an acquisition unit, a calculation unit, a determination unit, and a correction unit. The acquisition unit acquires measurement information from a load distribution sensor, which detects the distribution of load received from the sole of a subject's foot. The calculation unit calculates a total load value of a sole area based on the measurement information, wherein the sole area corresponds to the position of the sole of one leg of the subject. The determination unit determines the action state of the leg based on the total load value. In response to the determination that the action state is a first action state, the correction unit starts to offset the total load value using an offset filter, wherein the offset filter reduces the offset as time passes. In the first action state, the total load value tends to increase and is equal to or greater than a preset determination value.

Description

Technical Field

[0001] The invention relates to a detection system, a walking training system, a detection method and a storage medium. Background Art

[0002] A walking training system has been developed for rehabilitation patients to practice walking movements. In this system, a load distribution sensor installed in a treadmill measures the load distribution of the exerciser. For example, WO2006 / 106714 discloses a pressure distribution detection device comprising two annular electrode groups, an elastic body on the annular electrode groups, and a conductive material on the elastic body. The walking training system measures the exerciser's walking state based on the load values obtained from the measurements and assists in the movement of the exerciser's joints. Summary of the Invention

[0003] For example, if the load value exceeds the control threshold and then falls below it again, the walking training system detects that the user has begun to reduce their pedaling force and assists in stretching their joints. However, when the pressure distribution detection device described in WO2006 / 106714 is used in a walking training system, stress gradually propagates through the elastic body, creating a discrepancy between the output of the load distribution sensor and the actual applied load. This can lead to the problem of falsely detecting that the user has begun to reduce their pedaling force.

[0004] The present invention has been made to solve this problem, and provides a detection system, a walking training system, a detection method, and a storage medium, which improve the detection accuracy of the motion state of the legs.

[0005] The detection system according to the solution of the present invention includes an acquisition unit, a calculation unit, a determination unit, and a correction unit. The acquisition unit acquires measurement information from a load distribution sensor that detects the distribution of load received by a subject's foot sole. The calculation unit calculates a total load value for a sole area based on the measurement information, the sole area corresponding to the position of the sole of one of the subject's legs. The determination unit determines the movement state of the leg based on the total load value. In response to determining that the movement state is a first movement state, the correction unit begins to shift the total load value using an offset filter, wherein the offset filter reduces the offset over time. The first movement state is a state in which the total load value tends to increase and is equal to or greater than a preset determination value. Thus, the detection system can improve the detection accuracy of the leg's movement state. Furthermore, if the total load value after the offset tends to decrease and is less than the determination value, the determination unit can determine that the movement state is a second movement state. This can avoid erroneous detection that the exerciser has begun to reduce pedaling force.

[0006] The correction unit may generate the offset filter based on the output characteristics of the load distribution sensor with respect to the input load and the properties of the object. Thus, the output characteristics of the load distribution sensor corresponding to the input load pattern estimated from the properties of the object can be reflected in the offset.

[0007] In particular, the correction unit may generate the offset filter based on the output characteristic of the load distribution sensor with respect to the input load and the weight of the subject. By estimating the input load from the weight value, the offset filter can be easily and accurately generated.

[0008] Furthermore, when the movement state is determined to be the first movement state, the correction unit may generate the offset filter based on the state of the subject's sole when the sole begins to touch the ground. This allows the correction unit to perform offset correction appropriate for the exerciser's walking state.

[0009] When the area of the sole region is equal to or greater than a predetermined area threshold, the determination unit may determine that the action state is the first action state.

[0010] A walking training system according to an embodiment of the present invention includes: a control device configured to control the extension of a leg robot worn on at least one leg of a subject based on the motion state of the subject's leg; a load distribution sensor configured to detect the distribution of load received from the sole of the subject's foot; and a detection device. The detection device includes an acquisition unit, a calculation unit, a determination unit, and a correction unit. The acquisition unit acquires measurement information from the load distribution sensor. The calculation unit calculates a total load value for a sole region corresponding to the position of the sole of one of the subject's legs based on the measurement information. The determination unit determines the motion state of the leg based on the total load value. The correction unit, in response to a determination that the motion state is a first motion state, uses an offset filter to begin offsetting the total load value, the offset filter reducing the offset over time. The first motion state is a state in which the total load value tends to increase and is equal to or greater than a preset determination value. Thus, the walking training system can improve the accuracy of detecting the motion state of the leg. Furthermore, if the total load value after the offset tends to decrease and is less than the determination value, the determination unit can determine that the motion state is a second motion state. This can avoid erroneous detection that the exerciser has started to reduce the pedaling force.

[0011] The control device may control the extension of the legged robot in response to the detection of the second action state.

[0012] The detection method according to the solution of the present invention includes: obtaining measurement information from a load distribution sensor that detects the distribution of load received by a subject's foot sole; calculating a total load value for a sole area based on the measurement information, the sole area corresponding to the position of the sole of one of the subject's legs; determining the movement state of the leg based on the total load value; and, in response to determining that the movement state is a first movement state, starting to offset the total load value using an offset filter, the offset filter reducing the offset over time, the first movement state being a state in which the total load value tends to increase and is equal to or greater than a preset determination value. Thus, the detection accuracy of the movement state of the leg can be improved.

[0013] A storage medium storing a program according to the solution of the present invention causes a computer to execute a detection method. The detection method includes: a step of acquiring measurement information from a load distribution sensor, the load distribution sensor detecting the distribution of load received from the sole of a subject's foot; a step of calculating a total load value for a sole area based on the measurement information, the sole area corresponding to the position of the sole of one of the subject's legs; a step of determining the motion state of the one leg based on the total load value; and a step of starting to offset the total load value using an offset filter in response to a determination that the motion state is a first motion state, the offset filter reducing the offset over time, the first motion state being a state in which the total load value tends to increase and is equal to or greater than a preset determination value. Thus, the detection accuracy of the motion state of the leg can be improved.

[0014] The present invention can provide a detection system, a walking training system, a detection method and a storage medium, which improve the detection accuracy of the movement state of the legs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described hereinafter with reference to the accompanying drawings, wherein like reference numerals represent like elements, and wherein:

[0016] Figure 1 is a schematic perspective view of a walking training system according to an embodiment;

[0017] Figure 2 A schematic perspective view showing an exemplary configuration of a walking training device;

[0018] Figure 3 shows a side view and a top view of a treadmill according to an embodiment;

[0019] Figure 4 is a diagram showing an example of output characteristics of a load distribution sensor with respect to an input load;

[0020] Figure 5 is a diagram showing an example of output characteristics of a load distribution sensor when receiving a load from the sole of one leg of an exerciser during walking;

[0021] Figure 6 is a diagram for describing an example of an offset according to an embodiment;

[0022] Figure 7 is a block diagram showing a schematic configuration of a detection device according to an embodiment;

[0023] Figure 8 is a flow chart showing the steps of a detection method according to an embodiment;

[0024] Figure 9 is a diagram showing an example of an offset filter according to an embodiment;

[0025] Figure 10 is a diagram showing another example of an offset filter according to an embodiment;

[0026] Figure 11 is a diagram for describing an estimation process of a foot sole area according to an embodiment; and

[0027] Figure 12 is a schematic configuration diagram of a computer serving as a detection device and a system control unit according to the embodiment. DETAILED DESCRIPTION

[0028] The present invention will be described below using embodiments. The present invention according to the claims is not limited to the embodiments described below. Furthermore, not all of the configurations described in the embodiments are essential for solving the technical problems. For clarity, some omissions and simplifications may be made in the following description and accompanying drawings, as appropriate.

[0029] Figure 1 FIG1 is a schematic perspective view of a walking training system 1 according to an embodiment. Walking training system 1 is an example of a system to which the detection device (also referred to as a detection system) according to an embodiment can be applied. Walking training system 1 is a system for a person 900, who is a hemiplegic patient suffering from monoplegia, to perform walking training. Person 900 is also referred to as a subject. References to the up-down, left-right, and front-back directions in the following description are based on the orientation of person 900.

[0030] The walking training system 1 mainly includes a control panel 133 attached to a frame 130 forming an overall skeleton, a treadmill 131 for the exerciser 900 to walk on, and a walking assistance device 120 worn on at least one leg of the exerciser 900. In an embodiment, the at least one leg is the affected leg that is the paralyzed side of the exerciser 900.

[0031] Frame 130 is installed upright on a treadmill 131 mounted on a floor surface. Treadmill 131 rotates an endless belt 132 via a motor (not shown). This causes belt 132 to travel along a track. Treadmill 131 facilitates walking for the exerciser 900. During walking exercises, the exerciser 900 steps onto belt 132 and practices walking on the walking surface formed on it.

[0032] The frame 130 supports a control panel 133, an exercise monitor 138, and a voice output unit 139. The control panel 133 includes a detection device 100 and a system control unit 200. The detection device 100 is a computer device that detects the movement state of the legs of the exerciser 900 walking on the walking surface based on the measurement results of the sensors. The system control unit 200, also known as a control device, is a computer device that controls the sensors and motors. For example, the system control unit 200 controls the extension of the walking assistance device 120 based on the movement state of the legs of the exerciser 900 detected by the detection device 100.

[0033] The exercise monitor 138 is a display device that presents information related to exercise and measurement to the exerciser 900. For example, the exercise monitor 138 is a liquid crystal panel. The exercise monitor 138 is installed so that the exerciser 900 can visually recognize the exercise monitor 138 while walking on the belt 132 of the treadmill 131.

[0034] The voice output unit 139 outputs information related to exercise and measurement by voice and informs the exerciser 900. For example, the voice output unit 139 is a speaker. The voice output unit 139 is installed in a position so that the exerciser 900 can hear the voice while walking on the belt 132 of the treadmill 131.

[0035] Furthermore, the frame 130 supports a front pulling unit 135 near the upper front portion of the head of the practitioner 900, supports the harness pulling unit 112 near the upper portion of the head of the practitioner 900, and supports a rear pulling unit 137 near the upper rear portion of the head of the practitioner 900. Furthermore, the frame 130 may include a handrail 130a to be grasped by the practitioner 900.

[0036] Camera 140 is a front camera unit that captures images of exerciser 900 at a viewing angle that enables recognition of exerciser 900's gait from the front. Alternatively, camera 140 may include a side camera unit that captures images of exerciser 900 at a viewing angle that enables recognition of exerciser 900's gait from the side. In this embodiment, camera 140 includes a lens and an imaging element. The lens has a viewing angle that enables capture of the entire body of exerciser 900, including the head, while standing on belt 132. For example, the imaging element is a CMOS image sensor that converts the optical image formed on the imaging surface into an image signal. Camera 140 is mounted near exercise monitor 138 so as to face exerciser 900. If camera 140 includes a side camera unit, the side camera unit may be mounted on handrail 130a to capture exerciser 900 from the side.

[0037] One end of the front cord 134 is connected to the take-up mechanism of the front traction unit 135, and the other end is connected to the walking assistance device 120. The take-up mechanism of the front traction unit 135 turns on and off a motor (not shown) based on a command from the system control unit 200, thereby winding or unwinding the front cord 134 in accordance with the movement of the affected leg. Similarly, one end of the rear cord 136 is connected to the take-up mechanism of the rear traction unit 137, and the other end is connected to the walking assistance device 120. The take-up mechanism of the rear traction unit 137 turns on and off a motor (not shown) based on a command from the system control unit 200, thereby winding or unwinding the rear cord 136 in accordance with the movement of the affected leg. Through this coordinated action of the front traction unit 135 and the rear traction unit 137, the load from the walking assistance device 120 is offset, preventing the load from becoming a burden on the affected leg, and further assisting the movement of the affected leg according to the set level.

[0038] For a practitioner suffering from severe paralysis, the operator 910, who is the practitioner's assistant, sets the assistance level to a high level. The operator 910 is a physical therapist or doctor who has the authority to select, modify, and add setting items of the walking training system 1. When the assistance level is set to a high level, the front traction unit 135 reels the front wire 134 with a relatively large force at the same time as the movement of the affected leg begins. When the exercise is in progress and assistance is not needed, the operator sets the assistance level to the lowest level. When the assistance level is set to the lowest level, the front traction unit 135 reels the front wire 134 with a force that allows the walking assistance device 120 to offset its own weight at the same time as the movement of the affected leg begins.

[0039] The walking training system 1 includes a safety device having a safety harness 110, a harness line 111, and a harness pulling unit 112 as its main components. The safety harness 110 is a belt that wraps around the abdomen of the exerciser 900 and is secured to the waist, for example, using hook-and-loop fasteners. The harness line 111 is a wire with one end connected to the safety harness 110 and the other end connected to the take-up mechanism of the harness pulling unit 112. The take-up mechanism of the harness pulling unit 112 turns on and off a motor (not shown) to wind or unwind the harness line 111. With this configuration, if the exerciser 900 significantly loses their posture, the safety device winds up the harness line 111 in response to a command from the system control unit 200 that detects movement, thereby supporting the exerciser's upper body through the safety harness 110.

[0040] The management monitor 141 is a display device attached to the frame 130 and monitored and operated by the operator 910. The management monitor 141 is, for example, a liquid crystal panel, and a touch panel is superimposed on the surface of the management monitor 141 as an example of the input unit 142. The management monitor 141 displays various menu items related to exercise and measurement settings, various parameter values during exercise and measurement, and measurement results during exercise. Instead of a touch panel, the input unit 142 may be a keyboard, etc. Furthermore, the operator 910 uses the input unit 142 to select, modify, or add setting items. Furthermore, the management monitor 141 is mounted in a position such that the exerciser 900 cannot visually recognize the display on the management monitor 141 in the exercise trial position on the treadmill 131. The support unit supporting the management monitor 141 may include a rotation mechanism that inverts the display surface. In this case, the operator 910 can intentionally make the exerciser 900 see the display screen.

[0041] The walking assist device 120 is worn on the affected leg of the exerciser 900 and assists the exerciser 900 in walking by reducing the load of extension and flexion at the knee joint of the affected leg. The walking assist device 120 transmits data related to leg movement and obtained through walking exercises to the system control unit 200 and drives the joint portion according to instructions from the system control unit 200. The walking assist device 120 can be connected to the hip joint (including the connecting member of the rotating portion) attached to the safety equipment 110, which is part of the fall prevention harness system, via a wire or the like.

[0042] Figure 2 1 is a schematic perspective view showing an exemplary configuration of a walking assist device 120. The walking assist device 120 mainly includes a control unit 121 and a plurality of frames that support a portion of an affected leg. The walking assist device 120 is also referred to as a leg robot.

[0043] The control unit 121 includes an assist control unit 220 for controlling the walking assist device 120 and includes a motor (not shown) that generates a driving force for assisting the extension and flexion movements of the knee joint. The frame that supports the various parts of the affected leg includes a thigh frame 122 and a calf frame 123 pivotally coupled to the thigh frame 122. Furthermore, the frame includes a flat foot frame 124 pivotally coupled to the calf frame 123, a front engaging frame 127 for engaging the front cord 134, and a rear engaging frame 128 for engaging the rear cord 136.

[0044] The thigh frame 122 and the calf frame 123 pivot relative to each other around the illustrated hinge axis Ha. The motor of the control unit 121 rotates according to a command from the auxiliary control unit 220, and applies a force to cause the thigh frame 122 and the calf frame 123 to open relative to each other around the hinge axis Ha, or applies a force to cause the thigh frame 122 and the calf frame 123 to close relative to each other around the hinge axis Ha. The angle sensor 223 included in the control unit 121 is, for example, a rotary encoder, and detects the angle between the thigh frame 122 and the calf frame 123 around the hinge axis Ha. The calf frame 123 and the flat foot frame 124 pivot relative to each other around the illustrated hinge axis Hb. The angle range of the relative pivoting is pre-adjusted by the adjustment mechanism 126.

[0045] The front engaging frame 127 is provided so as to extend in the left-right direction on the front side of the thigh and connect its ends to the thigh frame 122. Furthermore, an engaging hook 127a for engaging the front thread 134 is provided near the center in the left-right direction on the front engaging frame 127. The rear engaging frame 128 is provided so as to extend in the left-right direction on the rear side of the calf and connect its ends to the calf frame 123, both of which extend in the vertical direction. Furthermore, an engaging hook 128a for engaging the rear thread 136 is provided near the center in the left-right direction on the rear engaging frame 128.

[0046] The thigh frame 122 includes a thigh belt 129. The thigh belt 129 is a belt integrally formed with the thigh frame and secures the thigh frame 122 to the thigh of the affected leg while being wrapped around the thigh. This prevents the entire walking assistance device 120 from deviating from the leg of the exerciser 900.

[0047] Figure 3 The treadmill 131 includes at least an endless belt 132 , pulleys 151 , and a motor (not shown).

[0048] Furthermore, load distribution sensor 150 is arranged inside belt 132, that is, on the opposite side of the surface of belt 132 on which exerciser 900 rides. Load distribution sensor 150 is fixed to treadmill 131 so as not to move due to movement of belt 132.

[0049] The load distribution sensor 150 is a load distribution sensor sheet with multiple pressure detection points. These pressure detection points are arranged in a matrix parallel to the walking surface W (the placement surface) that supports the sole of the foot of the practitioner 900 in a standing position. Furthermore, the load distribution sensor 150 is positioned in the center of the walking surface W in a left-right direction perpendicular to the forward-backward walking direction. The forward-backward walking direction is parallel to the direction of travel of the belt 132. The load distribution sensor 150 uses the output values of these multiple pressure detection points to detect the magnitude and distribution of the vertical load received by the sole of the practitioner 900's foot. Thus, through the belt 132, the load distribution sensor 150 detects the position of the ground contact area (sole area) SL of the sole of the practitioner 900 in a standing position, as well as the load distribution received by the sole of the practitioner 900's foot. The position of the sole area SL is also referred to as the practitioner 900's standing position or stepping position.

[0050] Load distribution sensor 150 is connected to detection device 100. Detection device 100 obtains load distribution information from load distribution sensor 150 as measurement information and measures the motion state of the leg of exerciser 900 based on the load distribution information. For example, the motion state of the leg may be the state of pedaling, the state of maximum pedaling, or the state of force beginning to weaken. Detection device 100 is connected to system control unit 200 via a wire or wirelessly and outputs the measured motion state to system control unit 200.

[0051] The system control unit 200 controls various drive units based on the leg motion state acquired from the detection device 100. For example, the system control unit 200 is connected via wires or wirelessly to the treadmill drive unit 211, the traction drive unit 214, the harness drive unit 215, and the assistance control unit 220 of the walking assistance device 120. The system control unit 200 sends drive signals to the treadmill drive unit 211, the traction drive unit 214, and the harness drive unit 215, and sends control signals to the assistance control unit 220.

[0052] The treadmill drive unit 211 includes the aforementioned motor that rotates the belt 132 of the treadmill 131, as well as a drive circuit for the motor. The system control unit 200 controls the rotation of the belt 132 by sending a drive signal to the treadmill drive unit 211. For example, the system control unit 200 adjusts the rotation speed of the belt 132 based on the walking speed set by the operator 910. Alternatively, the system control unit 200 adjusts the rotation speed of the belt 132 based on the leg movement status of the exerciser 900 output from the detection device 100.

[0053] The pulling drive unit 214 includes a motor and a drive circuit for pulling the front cable 134, located in the front pulling unit 135. It also includes a motor and a drive circuit for pulling the rear cable 136, located in the rear pulling unit 137. The system control unit 200 controls the winding of the front cable 134 and the winding of the rear cable 136 by sending drive signals to the pulling drive unit 214. In addition to the winding action, the system control unit 200 controls the tension of each cable by controlling the driving torque of the motors. Furthermore, for example, the system control unit 200 assists the movement of the affected leg by identifying the moment when the affected leg switches from the standing position to the idle position based on the movement state of the exerciser's leg output from the detection device 100, and increasing or decreasing the tension of each cable in sync with this moment.

[0054] The harness drive unit 215 includes a motor, which is provided in the harness pulling unit 112 and pulls the harness wire 111, and a drive circuit for the motor. The system control unit 200 controls the winding of the harness wire 111 and the tension of the harness wire 111 by sending a drive signal to the harness drive unit 215. For example, if the trainee 900 is predicted to fall, the system control unit 200 prevents the trainee 900 from falling by winding the harness wire 111 a certain amount.

[0055] The assist control unit 220 is, for example, a microprocessor unit (MPU), and controls the walking assist device 120 by executing a control program given from the system control unit 200. Furthermore, the assist control unit 220 notifies the system control unit 200 of the state of the walking assist device 120. Furthermore, the assist control unit 220 controls the walking assist device 120, for example, starting and stopping the walking assist device 120, in response to an instruction from the system control unit 200.

[0056] The auxiliary control unit 220 sends a drive signal to a joint drive unit including a motor of the control unit 121 and a drive circuit for the motor, and thereby applies a force so that the thigh frame 122 and the calf frame 123 are relatively opened around the hinge axis Ha, or applies a force so that the thigh frame 122 and the calf frame 123 are relatively closed around the hinge axis Ha. Through this action, the extension and flexion movements of the knee are assisted, and soft bending is prevented. The auxiliary control unit 220 receives a detection signal from an angle sensor (not shown) that detects the angle between the thigh frame 122 and the calf frame 123 around the hinge axis Ha, and calculates the opening angle of the knee joint.

[0057] In load distribution sensor 150, an elastic sheet 150c is inserted between two facing electrode sheets 150a and 150b. Elastic sheet 150c is an elastic member made of silicone sponge, polyurethane foam, or the like. Due to its viscosity, elastic sheet 150c slowly transmits stress and gradually deforms when an external force is applied.

[0058] Figure 4 This figure illustrates an example of the output characteristics of load distribution sensor 150 relative to an input load. This figure shows the temporal changes in the output value of load distribution sensor 150 when an external force (input load) having a constant load value p1 (kPa) is continuously applied to load distribution sensor 150. As shown in the figure, at the time the external force begins to be applied, the output value of load distribution sensor 150 is lower than the actual load value of the applied external force. This is because the presence of elastic sheet 150c included in load distribution sensor 150 causes stress to be transmitted through load distribution sensor 150 with a delay, resulting in a delay in the output of load distribution sensor 150. However, the output value of load distribution sensor 150 gradually approaches the load value p1 of the applied external force over time. In other words, the difference between the output value of load distribution sensor 150 and the load value p1 of the external force decreases over time. Consequently, after a predetermined period of time has passed since the start of the external force application, the output value of load distribution sensor 150 stabilizes at a value corresponding to the applied load value p1.

[0059] Figure 5 This figure shows an example of the output characteristics of load distribution sensor 150 when receiving a load from the sole of one leg of exerciser 900 during walking. In the embodiment, the one leg is the affected leg. In the figure, the ordinate indicates the load value (kPa) received from the sole of the leg during stance, while the abscissa indicates time (s).

[0060] The dotted line indicates the load value ("actual load") actually applied to the load distribution sensor 150 by the sole of the leg of the practitioner 900 during the leg stance period from landing to leaving. In the "actual load," the load value gradually increases from the start of pedaling, reaching a maximum at a certain point in time. At this point in time, the entire body weight is supported by the leg during the leg stance period, and therefore the maximum load value is almost equal to the body weight of the practitioner 900. Thereafter, in the later stage of the leg stance period, the practitioner 900 begins to gradually reduce the force, and therefore the load value gradually decreases. Subsequently, the practitioner 900 lifts the leg off the ground, resulting in a transition to the leg idle period.

[0061] The solid line indicates the load value received from the sole of the leg of exerciser 900 during leg stance and calculated based on the output of load distribution sensor 150 ("sensor output"). The "sensor output" is obtained by extracting the load value received from the sole of the leg during leg stance from the output of load distribution sensor 150. Specifically, the "sensor output" is obtained by calculating the sum of the load values output from the pressure detection points within the sole area SL of the leg during leg stance.

[0062] The "sensor output" increases during the start of the leg stance, and when the "sensor output" exceeds the first determination value (point P1), the detection device 100 detects the first action state. The first action state is the state of starting to step on. Subsequently, when the "sensor output" reaches the maximum (point P M ), the detection device 100 detects the maximum state. The maximum state is the state of maximum pedaling. In addition, the second judgment value is a value greater than the first judgment value and less than the weight value of the exerciser 900. At point P M Afterward, the "sensor output" gradually decreases, and when it falls below the second determination value (point P2), the detection device 100 detects the second motion state. The second motion state is a state in which the force begins to weaken. In response to detecting the second motion state, the system control unit 200 sends a control signal to the assistance control unit 220, and controls the extension of the walking assistance device 120. Alternatively, in response to detecting the second motion state, the system control unit 200 sends a drive signal to the traction drive unit 214, and assists the movement of the affected leg.

[0063] Due to the aforementioned output characteristics of the load distribution sensor 150, the value of the "sensor output" is lower than the actual applied load value until the output stabilizes. Consequently, there is a gap between the "sensor output" and the "actual load." This leads to erroneous detection of the second operating state. To address this issue, in an embodiment, the detection device 100 offsets the "sensor output" by a predetermined amount upon detecting the first operating state. Hereinafter, "offset" is sometimes referred to as "correction" or "performing offset correction."

[0064] Figure 6: is a diagram for describing an example of an offset according to an embodiment. For example, the detection device 100 offsets the "sensor output" by adding an offset (dashed line) that decreases monotonically with the passage of time to the "sensor output" (thin solid line). In the accompanying drawings, the "sensor output after offset" is indicated by a thick solid line. The method for determining the offset will be described later. Specifically, in response to detecting the first action state, the detection device 100 starts the offset of the "sensor output" (point P1'). As a result, the "sensor output after offset" becomes closer to the "actual load" than the "sensor output (before offset)". Therefore, the error in the detection moment of the second action state (point P2') of the "sensor output after offset" is smaller than the error in the detection moment of the second action state (point P2) of the "sensor output (before offset)". Therefore, erroneous detection of the second action state can be avoided, and the detection accuracy can be improved.

[0065] In the embodiment, the offset becomes zero after a predetermined time has passed, and therefore, the detection device 100 does not need to terminate (end) the offset after the offset starts. However, the detection device 100 may terminate the offset when the offset becomes equal to or less than a predetermined value. The predetermined value may be zero or a value greater than zero.

[0066] Figure 7 1 is a block diagram showing a schematic configuration of a detection apparatus 100 according to an embodiment. The detection apparatus 100 includes an acquisition unit 101, a calculation unit 102, a determination unit 103, a correction unit 104, an output unit 105, and a storage unit 106. The components of the detection apparatus 100 are connected to each other.

[0067] The acquisition unit 101 acquires load distribution information as measurement information from the load distribution sensor 150. For example, the acquisition unit 101 is connected to the load distribution sensor 150 and acquires the load distribution information from the load distribution sensor 150. Then, the acquisition unit 101 provides the load distribution information of the load distribution sensor 150 to the calculation unit 102.

[0068] Furthermore, acquisition unit 101 is connected to input unit 142 and acquires basic information used to generate an offset filter. The basic information is input from input unit 142. An offset filter is a filter used to determine the offset amount to be applied at a predetermined time point. In one embodiment, the offset filter has a characteristic in which the offset amount decreases over time. The basic information used to generate the offset filter includes at least information indicating the output characteristics of load distribution sensor 150 relative to the input load. For example, the information indicating the output characteristics may include a time function of the output value relative to the input load. Furthermore, the information indicating the output characteristics may include the elastic modulus, viscosity coefficient, and thickness of elastic sheet 150c included in load distribution sensor 150. Furthermore, the basic information used to generate the offset filter includes attribute information about exerciser 900, such as exerciser 900's weight. Additionally or alternatively, the basic information may include other attribute information such as exerciser 900's gender, age, foot length, and rehabilitation stage. Acquisition unit 101 provides the basic information received from input unit 142 to correction unit 104.

[0069] Based on the load distribution information, calculation unit 102 calculates the total load value for the sole area SL corresponding to the position of the sole of one leg of exerciser 900. Specifically, calculation unit 102 first extracts the load values of the pressure detection points within the sole area SL of one leg of exerciser 900 from the load distribution information. Then, based on the extracted load values, calculation unit 102 calculates the total load value, which is the sum of the loads in the sole area SL. The one leg is the leg being measured as the action state, also referred to as the subject leg. The subject leg may be an affected leg. Calculation unit 102 provides information regarding the calculated total load value to determination unit 103.

[0070] Based on at least one of the total load value calculated by calculation unit 102 and the total load value offset by correction unit 104 (described later), determination unit 103 detects various motion states of the subject leg of exerciser 900. For example, if the total load value calculated by calculation unit 102 tends to increase and is equal to or greater than a first determination value, determination unit 103 determines that the motion state of the subject leg is the first motion state. The total load value can be determined to be increasing when the change in the total load value based on the output of load distribution sensor 150 at consecutive measurement times has a positive value, or when the total load values based on the output of load distribution sensor 150 within a predetermined time period have a positive correlation. Alternatively, determination unit 103 can determine that the motion state of the subject leg is the first motion state when the area of the sole region SL of the subject leg is equal to or greater than a predetermined area threshold. That is, if the area of the sole region tends to increase and is equal to or greater than the predetermined area threshold, determination unit 103 can determine that the motion state of the subject leg is the first motion state. When the amount of change in the area of the sole area SL of the subject's leg at consecutive measurement times has a positive value, or when the areas of the sole area SL of the subject's leg within a predetermined time have a positive correlation, it can be determined that the area of the sole area SL tends to increase. Needless to say, in the first action state, the total load value calculated by the calculation unit 102 is less than the second determination value.

[0071] When the total load value calculated by the calculation unit 102 reaches the maximum, the determination unit 103 determines that the movement state of the subject leg is the maximum state. For example, when the trend of the total load value calculated by the calculation unit 102 changes from an increasing trend to a decreasing trend, the determination unit 103 may determine that the total load value calculated by the calculation unit 102 has reached the maximum.

[0072] Furthermore, for example, if the total load value after the offset correction performed by correction unit 104 tends to decrease and is less than the second determination value, determination unit 103 determines that the motion state of the subject's leg is the second motion state. The total load value after the offset correction performed by correction unit 104 is equal to the offset total load when the offset correction performed by correction unit 104 is performed, and is equal to the total load value calculated by calculation unit 102 when the offset correction performed by correction unit 104 is terminated. The total load value can be determined to be decreasing when the amount of change in the total load value based on the output of load distribution sensor 150 at consecutive measurement times has a negative value, or when the total load values based on the output of load distribution sensor 150 within a predetermined time period have a negative correlation. Determination unit 103 provides the determination result (detection result) to output unit 105.

[0073] Correction unit 104 generates an offset filter based on basic information used to generate the offset filter. For example, correction unit 104 generates the offset filter based on information indicating the output characteristics of load distribution sensor 150 relative to the input load and attribute information regarding exerciser 900. The output characteristics of load distribution sensor 150 vary depending on the pattern of the input load. Therefore, with this configuration, the output characteristics of load distribution sensor 150, which depend on the pattern of the input load estimated from the attributes of exerciser 900, can be reflected in the offset. In this embodiment, correction unit 104 generates the offset filter based on information indicating the output characteristics of load distribution sensor 150 relative to the input load and the weight of exerciser 900. By estimating the input load from the weight, correction unit 104 can easily and accurately generate the offset filter.

[0074] Correction unit 104 then uses the generated offset filter to offset the total load value. Specifically, in response to determining that the action state of the subject's leg is the first action state, correction unit 104 uses the offset filter to begin offsetting the total load value calculated by calculation unit 102. This allows the total load value to be brought closer to the actual load, and thus, for example, can avoid erroneous detection of the second action state, where the force begins to weaken, thereby improving the accuracy of action state detection.

[0075] The output unit 105 outputs a control signal to the system control unit 200 based on the detection result provided by the determination unit 103. In the embodiment, when the second motion state is detected, the output unit 105 outputs a control signal indicating the detection of the second motion state to the system control unit 200. However, this is not limited to the embodiment, and when the first motion state or the maximum state is detected, the output unit 105 may output a control signal corresponding to the motion state to the system control unit 200.

[0076] The storage unit 106 is a storage medium that stores information required for processing in the detection device 100 and generated information.

[0077] Figure 8 Flowchart showing the steps of the detection method according to the embodiment: First, the acquisition unit 101 acquires basic information from the trainee 900 or the operator 910 via the input unit 142 (step S10). Then, the correction unit 104 generates an offset filter based on the basic information (step S11).

[0078] Figure 9 is a diagram illustrating an example of an offset filter according to an embodiment. Figure 9The offset filter f(t) shown is a function of time t with respect to the offset. The offset filter f(t) can be a filter in which the initial offset value is o1 (>0 kPa), the offset decreases over time, and the offset reaches zero (kPa) at a predetermined time t1. The correction unit 104 can generate the offset filter f(t) based on the basic information by determining the initial value o1, time t1, and the slope. If the offset filter is a function of time t with respect to the offset, the correction unit 104 can perform offset correction by adding the value of the offset filter at the current time point to the total load value.

[0079] Figure 10 FIG. 4 is a diagram illustrating another example of an offset filter according to an embodiment. Figure 10 The offset filter g(t) shown is a function of time t with respect to the offset coefficient. The offset filter g(t) can be a function in which the initial value of the offset coefficient is o2 (>1), the offset coefficient decreases over time, and the offset coefficient reaches 1 (kPa) at a predetermined time t2. The correction unit 104 can generate the offset filter g(t) based on the basic information by determining the initial value o2, time t2, and the slope. If the offset filter is a function of time t with respect to the offset coefficient, the correction unit 104 can perform offset correction by multiplying the total load value by the value of the offset filter.

[0080] return Figure 8 , will be described further. The detection device 100 determines whether to start measurement (step S12). For example, measurement is started when exercise is started using the walking exercise system 1, or when detection processing is started by the detection device 100 through an operation from the operator 910. The detection device 100 repeats the process shown in step S12 until it determines that measurement has started. If it is determined that measurement has started ("YES" in step S12), the detection device 100 proceeds to step S13.

[0081] The acquisition unit 101 acquires load distribution information as measurement information from the load distribution sensor 150 (step S13). The load distribution information includes information indicating load values corresponding to pressure detection points at different positions. Subsequently, the acquisition unit 101 provides the load distribution information to the calculation unit 102. Next, the calculation unit 102 estimates the sole area SL of the subject's leg based on the load distribution information (step S14) and extracts the load value in the sole area SL of the subject's leg. Subsequently, the calculation unit 102 calculates a total load value that is the sum of the load values extracted in the sole area SL of the subject's leg (step S15).

[0082] Figure 111 is a diagram for describing the estimation process of the sole area SL according to the embodiment. For example, based on the position information about the pressure detection point and the load value detected from the pressure detection point, the calculation unit 102 generates Figure 11 The load distribution diagram shown. The calculation unit 102 can generate a load distribution diagram by extracting position information about load values equal to or greater than the detection threshold from the load values detected from the pressure detection point. Subsequently, the calculation unit 102 detects the sole area SL based on the load distribution diagram. Here, it is assumed that the subject leg is the right leg. The calculation unit 102 determines whether the sole area SL is the area of the subject leg based on the position of the sole area SL relative to the central axis D1 of the load distribution sensor 150 in the left and right directions. For example, the calculation unit 102 calculates the center of gravity position of the sole area SL, and in the case where the center of gravity position is located on the right side of the central axis D1, the calculation unit 102 determines that the sole area SL is the sole area SL of the subject leg. Subsequently, the calculation unit 102 calculates the sum of the load values included in the sole area SL as the total load value. The area from which the load value is extracted is not limited to the sole area SL, and may be a predetermined area A1 containing the sole area SL.

[0083] Furthermore, the computing unit 102 can determine whether the detected sole area SL is the sole area SL of the subject leg based on a captured image generated by capturing the gait of the exerciser 900 using the camera 140 as the front camera unit and the side camera unit. For example, if the captured image shows the exerciser 900 moving his right leg forward, or if the captured image shows the exerciser 900 landing his right leg, the computing unit 102 determines that the detected sole area SL is the sole area SL of the subject leg.

[0084] The above description describes a case where computing unit 102 detects one sole area SL, i.e., a case where the sole of one leg is touching the ground. However, if computing unit 102 detects two sole areas SL, i.e., a case where the soles of both legs are touching the ground, computing unit 102 can estimate the sole area SL of the subject's leg based on the relative positions of the two sole areas SL. For example, computing unit 102 can set the right sole area SL of the two sole areas SL as the sole area SL of the subject's leg. In this case as well, computing unit 102 can estimate the sole area SL of the subject's leg based on the captured image.

[0085] Furthermore, since the exerciser 900 walks while the soles of the right leg and the left leg touch the ground alternately, the calculation unit 102 can estimate the sole area SL of the target leg in accordance with the walking cycle.

[0086] Back to Figure 8, which will be described further. In step S16, the determination unit 103 determines whether the total load value is increasing. If the correction unit 104 does not perform offset correction, that is, if the offset is zero, the total load value is the total load value calculated by the calculation unit 102. If the correction unit 104 performs offset correction, that is, if the offset is not zero, the total load value is the offset total load value. If the determination unit 103 determines that the total load value is increasing ("Yes" in step S16), the determination unit 103 determines whether the total load value has become equal to or greater than the first judgment value for the first time in the first target period (step S17). The first target period is a period in the current gait cycle during which the total load value is increasing. If the determination unit 103 determines that the total load value has not yet become equal to or greater than the first judgment value in the first target period, or if the total load value has already become equal to or greater than the first judgment value in the past in the first target period ("No" in step S17), the determination unit 103 proceeds to step S23. On the other hand, if the determination unit 103 determines that the total load value has become equal to or greater than the first determination value for the first time in the first target period ("Yes" in step S17), the determination unit 103 determines that the action state of the subject leg is the first action state (step S18). Subsequently, the correction unit 104 starts offset correction for the total load value using the offset filter (step S19), and proceeds to step S23.

[0087] If the determination unit 103 determines that the total load value has not increased ("No" in step S16), the determination unit 103 determines whether the total load value has become less than the second determination value for the first time in the second target period (step S20). The second target period is a period in the current gait cycle during which the total load value has decreased. If the determination unit 103 determines that the total load value has not yet decreased below the second determination value in the second target period, or if the total load value has already decreased below the second determination value in the past in the second target period ("No" in step S20), the determination unit 103 proceeds with the process to step S23. On the other hand, if the determination unit 103 determines that the total load value has decreased below the second determination value for the first time in the second target period ("Yes" in step S20), the determination unit 103 determines that the motion state of the subject leg is the second motion state (step S21). Subsequently, the output unit 105 outputs a control signal to the system control unit 200 indicating that the second motion state has been detected (step S22), and proceeds with the process to step S23.

[0088] In step S23, detection device 100 determines whether measurement has ended. For example, measurement ends when exercise using walking training system 1 ends, or when detection processing by detection device 100 ends due to an operation by operator 910. Detection device 100 repeats the processes shown in steps S13 to S23 until measurement is determined to be complete.

[0089] Figure 8 The flow chart illustrates the case where the detection device 100 does not terminate the offset correction. However, for example, if the detection device 100 terminates the offset correction, the detection device 100 may perform the following processing immediately before step S23 (in step S19, in step S22, or after step S17 if step S17 is "No"). If the offset amount is equal to or less than a predetermined value, the correction unit 104 of the detection device 100 may terminate the offset correction and proceed to step S23. On the other hand, if the offset amount is greater than the predetermined value, the correction unit 104 may continue to perform the offset correction and proceed to step S23.

[0090] In this manner, using the embodiment, the detection device 100 can improve the detection accuracy of the motion state of the measured leg, and in particular, can improve the detection accuracy of the moment when the force starts to weaken.

[0091] In the above description, the correction unit 104 is Figure 8In step S11, an offset filter is generated based on the basic information. However, instead of or in addition to this, the correction unit 104 may generate an offset filter during the measurement. For example, in the period between step S15 and step S16 before determining that the action state of the subject leg is the first action state, the correction unit 104 may generate an offset filter based on the state of the sole of the practitioner 900 when the sole begins to touch the ground. Subsequently, the correction unit 104 may use the generated offset filter to perform offset correction. The state of the sole of the foot when it begins to touch the ground may be a "heel-landing state" when the heel touches the ground earlier, or may be a "toe-landing state" when the toe touches the ground earlier. The state of the sole of the foot when the sole of the foot begins to touch the ground may be determined by the calculation unit 102, and information about the state of the sole of the foot when the sole of the foot begins to touch the ground may be provided from the calculation unit 102 to the correction unit 104. For example, if the calculation unit 102 determines, based on the temporal change in the area of the sole area SL and the travel speed of the treadmill 131, that the sole area SL has expanded in the backward direction of walking over time after detection, the calculation unit 102 can determine that the state of the sole of the foot at the time of the sole's initial contact is the "toe-strike state." On the other hand, if the sole area SL has expanded in the forward direction of walking over time after detection, the calculation unit 102 can determine that the state of the sole of the foot at the time of the sole's initial contact is the "heel-strike state." The offset filters corresponding to the "toe-strike state" and the "heel-strike state" can differ from each other in at least one of their initial values, slopes, and time. This allows the correction unit 104 to perform offset correction appropriate for the walking state of the exerciser 900.

[0092] Figure 12 19 is a schematic diagram of a computer used as a detection device and system control unit 200 according to an embodiment. Computer 1900 includes a processor 1000, a read-only memory (ROM) 1010, a random access memory (RAM) 1020, and an interface (IF) unit 1030 as main hardware components. Processor 1000, ROM 1010, RAM 1020, and interface unit 1030 are interconnected via a data bus or the like.

[0093] The processor 1000 has a function as a computing device that performs control processing, calculation processing, etc. The processor 1000 can be a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or a combination thereof. The ROM 1010 has a function of storing control programs, calculation programs, etc. executed by the processor 1000. The RAM 1020 has a function of temporarily storing processing data, etc. The interface unit 1030 exchanges signals with the outside via wired or wireless communication. In addition, the interface unit 1030 accepts user input data operations and displays information to the user. For example, the interface unit 1030 communicates with the load distribution sensor 150, the input unit 142, and the system control unit 200.

[0094] In the above examples, the program includes commands (or software codes) that, when read by a computer, cause the computer to perform one or more functions described in the embodiments. The program can be stored in various non-transitory computer-readable media, each of which is an example of ROM 1010, or can be stored in tangible storage media. Although not limited, examples of computer-readable media or tangible storage media include memory technologies such as random access memory (RAM), read-only memory (ROM), flash memory, solid-state drives (SSD), such as CD-ROMs, digital versatile disks (DVDs), and The program may be transmitted via a transitory computer-readable medium or communication medium. Although not limiting, examples of transitory computer-readable media or communication media include electrical propagation signals, optical propagation signals, acoustic propagation signals, and other types of propagation signals.

[0095] In the above embodiment, computer 1900 is configured by a computer system including a personal computer, a word processor, etc. However, there is no restriction on this, and computer 1900 can be configured by a server in a local area network (LAN), a main frame for computer (PC) communication, a computer system connected to the Internet, etc. In addition, function distribution can be performed between devices on the network, and computer 1900 can be configured by the entire network. Therefore, the constituent elements of the detection device can be distributed in devices different from each other.

[0096] The present invention is not limited to the above-described embodiment and can be modified appropriately without departing from the main idea. For example, in the above-described embodiment, the detection device 100 detects the motion state of the diseased leg as the target leg. However, the detection device 100 may detect the motion state of the normal leg. In addition, the detection device 100 may detect the motion state of each of the left and right legs. In this case, for each leg, the detection device 100 performs Figure 8 The processing shown in steps S13 to S23.

[0097] Furthermore, in the above embodiment, the second determination value is greater than the first determination value. However, the second determination value may be equal to the first determination value.

[0098] In addition, the exerciser 900 can wear the walking assistance device 120 on both legs and can perform exercises. Alternatively, the exerciser 900 does not need to wear the walking assistance device 120 on any leg.

Claims

1. A detection system comprising: an acquisition unit configured to acquire measurement information from a load distribution sensor that detects a distribution of a load received from a sole of a subject; a calculation unit configured to calculate a total load value of a sole area corresponding to a position of a sole of one leg of the subject based on the measurement information; a determination unit configured to determine an action state of the one leg based on the total load value; as well as a correction unit configured to, in response to a determination that the action state is a first action state, start shifting the total load value using an offset filter, the offset filter reducing an offset amount over time, the first action state being a state in which the total load value tends to increase and is equal to or greater than a preset determination value; The correction unit generates the offset filter based on an output characteristic of the load distribution sensor with respect to an input load and the weight of the subject, wherein the output characteristic is that a difference between an output value of the load distribution sensor and the input load becomes smaller with time.

2. The detection system according to claim 1, wherein: In a case where the total load value after the shift tends to decrease and is smaller than the determination value, the determination unit determines that the operating state is a second operating state.

3. The detection system according to claim 1, wherein: When it is determined that the motion state is the first motion state, the correction unit generates the offset filter based on the state of the sole of the subject when the sole begins to land.

4. The detection system according to any one of claims 1 to 3, wherein: When the area of the sole region is equal to or greater than a predetermined area threshold, the determination unit determines that the action state is the first action state.

5. A walking training system comprising: a control device configured to control extension of a leg robot worn on at least one leg of the subject based on a motion state of the subject's legs; a load distribution sensor configured to detect a distribution of a load received from a sole of the subject; as well as Detection device, wherein The detection device comprises: an acquiring unit configured to acquire measurement information from the load distribution sensor; a calculation unit configured to calculate a total load value of a sole area corresponding to a position of a sole of one leg of the subject based on the measurement information; a determination unit configured to determine an action state of the one leg based on the total load value; and a correction unit configured to, in response to a determination that the action state is a first action state, start shifting the total load value using an offset filter, the offset filter reducing an offset amount over time, the first action state being a state in which the total load value tends to increase and is equal to or greater than a preset determination value; The correction unit generates the offset filter based on an output characteristic of the load distribution sensor with respect to an input load and the weight of the subject, wherein the output characteristic is that a difference between an output value of the load distribution sensor and the input load becomes smaller with time.

6. The walking training system according to claim 5, wherein: In a case where the total load value after the shift tends to decrease and is smaller than the determination value, the determination unit determines that the operating state is a second operating state.

7. The walking training system according to claim 6, wherein: The control device controls the extension of the legged robot in response to the detection of the second action state.

8. A detection method comprising: a step of acquiring measurement information from a load distribution sensor that detects a distribution of a load received from a sole of a subject's foot; a step of calculating a total load value of a sole area based on the measurement information, the sole area corresponding to a position of a sole of one leg of the subject; a step of determining the movement state of the one leg based on the total load value; as well as In response to a determination that the action state is a first action state, the step of starting to shift the total load value using a shift filter, the shift filter reducing an offset amount over time, the first action state being a state in which the total load value tends to increase and is equal to or greater than a preset determination value; The offset filter is generated based on an output characteristic of the load distribution sensor with respect to an input load and the weight of the subject, wherein the output characteristic is that a difference between an output value of the load distribution sensor and the input load becomes smaller with the passage of time.

9. A storage medium storing a program for causing a computer to execute a detection method, the detection method comprising: a step of acquiring measurement information from a load distribution sensor that detects a distribution of a load received from a sole of a subject's foot; a step of calculating a total load value of a sole area based on the measurement information, the sole area corresponding to a position of a sole of one leg of the subject; a step of determining the movement state of the one leg based on the total load value; as well as In response to a determination that the action state is a first action state, the step of starting to shift the total load value using a shift filter, the shift filter reducing an offset amount over time, the first action state being a state in which the total load value tends to increase and is equal to or greater than a preset determination value; The offset filter is generated based on an output characteristic of the load distribution sensor with respect to an input load and the weight of the subject, wherein the output characteristic is that a difference between an output value of the load distribution sensor and the input load becomes smaller with the passage of time.

Images