A hand exoskeleton control method, device and system

Abstract

The application provides a hand exoskeleton control method, device and system, and relates to the technical field of exoskeletons.The hand exoskeleton control method comprises the following steps: when the movement mode adapted to a target finger is an assisting mode, the movement characteristic data of the target finger is acquired, and it is judged whether the target finger needs to be assisted according to the movement characteristic data; if yes, corresponding assisting data is generated according to the movement gear adapted to the target finger and the flexion and extension intention, the hand exoskeleton is controlled to apply assisting force to the target finger according to the assisting data, and after detecting that the target finger moves by a preset displacement, the assisting force is cancelled, and the step of acquiring the movement characteristic data of the target finger and judging whether the target finger needs to be assisted according to the movement characteristic data is returned to be executed.The application can realize intelligent on-demand assistance, and improve the assisting effect of the hand exoskeleton.

Description

Technical Field

[0001] This invention relates to the field of exoskeleton technology, and more specifically, to a method, device, and system for controlling a hand exoskeleton. Background Technology

[0002] Hand exoskeletons are exoskeleton devices specifically designed to assist hand movements, and they have significant application value in hand function rehabilitation and hand motor enhancement training. Currently, hand exoskeletons can typically only perform passive training, that is, continuously provide external assistance to the hand. Their training modes are relatively simple, the rehabilitation effects are poor, and the level of intelligence is low, resulting in limited applicable scenarios and an inability to adapt to training subjects in different situations. Summary of the Invention

[0003] The problem addressed by this invention is how to improve the intelligence level of hand exoskeletons.

[0004] To address the above problems, this invention provides a method for controlling a hand exoskeleton, comprising:

[0005] When the target finger is adapted to the motion mode of the assist mode, the motion feature data of the target finger is acquired, and it is determined whether the target finger needs to be assisted based on the motion feature data.

[0006] If so, then generate corresponding assistance data based on the appropriate movement level and flexion / extension intention of the target finger, control the hand exoskeleton to apply assistance to the target finger according to the assistance data, cancel the assistance after detecting the preset displacement of the target finger, and return to the step of obtaining the motion feature data of the target finger and determining whether the target finger needs assistance based on the motion feature data.

[0007] Optionally, it also includes:

[0008] In automatic adaptation mode, motion feature data of each finger in a preset number of training sessions are acquired, wherein the motion feature data includes at least one of motion force, motion speed and motion displacement;

[0009] Based on the motion characteristic data of each finger in the preset training sessions, a motion capability representation value for each finger is generated.

[0010] The motor ability representation value of each finger is compared with the motor pattern threshold to determine the threshold range of the motor ability representation value of each finger.

[0011] The appropriate movement mode for each finger is determined based on the threshold range of the movement ability representation value of each finger. The movement mode includes an active mode, a passive mode, and the assisted movement mode.

[0012] Optionally, after determining the appropriate movement pattern for each finger based on the threshold range of the movement ability representation value of each finger, the method further includes:

[0013] In a preset sequence of exercise levels, the minimum exercise level that is greater than the exercise capacity representation value of the finger is determined; the minimum exercise level is then used as the exercise level that the finger is adapted to.

[0014] Optionally, determining whether the target finger needs assistance based on the motion feature data includes:

[0015] When the movement force of the target finger is less than the standard force, and the movement trajectory of the target finger is within the preset safe trajectory range, it is determined that the target finger needs to be assisted; wherein, the standard force is determined based on the movement level adapted to the target finger;

[0016] Alternatively, when the movement speed of the target finger is less than the standard speed and the trajectory of the target finger is within a preset safe trajectory range, it is determined that the target finger needs to be assisted; wherein, the standard speed is determined based on the movement level adapted to the target finger.

[0017] Optionally, it also includes:

[0018] Acquire motion feature data of the target finger in a preset number of training sessions, and determine whether the target finger meets the preset gear adjustment conditions;

[0019] When the target finger meets the upgrade conditions, the sports level adapted to the target finger is increased. The upgrade conditions include that the sports feature data of the target finger in a preset number of training sessions are all greater than the standard upgrade value corresponding to the current sports level.

[0020] When the target finger meets the downgrade condition, the motion level adapted to the target finger is reduced. The downgrade condition includes that the motion feature data of the target finger in a preset number of training sessions are all less than the standard downgrade value corresponding to the current motion level.

[0021] Optionally, it also includes:

[0022] When in a multi-finger collaborative working state, each collaborative finger is regarded as the target finger;

[0023] When in a single-finger working state, the single finger being worked is taken as the target finger.

[0024] Optionally, it also includes:

[0025] When in mirror control state, the position information, speed information and motion force information of the active side finger are collected, and the position information, speed information and motion force information of the active side finger are used as the target position information, target speed information and target motion force information of the corresponding passive side finger.

[0026] Obtain the current actual position information, actual speed information, and actual movement force information of the passive side finger;

[0027] Based on the target position information and actual position information, target speed information and actual speed information, and target motion force information and actual motion force information of the passive side finger, motion control information of the passive side finger is generated.

[0028] Optionally, it also includes:

[0029] Acquire the position, velocity, and force information of the target finger;

[0030] The position, speed, and force of movement of the target finger are mapped to the virtual reality device in real time.

[0031] The present invention also proposes a hand exoskeleton control device, including a computer-readable storage medium storing a computer program and a processor, wherein the computer program is read and executed by the processor to implement the hand exoskeleton control method as described above.

[0032] The present invention also proposes a hand exoskeleton control system, including a hand exoskeleton, a virtual reality device, and a hand exoskeleton control device as described above.

[0033] Compared with the prior art, the present invention has at least the following beneficial effects:

[0034] When the target finger is in the assisted movement mode, the motion characteristic data of the target finger is acquired, and the system automatically determines whether the target finger needs assistance based on the motion characteristic data. If assistance is needed, the system generates corresponding assistance data based on the motion level and flexion / extension intention of the target finger, and controls the hand exoskeleton to apply assistance to the target finger according to the assistance data. After detecting the preset displacement of the target finger, the assistance is canceled, and the system returns to acquiring the motion characteristic data of the target finger to re-determine whether the target finger needs assistance. This process is repeated in a loop. Specifically, in assistive mode, instead of providing continuous, single-intensity assistance, the entire movement stroke is discretized. Within a single discrete stroke, the actual condition of the target finger is determined based on its motion characteristic data, thereby judging whether the target finger needs assistance and the appropriate assistance intensity. If assistance is needed at a certain moment, corresponding assistance data is generated based on the target finger's appropriate movement level and flexion / extension intention. The hand exoskeleton is then controlled to apply appropriate assistance to the target finger according to the assistance data. After detecting the target finger's preset displacement, the assistance is canceled, and the process returns to retrieve the target finger's motion characteristic data. The assistance need is re-judged at the next moment. If assistance is still needed, the appropriate assistance for the current moment is applied precisely again. If no assistance is needed, it is temporarily withheld, allowing the target finger to train autonomously. Simultaneously, the process re-enters the next assistance need judgment, repeating the above process to achieve precise assistance implementation throughout the entire movement stroke. This intelligent, on-demand assistance improves the assistive effect of the hand exoskeleton and enhances the user experience.

[0035] Especially when applied to hand rehabilitation training, this invention enables intelligent, on-demand assistance. In assistive mode, instead of rigidly providing continuous, single assistance, the entire assisted movement is discretized. After a predetermined displacement of a target finger, the assistance is canceled, and the next adjacent stage reassesses whether assistance is needed. If needed, assistance of appropriate intensity is provided again, thus providing more flexible and adaptive discrete assistance; if not needed, no assistance is provided, allowing the target finger to train actively on its own, thereby achieving maximum active rehabilitation for patients with hand dysfunction. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of one embodiment of the hand exoskeleton control system of the present invention;

[0037] Figure 2 This is a schematic diagram of an embodiment of the hand exoskeleton control method of the present invention. Detailed Implementation

[0038] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0039] In this embodiment of the invention, see Figure 1 The hand exoskeleton control system includes a hand exoskeleton, a virtual reality device, and a hand exoskeleton control device. These three components can transmit data via wired communication or wireless communication technologies such as Bluetooth. The hand exoskeleton is worn on the hand, with corresponding placement areas for the wrist, palm, and fingers. The hand exoskeleton can be equipped with a glove-like shell. The virtual reality device is used to synchronously simulate the exoskeleton's movements based on the motion characteristic data transmitted from the hand exoskeleton, providing an immersive experience for the user through rich virtual scene tasks. The hand exoskeleton control device includes a computer-readable storage medium storing a computer program and a processor. The computer program is read and executed by the processor to implement the hand exoskeleton control method described below.

[0040] See Figure 2 In this embodiment of the invention, the hand exoskeleton control method includes:

[0041] When the target finger is adapted to the motion mode of the assist mode, step S100 is executed to obtain the motion feature data of the target finger, and step S200 is executed to determine whether the target finger needs to be assisted based on the motion feature data.

[0042] In this system, the target finger is a designated training finger. When multiple fingers are working together, each cooperating finger is designated as the target finger. When a single finger is working, the single finger is designated as the target finger. For each target finger, steps S100-S400 are executed to achieve independent passive / active / assisted movement of the single finger. A training task library can be pre-set, which can include independent training tasks for single fingers or joint training tasks for multiple fingers. For example, it can include five-finger grasping, multi-finger coordinated gestures, or object grasping.

[0043] The movement modes include active mode, passive mode and assisted mode. When the movement mode adapted to the target finger is assisted mode, steps S100-S400 are executed.

[0044] Motion characteristic data may include at least one of motion force, motion speed, and motion displacement. Motion force may include fingertip pressure or finger flexion / extension force, which can be detected by force sensors on the hand exoskeleton. Motion speed can be determined by the rotational speed of the corresponding motor on the hand exoskeleton. Motion displacement can be determined by the angular displacement of the corresponding finger drive motor on the hand exoskeleton. The finger motion displacement and the angular displacement of the corresponding finger drive motor have a defined mapping relationship; therefore, the angular displacement of the corresponding motor is used to determine the finger flexion / extension angle / displacement. Here, finger flexion / extension includes finger displacement during metacarpophalangeal joint movement and finger displacement during interphalangeal joint movement, which can be calculated based on the mechanical drive structure of the hand exoskeleton.

[0045] In one embodiment, the hand exoskeleton sends motion feature data detected by sensors, or data such as motor speed and angular displacement, to the hand exoskeleton control device. In step S100, motion feature data of the target finger is acquired, and the hand exoskeleton control device receives the motion feature data of the target finger sent by the hand exoskeleton. In another embodiment, the hand exoskeleton stores motion feature data detected by sensors, or data such as motor speed, in a preset database. In step S100, motion feature data of the target finger is acquired, and the hand exoskeleton control device reads the motion feature data of the target finger from the preset database.

[0046] Motion characteristic data of the target finger are used to characterize the state of the target finger, and based on the state of the target finger, it can be determined whether the target finger needs assistance. Whether the target finger needs assistance can be determined based on one or more of the motion force, motion speed, and motion displacement.

[0047] This method acquires motion feature data of the target finger at multiple moments and generates a motion capability representation value for the target finger based on this data. This value is then compared with a corresponding standard value to determine whether assistance is needed. Methods for generating the motion capability representation value based on multiple moments include, but are not limited to: calculating the average of the multiple motion feature data as the corresponding finger's motion capability representation value; using the median of the multiple motion feature data as the corresponding finger's motion capability representation value; or using 0.8 times the maximum value or 1.2 times the minimum value of the multiple motion feature data as the corresponding finger's motion capability representation value. Similarly, for motion force, methods for calculating the motion capability representation value of the target finger based on the motion force at multiple moments include, but are not limited to: calculating the average of the motion force at multiple moments as the corresponding finger's motion capability representation value; using the median of the motion force at multiple moments as the corresponding finger's motion capability representation value; or using 0.8 times the maximum value or 1.2 times the minimum value of the multiple motion force as the corresponding finger's motion capability representation value.

[0048] In one embodiment, determining whether the target finger needs assistance based on motion characteristic data includes:

[0049] When the movement force of the target finger is less than the standard force, it is determined that the target finger needs assistance; when the movement force of the target finger is greater than or equal to the standard force, it is determined that the target finger does not need assistance. The standard force is determined based on the appropriate movement level for the target finger.

[0050] Alternatively, if the target finger's movement speed is less than the standard speed, it is determined that the target finger needs assistance; if the target finger's movement speed is greater than or equal to the standard speed, it is determined that the target finger does not need assistance. The standard speed is determined based on the appropriate movement level for the target finger.

[0051] Alternatively, if the target finger's displacement is less than the standard displacement, it is determined that the target finger requires assistance; if the target finger's displacement is greater than or equal to the standard displacement, it is determined that the target finger does not require assistance. The standard displacement is determined based on the appropriate movement level for the target finger.

[0052] Alternatively, the movement force and speed of the target finger can be compared with the corresponding standard values. If at least one of the following conditions is met, the target finger needs assistance; if both the movement force and speed are greater than or equal to the corresponding standard values, the target finger does not need assistance.

[0053] Alternatively, the movement force and displacement of the target finger can be compared with the corresponding standard values. If the movement force is less than the standard force or the movement displacement is less than the standard displacement, it is determined that the target finger needs assistance. If both the movement force and the movement displacement are greater than or equal to the corresponding standard values, it is determined that the target finger does not need assistance.

[0054] Alternatively, the movement speed and displacement of the target finger can be compared with the corresponding standard values. If the movement speed is less than the standard speed or the movement displacement is less than the standard displacement, it is determined that the target finger needs assistance. If both the movement speed and the movement displacement are greater than or equal to the corresponding standard values, it is determined that the target finger does not need assistance.

[0055] Alternatively, the movement force, speed, and displacement of the target finger can be compared with the corresponding standard values. If at least one of the following conditions is met: the movement force is less than the standard force, the movement speed is less than the standard speed, and the movement displacement is less than the standard displacement, then the target finger is determined to require assistance. If all three are greater than the corresponding standard values, then the target finger is determined not to require assistance.

[0056] Specifically, if the target finger's movement force is below the standard value, it indicates insufficient force, and assistance is needed to aid its movement. Similarly, if the target finger's movement speed is below the standard value, assistance is applied to accelerate its movement. If the target finger's displacement is below the standard value, meaning it cannot move to the standard displacement over a prolonged period (e.g., unable to move from point A to point B), assistance is required to aid its movement.

[0057] The standard force, standard speed, and standard displacement corresponding to each motion level can be preset.

[0058] By comparing the motion force, motion speed, and / or motion displacement with corresponding standard values, the system determines whether the target finger needs assistance based on the comparison results. The judgment logic is simple and the response speed is fast. The system allows users to select whether to make a judgment using a single motion feature parameter or a combination of multiple motion feature parameters, and users can also select the specific type of motion feature parameter used for assistance judgment to adapt to users' personalized needs, broaden the scope of application of the embodiments of the present invention, and improve user-friendliness.

[0059] In another implementation, when the activation of the safety mode is detected, the system combines a preset safety trajectory range with motion characteristic data to determine whether assistance is needed for the target finger. Specifically, this includes:

[0060] When the movement force of the target finger is less than the standard force and the trajectory of the target finger is within the preset safe trajectory range, it is determined that the target finger needs assistance; when the movement force of the target finger is greater than or equal to the standard force, or the trajectory of the target finger is outside the preset safe trajectory range, it is determined that the target finger does not need assistance. The standard force is determined based on the appropriate movement level for the target finger.

[0061] Alternatively, if the target finger's movement speed is less than the standard speed and its trajectory is within a preset safe trajectory range, it is determined that assistance needs to be applied to the target finger; if the target finger's movement speed is greater than or equal to the standard speed, or its trajectory is outside the preset safe trajectory range, it is determined that no assistance needs to be applied to the target finger. The standard speed is determined based on the appropriate movement level for the target finger.

[0062] Alternatively, if the movement force of the target finger is less than the standard force and the movement speed of the target finger is less than the standard speed, and the trajectory of the target finger is within the preset safe trajectory range, it is determined that the target finger needs to be assisted; if the movement force of the target finger is greater than or equal to the standard force, or the movement speed is greater than or equal to the standard speed, or the trajectory of the target finger is outside the preset safe trajectory range, it is determined that the target finger does not need to be assisted.

[0063] The preset safe trajectory range can be characterized by the finger flexion and extension angles, for example, metacarpophalangeal joints: (-10°, +5°), interphalangeal joints: (-10°, +0°). The preset safe trajectory range is obtained by acquiring the average value of the target finger's motion trajectory range from multiple training data points in active mode, and stored as the preset safe trajectory range. The user can choose whether to activate the safe mode. If activated, the preset safe trajectory range is used in conjunction with motion characteristic data to determine whether the target finger requires assistance.

[0064] The preset safety trajectory range ensures that the hand exoskeleton cannot apply assistance outside of this range, thus preventing finger injuries and guaranteeing the safety of the hand exoskeleton.

[0065] If so, proceed to step S300, generate corresponding assistance data based on the target finger's appropriate movement level and flexion / extension intention, control the hand exoskeleton to apply assistance to the target finger according to the assistance data, and proceed to step S400, after detecting the target finger's preset displacement, cancel the assistance, and return to step S100 to obtain the target finger's movement feature data and determine whether the target finger needs assistance based on the movement feature data.

[0066] The assistance data includes the level of assistance and the direction of assistance. The level of assistance is determined according to the appropriate movement level for the target finger, and the direction of assistance is determined according to the flexion and extension intention of the target finger.

[0067] Specifically, the level of assistance can be determined based on the difference between the actual motion characteristics of the target finger and the standard range of its corresponding motion level (such as standard force range / standard speed range). For example, if the target finger is suited for motion level 2, with a corresponding standard force range of 5N-7N, and the user's actual motion force is 2N, then the difference range is 3N-5N. In this case, the minimum value within the difference range (e.g., 3N) can be selected as the assistance value, or the minimum value within the difference range can be taken plus 0.5N (e.g., 3.5N). This provides the target finger with the minimum or relatively small assistance needed, allowing the user's target finger to move as actively as possible, ensuring the training effect.

[0068] Flexion-extension intention refers to the user's intention to flex or extend their fingers, i.e., the direction of the target finger's intended movement. In one embodiment, after determining that the target finger requires assistance based on motion feature data, the flexion-extension intention of the target finger is determined based on the motion feature data. Specifically, the direction of the target finger can be determined by the direction of the motion force / speed / displacement in the motion feature data, thereby determining its flexion-extension intention. In another embodiment, at the beginning of a movement stroke, the hand exoskeleton does not apply external force, and the target finger moves voluntarily. Motion feature data of this voluntary movement is collected to determine the flexion-extension intention. Subsequently, during the assistance process of this movement stroke, the flexion-extension intention is directly acquired to determine the direction of assistance until the end of the movement stroke. For example, after the hand exoskeleton is activated, the first movement stroke begins. The hand exoskeleton does not apply external force, and the target finger moves voluntarily. Based on its motion feature data, if the user's intention is to flex the target finger, then this movement stroke is defined as a flexion stroke. During the subsequent assistance process of this movement stroke, the direction of assistance is directly determined based on the direction of the flexion stroke. Among them, the movement can be switched or ended after the target finger stops for a certain period of time or the flexion and extension angle of the target finger reaches a preset angle threshold. For example, the target finger stops moving for 5 seconds, or the target finger flexes to a preset limit angle or extends to a preset limit angle.

[0069] The movement displacement of the target finger is determined by the rotational angular displacement of the corresponding motor on the hand exoskeleton. After detecting the preset displacement of the target finger, the assistance is canceled. That is, the assistance is canceled immediately after the target finger moves the preset displacement. Then, the judgment on whether to apply assistance is re-evaluated. If assistance is still needed, it is applied again. After the target finger moves the preset displacement, the assistance is canceled again, and the judgment on whether to apply assistance is re-evaluated. This process is repeated cyclically to detect the state of the target finger in a timely manner and provide the optimal adaptive assistance to the target finger. This achieves discrete assistance, improves the assistive effect of the hand exoskeleton, and enhances the user experience.

[0070] If the motion feature data indicates that the target finger does not require assistance, then return to step S100 to obtain the motion feature data of the target finger and determine whether the target finger requires assistance based on the motion feature data.

[0071] This invention, in its embodiments, acquires motion characteristic data of the target finger when the target finger's adaptation motion mode is in assistive mode. Based on this data, it automatically determines whether the target finger requires assistance. If assistance is needed, it generates corresponding assistance data according to the target finger's adaptation motion level and flexion / extension intention. The hand exoskeleton is then controlled to apply assistance to the target finger according to this data. Upon detecting a preset displacement of the target finger, the assistance is canceled, and the process returns to acquiring the target finger's motion characteristic data to re-determine whether assistance is needed. This cycle repeats. In other words, the assistive mode does not provide continuous, single-intensity assistance, but rather discretizes the entire motion stroke. In each stroke, the actual condition of the target finger is determined based on its motion characteristic data, thereby determining whether assistance is needed. The system determines whether the target finger needs assistance and the appropriate assistance intensity. If assistance is needed at a given moment, corresponding assistance data is generated based on the target finger's appropriate movement level and flexion / extension intention. The hand exoskeleton is then controlled to apply the appropriate assistance to the target finger according to the assistance data. Once a preset displacement of the target finger is detected, the assistance is canceled, and the process returns to retrieve the target finger's movement feature data. The assistance need is reassessed at the next moment. If assistance is still needed, the appropriate assistance for the current moment is applied precisely again. If no assistance is needed, it is temporarily withheld, allowing the target finger to train autonomously. Simultaneously, the system re-enters the next assistance need assessment, repeating the above process. This achieves discrete and precise assistance implementation, enabling intelligent on-demand assistance, thereby improving the assistive effect of the hand exoskeleton and enhancing the user experience.

[0072] Especially when applied to hand rehabilitation training, this invention enables intelligent, on-demand assistance. In assistive mode, instead of rigidly providing continuous, single assistance, the entire assisted movement is discretized. After a predetermined displacement of a target finger, the assistance is canceled, and the next adjacent stage reassesses whether assistance is needed. If needed, assistance of appropriate intensity is provided again, thus providing more flexible and adaptive discrete assistance; if not needed, no assistance is provided, allowing the target finger to train actively on its own, thereby achieving maximum active rehabilitation for patients with hand dysfunction.

[0073] Alternatively, the hand exoskeleton control method also includes:

[0074] Step S1: In automatic adaptation mode, acquire motion feature data of each finger in a preset training session, wherein the motion feature data includes at least one of motion force, motion speed and motion displacement.

[0075] The automatic adaptation mode refers to the movement pattern that the hand exoskeleton intelligently provides for the target finger. The preset training here refers to test training without applying any external assistance.

[0076] Optionally, in automatic adaptation mode, the virtual reality device can output the instruction information of the preset gesture, or the voice prompt can output the instruction information of the preset gesture, so as to prompt the target finger to make the preset gesture a preset number of times and obtain the motion characteristic data of each finger.

[0077] Step S2: Generate the motion capability representation value of each finger based on the motion feature data of each finger in the preset training.

[0078] The value representing the motor ability of the finger can be expressed using at least one of force representation data, velocity representation data, and displacement representation data.

[0079] Based on the motion feature data of each finger in a preset training session, a motion capability representation value for each finger is generated. The methods that can be used include, but are not limited to: calculating the average value of the motion feature data in the preset training session as the motion capability representation value of the corresponding finger; taking the median of the motion feature data in the preset training session as the motion capability representation value of the corresponding finger; or taking 0.8 times the maximum value or 1.2 times the minimum value of the motion feature data in the preset training session as the motion capability representation value of the corresponding finger.

[0080] Step S3: Compare the motor ability representation value of each finger with the motor pattern threshold to determine the threshold range of the motor ability representation value of each finger.

[0081] Step S4: Determine the appropriate movement mode for each finger based on the threshold range of the movement ability representation value of each finger. The movement mode includes active mode, passive mode and assisted mode.

[0082] The exoskeleton provides no assistance, allowing the user to actively perform gestures according to the task requirements. The passive mode provides continuous assistance to the fingers, helping the user complete gestures. The assisted mode provides fine-grained assistance to the fingers only when needed, and no assistance when not required.

[0083] The appropriate movement pattern for each finger is determined based on the threshold range of its motor ability representation value, and may include:

[0084] When the force intensity of a finger is greater than or equal to a first force threshold, the movement mode adapted to that finger is determined to be an active mode; when the force intensity of a finger is less than the first force threshold but greater than or equal to a second force threshold, the movement mode adapted to that finger is determined to be an assisted mode; when the force intensity of a finger is less than the second force threshold, the movement mode adapted to that finger is determined to be a passive mode. The movement mode thresholds include the first force threshold and the second force threshold.

[0085] or,

[0086] When the velocity data of a finger is greater than or equal to a first velocity threshold, the movement mode adapted to that finger is determined to be an active mode; when the velocity data of a finger is less than the first velocity threshold but greater than or equal to a second velocity threshold, the movement mode adapted to that finger is determined to be an assisted mode; when the velocity data of a finger is less than the second velocity threshold, the movement mode adapted to that finger is determined to be a passive mode. The movement mode thresholds include the first velocity threshold and the second velocity threshold.

[0087] or,

[0088] When the displacement data of a finger is greater than or equal to a first displacement threshold, the movement mode adapted to that finger is determined to be an active mode; when the displacement data of a finger is less than the first displacement threshold but greater than or equal to a second displacement threshold, the movement mode adapted to that finger is determined to be an assisted mode; when the displacement data of a finger is less than the second displacement threshold, the movement mode adapted to that finger is determined to be a passive mode. The movement mode thresholds include the first displacement threshold and the second displacement threshold.

[0089] or,

[0090] When both the force and velocity data of a finger are within the first threshold range, the movement mode adapted to that finger is determined to be the active mode; when both the force and velocity data of a finger are within the second threshold range, or one is within the first threshold range and the other is within the second threshold range, the movement mode adapted to that finger is determined to be the assisted mode; when at least one of the force and velocity data of a finger is within the third threshold range, the movement mode adapted to that finger is determined to be the passive mode. The first threshold range refers to force data greater than or equal to the first force threshold and velocity data greater than or equal to the first velocity threshold; the second threshold range refers to velocity data less than the first velocity threshold and greater than or equal to the second velocity threshold, and force data less than the first force threshold and greater than or equal to the second force threshold; the third threshold range refers to force data less than the second force threshold and velocity data less than the second velocity threshold.

[0091] In addition, force characterization data and displacement characterization data, or velocity characterization data and displacement characterization data, can also be combined to determine the required motion pattern. The relevant logic is as described above and will not be repeated here.

[0092] By using the motion characteristic data of each finger in the preset training, the motion capability representation value of each finger is generated to determine the motion state of each finger, and then the most suitable motion mode is intelligently matched for each finger, so as to provide assistance according to the actual needs of each finger and ensure the auxiliary effect of the hand exoskeleton.

[0093] Optionally, after step S4, the method further includes: determining the smallest motion level greater than the motion capability representation value of the finger in a preset motion level sequence; and using the smallest motion level as the motion level adapted to the finger.

[0094] For example, suppose the preset speed gear sequence is: gear 1 is 1° / s-3° / s, gear 2 is 4° / s-6° / s, gear 3 is 7° / s-9° / s, gear 4 is 10° / s-12° / s, and gear 5 is 13° / s-15° / s, and the characteristic value of the finger's movement ability is 3.5° / s, then the movement gear closest to it in the preset speed gear sequence is gear 2.

[0095] Optionally, after using the hand exoskeleton for a period of time, the hand exoskeleton control device can automatically collect motion characteristic data from a preset training session to determine whether the target finger meets the preset adjustment conditions. Alternatively, an automatic adaptation mode can be manually activated to adjust the adjustment level. Specifically, the hand exoskeleton control method also includes:

[0096] Acquire motion characteristic data of the target finger during a preset number of training sessions to determine whether the target finger meets preset adjustment conditions. When the target finger meets the upshift condition, increase the appropriate motion level for the target finger. The upshift condition includes that the motion characteristic data of the target finger during the preset number of training sessions are all greater than the standard upshift value corresponding to the current motion level. When the target finger meets the downshift condition, decrease the appropriate motion level for the target finger. The downshift condition includes that the motion characteristic data of the target finger during the preset number of training sessions are all less than the standard downshift value corresponding to the current motion level. When the target finger does not meet either the upshift or downshift condition, maintain the appropriate motion level for the target finger.

[0097] The motion characteristic data includes, but is not limited to, at least one of motion force and motion speed.

[0098] The configurable conditions for increasing the intensity of the target finger's movement in a preset training session are: both the intensity and speed of the movement are greater than the standard increase value corresponding to the current intensity level. The configurable conditions for decreasing the intensity of the target finger's movement in a preset training session are: both the intensity and speed of the movement are less than the standard decrease value corresponding to the current intensity level.

[0099] Different target fingers correspond to their respective exercise levels, and different exercise levels indicate different exercise intensities. Each exercise level has a corresponding standard upshift force and standard upshift speed.

[0100] Therefore, when the motion characteristic parameters such as motion intensity and speed meet the conditions for upgrading, the motion level adapted to the target finger can be increased; when the motion characteristic parameters such as motion intensity and speed meet the conditions for downgrading, the motion level adapted to the target finger can be decreased. This allows the motion level to be adjusted according to the motion state of the target finger, enabling the hand exoskeleton to provide more precise assistance that better meets the needs of the target finger.

[0101] Alternatively, the hand exoskeleton control method also includes:

[0102] When in mirror control mode, the position, speed and motion force information of the active side finger are collected, and the position, speed and motion force information of the active side finger are used as the target position, target speed and target motion force information of the corresponding passive side finger.

[0103] Obtain the current actual position, actual speed, and actual movement force information of the passive side finger;

[0104] Based on the target position information and actual position information, target speed information and actual speed information, and target motion force information and actual motion force information of the passive finger, motion control information of the passive finger is generated.

[0105] In this context, mirror control refers to a state where one hand exoskeleton actively moves, synchronously controlling the other hand exoskeleton to perform a synchronized passive movement. Motion control information for the passive hand fingers is generated by using the differences between target position information and actual position information, target velocity information and actual velocity information, and target motion force information and actual motion force information. This motion control information is then used to control the movement of the hand exoskeleton. The active hand fingers move the hand exoskeleton through active gestures. The position and motion force information of the active hand fingers are transmitted to the active hand exoskeleton, which, through synchronous mapping program code, drives the corresponding mapped motors in the passive hand exoskeleton to move, thus driving the passive hand exoskeleton to perform a corresponding mirror movement, thereby achieving passive mirror movement on the passive side.

[0106] In this embodiment of the invention, the hand exoskeleton control device can also be connected to other control signal input devices, such as myoelectric bracelets worn on the active side of the hand, other forms of hand exoskeletons and other wearable devices. After receiving the control signal transmitted by the external input device, the device generates motion control information for the passive side fingers based on the control signal to perform passive assisted movement.

[0107] In this embodiment of the invention, the hand exoskeleton control device can also be connected to other terminal devices, such as robotic hands. The hand exoskeleton control device can collect the position, speed, and force information of each finger of the hand exoskeleton and transmit it to the connected terminal devices to control them to perform corresponding actions. For example, if the hand exoskeleton performs a grasping action, the hand exoskeleton control device can control the robotic hand to perform the same grasping action. This expands the application scenarios of the hand exoskeleton control method, device, and system described in this embodiment of the invention.

[0108] Alternatively, the hand exoskeleton control method also includes:

[0109] Acquire the position, velocity, and force of the target finger; map the position, velocity, and force of the target finger to the virtual reality device in real time.

[0110] The virtual reality device is pre-set with a training task library integrated with 3D animation. During the training process, the virtual reality device displays the training task and, at the same time, receives or reads the position information and movement force information of the target finger through real-time communication with the hand exoskeleton. Through its own control attributes, it realizes real-time mapping and immersive guidance between the VR animation task and the actual rehabilitation hand exoskeleton.

[0111] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for controlling a hand exoskeleton, characterized in that, Includes the following steps: When the target finger is adapted to the motion mode of the assist mode, the motion feature data of the target finger is acquired, and it is determined whether the target finger needs to be assisted based on the motion feature data. If so, then generate corresponding assistance data according to the movement level and flexion / extension intention of the target finger, control the hand exoskeleton to apply assistance to the target finger according to the assistance data, cancel the assistance after detecting the preset displacement of the target finger, and return to the step of obtaining the movement feature data of the target finger and determining whether the target finger needs to be assisted according to the movement feature data; The step of determining whether the target finger needs assistance based on the motion feature data includes: When the movement force of the target finger is less than the standard force, and the trajectory of the target finger is within the preset safe trajectory range, it is determined that the target finger needs to be assisted; wherein, the standard force is determined based on the movement level adapted to the target finger; Alternatively, when the movement speed of the target finger is less than the standard speed and the trajectory of the target finger is within a preset safe trajectory range, it is determined that the target finger needs to be assisted; wherein, the standard speed is determined based on the movement level adapted to the target finger; In this process, at the beginning of a movement, the exoskeleton of the hand does not apply external force, and the target finger moves on its own. The motion characteristic data of the active movement is collected to determine the flexion and extension intention. Then, during the assistance process of this movement, the flexion and extension intention is directly obtained to determine the direction of assistance until the end of the movement. The assistance data includes the level of assistance and the direction of assistance. The level of assistance is determined according to the appropriate movement level of the target finger, and the direction of assistance is determined according to the flexion and extension intention of the target finger. The amount of assistance is determined based on the difference between the actual motion characteristic data of the target finger and the standard value range corresponding to its motion level. The hand exoskeleton control method further includes: Acquire motion feature data of the target finger in a preset number of training sessions, and determine whether the target finger meets the preset gear adjustment conditions; When the target finger meets the upgrade conditions, the sports level adapted to the target finger is increased. The upgrade conditions include that the sports feature data of the target finger in a preset number of training sessions are all greater than the standard upgrade value corresponding to the current sports level. When the target finger meets the downgrade condition, the motion level adapted to the target finger is reduced. The downgrade condition includes that the motion feature data of the target finger in a preset number of training sessions are all less than the standard downgrade value corresponding to the current motion level. Furthermore, in the assisted mode, once the preset displacement of the target finger is detected, the assistance is immediately canceled, allowing the target finger to move on its own.

2. The hand exoskeleton control method as described in claim 1, characterized in that, Also includes: In automatic adaptation mode, motion feature data of each finger in a preset number of training sessions are acquired, wherein the motion feature data includes at least one of motion force, motion speed and motion displacement; Based on the motion characteristic data of each finger in the preset training sessions, a motion capability representation value for each finger is generated. The motor ability representation value of each finger is compared with the motor pattern threshold to determine the threshold range of the motor ability representation value of each finger. The appropriate movement mode for each finger is determined based on the threshold range of the movement ability representation value of each finger. The movement mode includes an active mode, a passive mode, and the assisted movement mode.

3. The hand exoskeleton control method as described in claim 2, characterized in that, After determining the appropriate movement pattern for each finger based on the threshold range of the movement ability representation value of each finger, the method further includes: In the preset sequence of exercise levels, determine the minimum exercise level that is greater than the exercise capacity representation value of the finger; The minimum motion level is used as the motion level that the finger is adapted to.

4. The hand exoskeleton control method as described in any one of claims 1 to 3, characterized in that, Also includes: When in a multi-finger collaborative working state, each collaborative finger is regarded as the target finger; When in a single-finger working state, the single finger being worked is taken as the target finger.

5. The hand exoskeleton control method as described in any one of claims 1 to 3, characterized in that, Also includes: When in mirror control state, the position information, speed information and motion force information of the active side finger are collected, and the position information, speed information and motion force information of the active side finger are used as the target position information, target speed information and target motion force information of the corresponding passive side finger. Obtain the current actual position information, actual speed information, and actual movement force information of the passive side finger; Based on the target position information and actual position information, target speed information and actual speed information, and target motion force information and actual motion force information of the passive side finger, motion control information of the passive side finger is generated.

6. The hand exoskeleton control method as described in any one of claims 1 to 3, characterized in that, Also includes: Acquire the position, velocity, and force information of the target finger; The position, speed, and force of movement of the target finger are mapped to the virtual reality device in real time.

7. A hand exoskeleton control device, characterized in that, The method includes a computer-readable storage medium storing a computer program and a processor, the computer program being read and executed by the processor to implement the hand exoskeleton control method as described in any one of claims 1 to 6.

8. A hand exoskeleton control system, characterized in that, This includes hand exoskeletons, virtual reality devices, and the hand exoskeleton control device as described in claim 7.

Images