Rehabilitation assisting system and method for patient with unilateral space neglect

Abstract

The invention belongs to the technical field of rehabilitation assistance, and provides a rehabilitation assistance system and method for a unilateral space neglect patient, and the system comprises a head-mounted device, an upper limb device and a lower limb device; the head-mounted device comprises a head controller, and the head controller is connected with a head posture detection module, a voice module, a head vibration module and a dry electrode type electrical stimulation module. The lower limb equipment comprises a lower limb controller; the lower limb controller is connected with a lower limb posture detection module, a lower limb vibration module and a distance measurement module; the upper limb equipment comprises an upper limb controller, and the upper limb controller is connected with an upper limb posture detection module and an upper limb vibration module; according to the invention, through the head-mounted device and the limb wearable device, an electrical stimulation function and a vibration feedback function are achieved; the head-mounted equipment and the limb wearable equipment both have a distance detection function and can detect the distances of objects in front of the body and on the side surfaces of the body; and effective assistance can be provided for rehabilitation of the patient with unilateral space neglect.

Description

Technical Field

[0001] The present invention belongs to the field of rehabilitation assistance technology, and in particular relates to a rehabilitation assistance system and method for patients with unilateral spatial neglect. Background Art

[0002] Unilateral spatial neglect, also known as unilateral neglect or hemi-neglect, is a neuropsychological phenomenon that usually occurs after damage to the right hemisphere of the brain. The patient lacks attention and awareness to one side of the body (usually the left) or one side of space (usually the left). This condition is common in stroke, brain trauma or other diseases that affect brain function. The mechanism of unilateral spatial neglect is related to the brain's spatial processing function. Usually, the right hemisphere is responsible for processing spatial and global information, while the left hemisphere pays more attention to details. Damage to the right hemisphere may lead to a lack of attention to the left side of the space. This phenomenon is not limited to visual perception, but may also affect senses such as hearing and touch.

[0003] The present invention solves the problem that patients with unilateral spatial neglect are prone to neglecting the existence of the affected side in daily life, and that they may neglect the vision or space or objects on the affected side in daily life (such as walking and neglecting obstacles on the affected side and bumping into them, neglecting walls on the affected side and bumping into them, etc.), thereby causing dangerous situations. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a rehabilitation assistance system and method for patients with unilateral spatial neglect to solve the problems in the prior art. The technical solution adopted by the present invention is:

[0005] A rehabilitation assistive system for patients with unilateral spatial neglect, comprising a head-mounted device, an upper limb device, and a lower limb device;

[0006] The head-mounted device includes a head controller, a head posture detection module, a voice module, a head vibration module and a dry electrode electrical stimulation module;

[0007] The lower limb device includes a lower limb controller, a lower limb posture detection module, a lower limb vibration module and a distance measurement module;

[0008] The upper limb device includes an upper limb controller, an upper limb posture detection module and an upper limb vibration module;

[0009] The head-mounted device is used to be worn on the patient's head, the lower limb device is used to be worn on the patient's legs, and the upper limb device is used to be worn on the patient's arms;

[0010] When the distance between the affected side of the patient and the obstacle is lower than a threshold, the voice module plays an alarm audio, the head vibration module vibrates, and the lower limb vibration module and the upper limb vibration module on the corresponding side vibrate;

[0011] When the distance between the affected side of the patient and the obstacle continues to shorten after falling below a threshold, the dry electrode electrical stimulation module stimulates the head.

[0012] Furthermore, the lower limb device, the upper limb device and the head-mounted device all include a power supply and a wireless communication module.

[0013] Furthermore, two of the lower limb devices and two of the upper limb devices are provided, distributed on both sides of the patient's body.

[0014] A rehabilitation assistive method for patients with unilateral spatial neglect, comprising:

[0015] Step 1, data collection;

[0016] Step 2, data preprocessing;

[0017] Step 3, angle calculation;

[0018] Step 4: Calculate the movement amplitude: Calculate the average movement amplitude difference coefficient of the left and right hands, and use the average movement amplitude difference coefficient to reflect the movement of the left and right sides;

[0019] Step 5, movement time calculation: calculate the average movement time difference coefficient of the left and right hands, and reflect the movement situation of the left and right sides through the average movement time difference coefficient;

[0020] Step 6, reminder trigger calculation; comprehensive average movement amplitude difference coefficient and average movement time difference coefficient, identify the patient's left and right side movement status, and remind the patient.

[0021] Furthermore, the step 1 includes: acquiring acceleration data, angular velocity data, and magnetic field data from a 3-axis acceleration, a 3-axis gyroscope, and a 3-axis geomagnetic sensor of an upper limb posture detection module, respectively;

[0022] Among them, the acceleration data is: a x ,a y ,a z ; Angular velocity data is: ω x ,ω y ,ω z ; Acceleration data: m x ,m y ,m z ;

[0023] The step 2 includes:

[0024] Normalization of acceleration data:

[0025]

[0026] Where: a: acceleration vector; |a|: modulus of acceleration vector, representing the composite value of three-axis acceleration; anorm: normalized acceleration vector;

[0027] Normalization of geomagnetic data:

[0028]

[0029] Where: m: geomagnetic vector, |m|: modulus of geomagnetic vector, mnorm: normalized geomagnetic vector;

[0030] Gyroscope data calibration:

[0031] ω'=ω-b g

[0032] Where: b g =[b x b y b z ] T , which is the zero bias error vector of the gyroscope, used to calibrate the gyroscope data; ω = [ω x ω y ω z ] T is the angular velocity vector.

[0033] Furthermore, the step 3 includes:

[0034] Step 3.1, initialize the attitude quaternion:

[0035] q=[q0 q1 q2 q3] T =[1 0 0 0] T

[0036] Step 3.2, initialize the state variables and covariance matrix of the Kalman filter, including:

[0037] Step 3.2.1, state vector:

[0038] x=[φ θ ψ] T

[0039] Where φ is the roll angle, θ is the pitch angle, and ψ is the yaw angle;

[0040] Step 3.2.2, process noise covariance matrix:

[0041]

[0042] Step 3.2.3, measurement noise covariance matrix:

[0043]

[0044] Step 3.3, initialize the covariance matrix:

[0045]

[0046] Step 3.4, state prediction, includes:

[0047] Step 3.4.1, use the gyroscope data to calculate the changes in roll angle, pitch angle and yaw angle:

[0048] Δφ=ω′ x Δt, Δθ=ω′ y Δt, Δψ=ω′ z ΔtΔt is the sampling time interval, which represents the time difference between two data samples;

[0049] Prediction status:

[0050]

[0051] Where: x k|k-1 is the predicted state vector, which represents the estimated value of the state at the current moment; k-1 The state vector at the previous moment;

[0052] Forecast covariance matrix:

[0053] P k|k-1 =P k-1 +Q

[0054] P k|k-1 Updated state covariance matrix;

[0055] P k-1 The state covariance matrix at the previous moment;

[0056] Step 3.4.2, Status Update:

[0057] Correct the state using accelerometer and magnetometer measurements:

[0058] Roll and pitch angle calculation based on accelerometer:

[0059]

[0060] Yaw angle calculation based on magnetometer:

[0061]

[0062] Yaw angle:

[0063]

[0064] Update Kalman filter observations:

[0065] Measurements:

[0066]

[0067] Kalman gain:

[0068] K=P k|k-1 (P k|k-1 +R) -1

[0069] Status Update:

[0070] x k =x k|k-1 +K(zx k|k-1 )

[0071] x k is the updated state vector;

[0072] Covariance update:

[0073] P k =(IK)P k|k-1

[0074] P k predicted state covariance matrix;

[0075] I identity matrix, used to update the covariance matrix;

[0076] Step 3.5, AHRS algorithm fusion, uses quaternion update and gyroscope integration to fuse data, including:

[0077] Step 3.5.1 Quaternion update:

[0078] Quaternion differential equation:

[0079]

[0080] in:

[0081] q=[q0 q1 q2 q3] T is a quaternion used to represent three-dimensional posture, q0 is the scalar part of the quaternion, q1 q2 q3 is the vector part of the quaternion, is the quaternion multiplication symbol, used to describe the update of the quaternion;

[0082]

[0083] Use discretized integral update:

[0084]

[0085] Step 3.5.2, normalize the quaternion:

[0086]

[0087] Step 3.5.3, calculate Euler angles from quaternions:

[0088]

[0089] θ=arcsin(2(q0q2-q3q1))

[0090]

[0091] Step 3.6, real-time loop: Repeat steps 3.1 to 3.5 to calculate the pose in real time.

[0092] Furthermore, the step 4 includes:

[0093] Step 4.1: For each time point, calculate the angle change of the left and right arms:

[0094] Δθ L (t) = θ L (t)-θ L (t-1)

[0095] Δθ R (t) = θ R (t)-θ R (t-1)

[0096] Where Δθ L (t) and Δθ R (t) are the angles of the left and right hands at time t, respectively;

[0097] Step 4.2: Calculate the total angle change. Calculate the total angle change over a period of time:

[0098]

[0099] Among them, ΔΘ L is the total angle change of the left hand, ΔΘ R is the total angle change of the right hand;

[0100] Step 4.3: Calculate the mean motion amplitude difference:

[0101] Calculate the difference coefficient M of the average movement amplitude of the left and right hands error :

[0102]

[0103] Furthermore, in step 4, the average motion time difference coefficient T error Calculated using the following formula:

[0104]

[0105] Among them, T L is the left hand movement time, T R Right hand movement time.

[0106] Furthermore, in step 5, the following formula is used to identify the movement of the left and right sides of the patient's body:

[0107]

[0108] Let M t ∈[0,1] and T t ∈[0,1]

[0109] If M error >M t , ΔΘ L >Θ R , the voice module prompts: Please move your right hand;

[0110] If M error <M t , ΔΘ L <Θ R , then the voice module prompts: Please move your left hand;

[0111] If T error >T t , TL>TR, then the voice module prompts: Please use your right hand more;

[0112] If T error <T t , TL<TR, then the voice module prompts: Please use your left hand more;

[0113] If the voice module prompts for more than N cycles, M error and T error If there is still no change, the dry electrode electrical stimulation module stimulates the head, the head vibration module vibrates, and the lower limb vibration module and upper limb vibration module on the corresponding side vibrate.

[0114] The present invention has the following beneficial effects:

[0115] The present invention uses a head-mounted device and a limb wearable device. The head-mounted device has a bilateral voice playback function, a bilateral independent electrical stimulation function, and a vibration feedback function; the limb wearable device has a bilateral independent electrical stimulation function and a vibration feedback function; the head-mounted device and the limb wearable device both have a distance detection function, and can detect the distance of objects in front of and on the side of the body; they have a posture detection function, which can detect the amplitude of limb movements; and can provide effective assistance for the rehabilitation of patients with unilateral spatial neglect. BRIEF DESCRIPTION OF THE DRAWINGS

[0116] Figure 1This is a schematic structural diagram of the head-mounted device of the present invention;

[0117] Figure 2 This is a schematic structural diagram of the lower limb device of the present invention;

[0118] Figure 3 This is a schematic structural diagram of the upper limb device of the present invention. DETAILED DESCRIPTION

[0119] The following is a combination of the embodiments of the present invention Figure 1-Figure 3 , the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Unless otherwise specified, the technical means used in the embodiments are conventional means well known to those skilled in the art.

[0120] A rehabilitation assistive system for patients with unilateral spatial neglect, comprising a head-mounted device, an upper limb device, and a lower limb device;

[0121] The head-mounted device includes a head controller connected to a head posture detection module, a voice module, a head vibration module and a dry electrode electrical stimulation module;

[0122] The lower limb device includes a lower limb controller, which is connected to a lower limb posture detection module, a lower limb vibration module and a distance measurement module;

[0123] The upper limb device includes an upper limb controller connected to an upper limb posture detection module and an upper limb vibration module;

[0124] The head-mounted device is used to be worn on the patient's head, the lower limb device is used to be worn on the patient's legs, and the upper limb device is used to be worn on the patient's arms;

[0125] When the distance between the affected side of the patient and the obstacle is lower than a threshold, the voice module plays an alarm audio, the head vibration module vibrates, and the lower limb vibration module and the upper limb vibration module on the corresponding side vibrate;

[0126] When the distance between the affected side of the patient and the obstacle continues to shorten after falling below a threshold, the dry electrode electrical stimulation module stimulates the head.

[0127] Furthermore, the lower limb device, the upper limb device and the head-mounted device all include a power supply and a wireless communication module.

[0128] Furthermore, two of the lower limb devices and two of the upper limb devices are provided, distributed on both sides of the patient's body.

[0129] Among them, two dry electrode electrical stimulation modules, head vibration modules, and voice modules can be set up and distributed on the left and right sides of the patient's head. Figure 1The CH1 and CH2 represent the left and right sides.

[0130] The head-mounted device, lower limb device and upper limb device can all be worn on the patient's body using a ring-belt structure.

[0131] A rehabilitation assistive method for patients with unilateral spatial neglect, comprising:

[0132] Step 1, data collection;

[0133] Step 2, data preprocessing;

[0134] Step 3, motion amplitude calculation: calculate the average motion amplitude difference coefficient of the left and right hands, and use the average motion amplitude difference coefficient to reflect the motion situation of the left and right sides;

[0135] Step 4, movement time calculation: calculate the average movement time difference coefficient of the left and right hands, and use the average movement time difference coefficient to reflect the movement of the left and right sides;

[0136] Step 5, reminder trigger calculation; comprehensive average movement amplitude difference coefficient and average movement time difference coefficient, identify the patient's left and right side movement status, and remind the patient.

[0137] Furthermore, the step 1 includes:

[0138] Acquire acceleration data, angular velocity data, and magnetic field data from the 3-axis accelerometer, 3-axis gyroscope, and 3-axis geomagnetic sensor of the upper limb posture detection module respectively;

[0139] Among them, the acceleration data is: a x ,a y ,a z ; Angular velocity data is: ω x ,ω y ,ω z ; Acceleration data: m x ,m y ,m z ;

[0140] The step 2 includes:

[0141] Normalization of acceleration data:

[0142]

[0143] Where: a: acceleration vector; |a|: modulus of acceleration vector, representing the composite value of three-axis acceleration; anorm: normalized acceleration vector;

[0144] Normalization of geomagnetic data:

[0145]

[0146] Where: m: geomagnetic vector, |m|: modulus of geomagnetic vector, mnorm: normalized geomagnetic vector;

[0147] Gyroscope data calibration:

[0148] ω'=ω-b g

[0149] Where: b g =[b x b y b z ] T , which is the zero bias error vector of the gyroscope, used to calibrate the gyroscope data;

[0150] ω=[ω x ω y ω z ] T is the angular velocity vector.

[0151] Step 3 includes:

[0152] Step 3.1, initialize the attitude quaternion:

[0153] q=[q0 q1 q2 q3] T =[1 0 0 0] T

[0154] Step 3.2, initialize the state variables and covariance matrix of the Kalman filter, including:

[0155] Step 3.2.1, state vector:

[0156] x=[φ θ ψ] T

[0157] Where φ is the roll angle, θ is the pitch angle, and ψ is the yaw angle;

[0158] Step 3.2.2, process noise covariance matrix:

[0159]

[0160] Step 3.2.3, measurement noise covariance matrix:

[0161]

[0162] Step 3.3, initialize the covariance matrix:

[0163]

[0164] Step 3.4, state prediction, includes:

[0165] Step 3.4.1, use the gyroscope data to calculate the changes in roll angle, pitch angle and yaw angle:

[0166] Δφ=ω′ x Δt, Δθ=ω′ y Δt, Δψ=ω′ z Δt

[0167] Δt is the sampling time interval, which represents the time difference between two data samples;

[0168] Prediction status:

[0169]

[0170] Where: x k|k-1 is the predicted state vector, which represents the estimated value of the state at the current moment;

[0171] x k-1 The state vector at the previous moment;

[0172] Forecast covariance matrix:

[0173] P k|k-1 =P k-1 +Q

[0174] P k|k-1 Updated state covariance matrix;

[0175] P k-1 The state covariance matrix at the previous moment;

[0176] Step 3.4.2, Status Update:

[0177] Correct the state using accelerometer and magnetometer measurements:

[0178] Roll and pitch angle calculation based on accelerometer:

[0179]

[0180] Yaw angle calculation based on magnetometer:

[0181]

[0182] Yaw angle:

[0183]

[0184] Update Kalman filter observations:

[0185] Measurements:

[0186]

[0187] Kalman gain:

[0188] K=P k|k-1 (P k|k-1 +R) -1

[0189] Status Update:

[0190] x k =x k|k-1 +K(zx k|k-1 )

[0191] x k is the updated state vector;

[0192] Covariance update:

[0193] P k =(IK)P k|k＝1

[0194] P k predicted state covariance matrix;

[0195] I identity matrix, used to update the covariance matrix;

[0196] Step 3.5, AHRS algorithm fusion, uses quaternion update and gyroscope integration to fuse data, including:

[0197] Step 3.5.1 Quaternion update:

[0198] Quaternion differential equation:

[0199]

[0200] in:

[0201] q=[q0 q1 q2 q3] T is a quaternion used to represent three-dimensional posture, q0 is the scalar part of the quaternion, q1 q2 q3 is the vector part of the quaternion, is the quaternion multiplication symbol, used to describe the update of the quaternion;

[0202]

[0203] Use discretized integral update:

[0204]

[0205] Step 3.5.2, normalize the quaternion:

[0206]

[0207] Step 3.5.3, calculate Euler angles from quaternions:

[0208]

[0209] θ=arcsin(2(q0q2-q3q1))

[0210]

[0211] Step 3.6, real-time loop: Repeat steps 3.1 to 3.5 to calculate the pose in real time.

[0212] The step 4 comprises:

[0213] Step 4.1: For each time point, calculate the angle change of the left and right arms:

[0214] Δθ L (t) = θ L (t)-θ L (t-1)

[0215] Δθ R (t) = θ R (t)-θ R (t-1)

[0216] Where Δθ L (t) and Δθ R (t) are the angles of the left and right hands at time t, respectively;

[0217] Step 4.2: Calculate the total angle change. Calculate the total angle change over a period of time:

[0218]

[0219] Among them, ΔΘ L is the total angle change of the left hand, ΔΘ R is the total angle change of the right hand;

[0220] Step 4.3: Calculate the mean motion amplitude difference:

[0221] Calculate the difference coefficient M of the average movement amplitude of the left and right hands error :

[0222]

[0223] Furthermore, in step 4, a threshold θ is set threshold , and then count the time the left and right hands exceed the threshold within a period of time:

[0224]

[0225] Average movement time variation coefficient T error Calculated using the following formula:

[0226]

[0227] Among them, T L is the left hand movement time, T R Right hand movement time.

[0228] Furthermore, in step 5, the following formula is used to identify the movement of the left and right sides of the patient's body:

[0229]

[0230] Let M t ∈[0,1] and T t ∈[0,1]

[0231] If M error >M t , ΔΘ L >Θ R , the voice module prompts: Please move your right hand;

[0232] If M error <M t , ΔΘ L <Θ R , then the voice module prompts: Please move your left hand;

[0233] If T error >T t , T L >T R , the voice module prompts: Please use your right hand more;

[0234] If T error <T t , T L <T R , the voice module prompts: Please use your left hand more;

[0235] If the voice module prompts more than N cycles, M error and T error If there is still no change, the dry electrode electrical stimulation module stimulates the head, the head vibration module vibrates, and the lower limb vibration module and upper limb vibration module on the corresponding side vibrate.

[0236] For the distance measuring module of the present invention, a laser distance measuring sensor or an ultrasonic distance measuring module can be used. Assuming that the affected side is the left side, an alarm prompt is given according to the following rule table.

[0237]

[0238] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various deformations, modifications, and substitutions made to the technical solutions of the present invention by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. A rehabilitation assistance system for patients with unilateral spatial neglect, characterized in that: Including head-mounted equipment, upper limb equipment and lower limb equipment; The head-mounted device includes a head controller, a head posture detection module, a voice module, a head vibration module and a dry electrode electrical stimulation module; The lower limb device includes a lower limb controller, a lower limb posture detection module, a lower limb vibration module and a distance measurement module; The upper limb device includes an upper limb controller, an upper limb posture detection module and an upper limb vibration module; The head-mounted device is used to be worn on the patient's head, the lower limb device is used to be worn on the patient's legs, and the upper limb device is used to be worn on the patient's arms; When the distance between the affected side of the patient and the obstacle is lower than a threshold, the voice module plays an alarm audio, the head vibration module vibrates, and the lower limb vibration module and the upper limb vibration module on the corresponding side vibrate; When the distance between the affected side of the patient and the obstacle continues to shorten after falling below a threshold, the dry electrode electrical stimulation module stimulates the head.

2. A rehabilitation assistance system for patients with unilateral spatial neglect according to claim 1, characterized in that: The lower limb device, the upper limb device and the head-mounted device also include a power supply and a wireless communication module.

3. The rehabilitation assistance system for patients with unilateral spatial neglect according to claim 1, characterized in that: There are two lower limb devices and two upper limb devices, which are distributed on both sides of the patient's body.

4. A rehabilitation assistance method for patients with unilateral spatial neglect, characterized in that: include: Step 1, data collection; Step 2, data preprocessing; Step 3, angle calculation; Step 4: Calculate the movement amplitude: Calculate the average movement amplitude difference coefficient of the left and right hands, and use the average movement amplitude difference coefficient to reflect the movement of the left and right sides; Step 5, movement time calculation: calculate the average movement time difference coefficient of the left and right hands, and reflect the movement situation of the left and right sides through the average movement time difference coefficient; Step 6, reminder trigger calculation; comprehensive average movement amplitude difference coefficient and average movement time difference coefficient, identify the patient's left and right side movement status, and remind the patient.

5. The rehabilitation assistance method for patients with unilateral spatial neglect according to claim 4, characterized in that: The step 1 comprises: acquiring acceleration data, angular velocity data and magnetic field data from the 3-axis acceleration, 3-axis gyroscope and 3-axis geomagnetic sensor of the upper limb posture detection module respectively; Among them, the acceleration data is: a x ,a y ,a z ; Angular velocity data is: ω x ,ω y ,ω z ; Acceleration data: m x ,m y ,m z ; The step 2 includes: Normalization of acceleration data: Where: a: acceleration vector; |a|: modulus of acceleration vector, representing the composite value of three-axis acceleration; anorm: normalized acceleration vector; Normalization of geomagnetic data: Where: m: geomagnetic vector, |m|: modulus of geomagnetic vector, mnorm: normalized geomagnetic vector; Gyroscope data calibration: ω'=ω-b g Where: b g =[b x b y b z ] T , which is the zero bias error vector of the gyroscope, used to calibrate the gyroscope data; ω=[ω x ω y ω z ] T is the angular velocity vector.

6. The rehabilitation assistance method for patients with unilateral spatial neglect according to claim 5, characterized in that: The step 3 comprises: Step 3.1, initialize the attitude quaternion: <h2 style=";text-align:left;direction:ltr">q=[q0 q1 q2 q3]<h2 style=";text-align:left;direction:ltr"> T <h2 style=";text-align:left;direction:ltr"> =[1 0 0 0]<h2 style=";text-align:left;direction:ltr"> T Step 3.2, initialize the state variables and covariance matrix of the Kalman filter, including: Step 3.2.1, state vector: x=[φ θ ψ] T Where φ is the roll angle, θ is the pitch angle, and ψ is the yaw angle; Step 3.2.2, process noise covariance matrix: Step 3.2.3, measurement noise covariance matrix: Step 3.3, initialize the covariance matrix: Step 3.4, state prediction, includes: Step 3.4.1, use the gyroscope data to calculate the changes in roll angle, pitch angle and yaw angle: Δφ=ω′ x Δt, Δθ=ω′yΔt, Δψ=ω′ z Δt Δt is the sampling time interval, which represents the time difference between two data samples; Prediction status: Where: x k|k-1 is the predicted state vector, which represents the estimated value of the state at the current moment; x k-1 The state vector at the previous moment; Forecast covariance matrix: P k|k-1 =P k-1 +Q P k|k-1 Updated state covariance matrix; P k-1 The state covariance matrix at the previous moment; Step 3.4.2, Status Update: Correct the state using accelerometer and magnetometer measurements: Roll and pitch angle calculation based on accelerometer: Yaw angle calculation based on magnetometer: m′ x =m x cosθ acc +m y sinφ acc sinθ acc +m z cosφ acc sinθ acc m y =m y cosφ acc -m z sinφ acc Yaw angle: Update Kalman filter observations: Measurements: Kalman gain: K=P k|k-1 (P k|k-1 +R) -1 Status Update: x k =x k|k-1 +K(z-x k|k—1 ) x k is the updated state vector; Covariance update: P k =(I-K)P k|k-1 P k predicted state covariance matrix; I identity matrix, used to update the covariance matrix; Step 3.5, AHRS algorithm fusion, uses quaternion update and gyroscope integration to fuse data, including: Step 3.5.1 Quaternion update: Quaternion differential equation: in: q=[q0 q1 q2 q3] T is a quaternion used to represent three-dimensional posture, q0 is the scalar part of the quaternion, q1 q2 q3 is the vector part of the quaternion, is the quaternion multiplication symbol, used to describe the update of the quaternion; Use discretized integral update: Step 3.5.2, normalize the quaternion: Step 3.5.3, calculate Euler angles from quaternions: θ=arcsin(2(q0q2-q3q1)) Step 3.6, real-time loop: Repeat steps 3.1 to 3.5 to calculate the pose in real time.

7. The rehabilitation assistance method for patients with unilateral spatial neglect according to claim 4, characterized in that: The step 4 comprises: Step 4.1: For each time point, calculate the angle change of the left and right arms: Dth L (t)=θ L (t)-θ L (t-1) Dth R (t)=θ R (t)-θ R (t-1) Where Δθ L (t) and Δθ R (t) are the angles of the left and right hands at time t, respectively; Step 4.2: Calculate the total angle change. Calculate the total angle change over a period of time: Among them, ΔΘ L is the total angle change of the left hand, ΔΘ R is the total angle change of the right hand; Step 4.3: Calculate the mean motion amplitude difference: Calculate the difference coefficient M of the average movement amplitude of the left and right hands error :

8. The rehabilitation assisting method for patients with unilateral spatial neglect according to claim 7, characterized in that: In step 4, the average motion time difference coefficient T error Calculated using the following formula: Among them, T L is the left hand movement time, T R Right hand movement time.

9. The rehabilitation assistance method for patients with unilateral spatial neglect according to claim 8, characterized in that: In step 5, the following formula is used to identify the movement of the left and right sides of the patient's body: Let M t ∈[0,1] and T t ∈[0,1] If M error >M t , ΔΘ L >Θ R , the voice module prompts: Please move your right hand; If M error <M t , ΔΘ L <Θ R , then the voice module prompts: Please move your left hand; If T error >T t , TL>TR, then the voice module prompts: Please use your right hand more; If T error <T t , TL<TR, then the voice module prompts: Please use your left hand more; If the voice module prompts more than N cycles, M error and T error If there is still no change, the dry electrode electrical stimulation module stimulates the head, the head vibration module vibrates, and the lower limb vibration module and upper limb vibration module on the corresponding side vibrate.

Images