Unmanned aerial vehicle operating upper limb assisting exoskeleton

Abstract

The present application relates to the technical field of power-assisted device, in particular to a kind of unmanned aerial vehicle operating upper limb power-assisted exoskeleton, including exoskeleton body, main control device and electromyographic sensor.Exoskeleton body contains the shoulder support component of adaptation shoulder and the arm component of adaptation arm, shoulder support component two sides each is equipped with an arm component, is connected by first flexible connecting piece.Arm component is equipped with electromyographic sensor, its signal and the rocker push signal of unmanned aerial vehicle remote controller are all transmitted to main control device;The first fixed part of arm component and second fixed part are connected by driving element, main control device can control driving element drive the relative rotation of both, and the winding mechanism of shoulder support component is wound or released to first connecting piece.The exoskeleton can accurately identify operation intention, cooperate upper limb movement to provide targeted power assistance, effectively reduce shoulder, elbow muscle load, reduce arm swing, delay fatigue, improve unmanned aerial vehicle control stability and operation durability, applicable to electric power inspection, surveying and mapping and other long-time work scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power-assisted devices, in particular to a power-assisted exoskeleton for operating upper limbs of unmanned aerial vehicles. BACKGROUND

[0002] With the rapid development of unmanned aerial vehicle technology, its application scenarios have expanded from the initial entertainment aerial photography to surveying and mapping, power inspection, agricultural plant protection, emergency rescue and other professional fields. In these professional scenarios, the operator often needs to control the unmanned aerial vehicle for a long time and with high precision. For example, in the power inspection task, the operator needs to control the unmanned aerial vehicle to fly along the power line for several hours, accurately avoid obstacles on the line, and capture equipment details; in the surveying and mapping operation, the unmanned aerial vehicle needs to be kept flying stably along the preset route to obtain accurate data.

[0003] In this process, the operator's upper limbs need to be kept in the air for a long time to control the remote controller, and need to make fine actions such as pushing, pulling and rotating frequently. This continuous muscle tension and repeated action can cause the shoulder, elbow and wrist muscles to increase dramatically, which can easily cause muscle fatigue. Long-term operation not only reduces the accuracy and efficiency of unmanned aerial vehicle control, increases the risk of task failure, but also can cause chronic muscle damage, such as periarthritis and tendonitis, which seriously affects the health and professional life of the operator.

[0004] Exoskeleton technology, as a mechanical device that can enhance human movement ability and reduce muscle load, has been widely used in industrial assembly, medical rehabilitation and other fields since its origin in the 1960s. It can effectively share the load of the human body and assist limb movement through mechanical structure and human movement coordination.

[0005] In the field of unmanned aerial vehicle control, some attempts related to exoskeleton have appeared in the prior art, such as the force tactile regeneration exoskeleton structure provided by Chinese patent CN110096066A, which assists in controlling the flight attitude of the unmanned aerial vehicle by detecting the signal of the upper limb movement pressing pressure strain sheet; the unmanned aerial vehicle remote operation tactile perception type wearable upper limb exoskeleton system provided by Chinese utility model patent CN209373435U takes the upper limb exoskeleton movement as the control input of the unmanned aerial vehicle. However, these prior arts focus more on using exoskeleton as a control input means of the unmanned aerial vehicle, and fail to fully address the problem of upper limb muscle load when the operator controls the unmanned aerial vehicle for a long time, and provide effective power support to reduce muscle fatigue and injury risk.

[0006] Therefore, there is an urgent need for an exoskeleton device that can accurately perceive the operator's control intention and provide targeted power assistance to the upper limbs, thereby reducing the muscle burden of the upper limbs when controlling the unmanned aerial vehicle for a long time. SUMMARY

[0007] The unmanned aerial vehicle operating upper limb assisting exoskeleton can sense the myoelectric signal of the human upper limb in real time, simultaneously receives the rocker pushing force signal of the unmanned aerial vehicle remote controller, and the main control device can accurately identify the operation intention of the operator based on the two types of signals.

[0008] The unmanned aerial vehicle operating upper limb assisting exoskeleton comprises an exoskeleton body, a main control device, a power module and a myoelectric sensor.

[0009] The exoskeleton body is composed of a shoulder supporting assembly adapted to the human shoulder and back and an arm assembly adapted to the human arm.

[0010] The arm assembly is provided with the myoelectric sensor, the signal output end of the myoelectric sensor is connected to the first signal input end of the main control device, so as to send the sensed myoelectric signal to the main control device, and the second signal input end of the main control device is connected to the signal output end of the unmanned aerial vehicle remote controller, so as to receive the rocker pushing force signal sent by the unmanned aerial vehicle remote controller.

[0011] The arm assembly comprises a first fixed part adapted to the upper arm and a second fixed part adapted to the forearm, the first fixed part and the second fixed part are connected through a driving member, the signal input end of the driving member is connected to the first signal output end of the main control device, so as to receive the first control signal sent by the main control device and drive the first fixed part and the second fixed part to rotate relative to each other.

[0012] The first flexible connecting piece comprises a first connecting piece, the shoulder supporting assembly is provided with a rotary winding mechanism, the first end of the first connecting piece is connected to the second fixed part, the second end of the first connecting piece is wound around the winding wheel of the rotary winding mechanism, the signal input end of the rotary winding mechanism is connected to the second signal output end of the main control device, so as to receive the second control signal sent by the main control device and drive the first connecting piece to wind around the winding wheel or release from the winding wheel.

[0013] The main control device, the myoelectric sensor, the rotary winding mechanism and the driving member are electrically connected to the power module.

[0014] According to a preferred embodiment, the first flexible connecting member further comprises a second connecting member, a first end of the second connecting member being connected to the first fixing part, and a second end of the second connecting member being connected to the shoulder support assembly.

[0015] According to a preferred embodiment, a waist assembly adapted to the waist of a human body is further included, the waist assembly being connected to the shoulder support assembly through a second flexible connecting member, the second flexible connecting member being configured to be adjustable in length.

[0016] According to a preferred embodiment, the first flexible connecting member and the second flexible connecting member are both made of nylon webbing.

[0017] According to a preferred embodiment, a support frame provided on the front surface of the waist assembly is further included, a lower end of the support frame being connected to the front surface of the waist assembly, and an upper end of the support frame being provided with a placement table adapted to a remote controller of a UAV.

[0018] According to a preferred embodiment, the lower end of the support frame is hingedly connected to the front surface of the waist assembly.

[0019] According to a preferred embodiment, the second flexible connecting member comprises a third connecting member provided on the front surface of the shoulder support assembly and a fourth connecting member provided on the back surface of the shoulder support assembly, the third connecting member being a three-point structure, two upper ends of the three-point structure being respectively connected to two sides of the front surface of the shoulder support assembly, and a lower end of the three-point structure being connected to the front surface of the waist assembly.

[0020] According to a preferred embodiment, the first fixing part comprises an upper arm module and a first arm ring connected to the upper arm module, the second fixing part comprises a forearm module and a second arm ring connected to the forearm module, and the electromyographic sensor is provided on the inner side of the first arm ring and / or the second arm ring.

[0021] According to a preferred embodiment, the upper arm module and the forearm module are configured to be adjustable in length.

[0022] According to a preferred embodiment, the master control device comprises:

[0023] a data receiving module, configured to receive the electromyographic signal and the joystick thrust signal;

[0024] a preprocessing module, configured to perform preprocessing and feature extraction on the electromyographic signal to obtain an electromyographic feature vector, and perform preprocessing and feature extraction on the joystick thrust signal to obtain a thrust feature vector;

[0025] a prediction module, configured to perform fusion processing on the electromyographic feature vector and the thrust feature vector by using a multi-modal information fusion algorithm to identify an operation intention;

[0026] A strategy generation module determines an exoskeleton control strategy based on the operation intention;

[0027] An execution module executes control based on the exoskeleton control strategy.

[0028] The technical solution of the unmanned aerial vehicle operation upper limb assisting exoskeleton provided by the present application has at least the following advantages and beneficial effects: (1) By arranging the electromyographic sensor on the arm assembly, the electromyographic signal of the human upper limb can be sensed in real time, and the rocker push signal of the unmanned aerial vehicle remote controller can be received at the same time. The main control device can accurately identify the operation intention of the operator based on the two types of signals. On this basis, the first fixed part and the second fixed part are driven to rotate relative to each other by the driving member, and the winding or releasing control of the first connecting piece by the rotating winding mechanism can make the exoskeleton and the human upper limb move cooperatively, provide targeted assistance for the flexion, swing and other movements of the arm, and effectively reduce the muscle load of the operator's shoulder and elbow during the operation of the unmanned aerial vehicle; (2) The shoulder support assembly of the exoskeleton body cooperates with the arm assembly to provide stable support for the upper limb and reduce the shaking when the arm is suspended during operation. At the same time, the assistance function reduces the strength of the continuous muscle force and delays the occurrence of muscle fatigue, so that the operator can maintain accurate operation for a long time, improve the stability of the unmanned aerial vehicle operation and the persistence of the task execution, and is especially suitable for scenes such as power inspection and aerial photography that require long-time operation; (3) The arm assembly is connected to the shoulder support assembly through the first flexible connecting piece. The flexible connection mode can not only transmit assistance, but also will not excessively limit the natural movement range of the human upper limb. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 The overall structure schematic diagram of the unmanned aerial vehicle operation upper limb assisting exoskeleton provided for the present application embodiment 1 is shown in the figure;

[0030] Figure 2 The control logic schematic diagram of the main control device provided for the present application embodiment 5 is shown in the figure;

[0031] The figure mark: 100-exoskeleton body, 110-shoulder support assembly, 111-rotating winding mechanism, 120-arm assembly, 121-first fixed part, 122-second fixed part, 123-driving member, 130-first flexible connecting piece, 131-first connecting piece, 132-second connecting piece, 140-waist assembly, 150-second flexible connecting piece, 151-third connecting piece. DETAILED DESCRIPTION

[0032] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0033] Embodiment 1

[0034] The present embodiment provides a UAV-operated upper limb assisting exoskeleton, Figure 1 The overall structure of the UAV-operated upper limb assisting exoskeleton is shown in the figure. Figure 1 As shown in the figure, the UAV-operated upper limb assisting exoskeleton comprises an exoskeleton body 100, a main control device, a power module and an electromyographic sensor.

[0035] The exoskeleton body 100 is a mechanical structure for supporting and moving, which is composed of a shoulder support assembly 110 adapted to the human shoulder and back and an arm assembly 120 adapted to the human arm. The shoulder support assembly 110 can provide basic support for the overall structure. Each of the two sides of the front of the shoulder support assembly 110 is provided with an arm assembly 120, and the arm assembly 120 is connected to the shoulder support assembly 110 through a first flexible connecting piece 130, so that the exoskeleton can provide assistance to the user's arms at the same time. In addition, the arm assembly 120 is connected to the shoulder support assembly 110 through the first flexible connecting piece 130. The flexible connection can not only transmit the assistance, but also not excessively limit the natural movement range of the human upper limb, thereby improving the flexibility of the exoskeleton.

[0036] The shoulder support assembly 110 and the arm assembly 120 cooperate with each other to provide stable support for the upper limb, reduce the shaking when the arm is suspended for control, and at the same time, the assistance function reduces the strength of the continuous muscle force, delays the generation of muscle fatigue, so that the operator can maintain precise operation for a long time, improves the stability of the UAV control and the persistence of the task execution, and is especially suitable for scenes such as power inspection and aerial photography which require long-time operation.

[0037] Regarding signal acquisition and input, the electromyographic sensor is arranged on the arm assembly 120 to detect the electromyographic signal of the upper limb in real time. The signal output end of the electromyographic sensor is connected to the first signal input end of the main control device to send the sensed electromyographic signal to the main control device. Further, the second signal input end of the main control device is connected to the signal output end of the UAV remote controller to receive the rocker thrust signal sent by the UAV remote controller, and then the operation intention of the user can be analyzed according to the electromyographic signal and the rocker thrust signal. Compared with a single signal, the fusion of multiple information can effectively improve the recognition accuracy of the operation intention, thereby laying a foundation for the assistance control of the exoskeleton.

[0038] Regarding the control of the actuator, the arm assembly 120 in this embodiment includes a first fixing part 121 adapted to the upper arm and a second fixing part 122 adapted to the forearm. The first fixing part 121 and the second fixing part 122 are connected by a drive member 123. The signal input terminal of the drive member 123 is connected to the first signal output terminal of the main control device to receive the first control signal sent by the main control device and drive the first fixing part 121 and the second fixing part 122 to rotate relative to each other, assisting the elbow in completing the flexion and extension action and providing additional torque.

[0039] The first flexible connector 130 includes a first connector 131. The shoulder support assembly 110 is provided with a rotating winding mechanism 111. The first end of the first connector 131 is connected to the second fixing part 122. The second end of the first connector 131 is wound around the winding wheel of the rotating winding mechanism 111. The signal input end of the rotating winding mechanism 111 is connected to the second signal output end of the main control device to receive the second control signal sent by the main control device and drive the first connector 131 to wind around the winding wheel or release from the winding wheel.

[0040] When the first connector 131 is tightened during winding, it provides upward and backward pulling force to the forearm, assisting in the arm-raising action; when released, the first connector 131 is relaxed, allowing the arm to hang down and extend forward naturally. This enables the exoskeleton to coordinate with the movement of the human upper limb, providing targeted assistance for the flexion, extension, and swinging movements of the arm, effectively reducing the muscle load on the shoulder and elbow of the operator during the operation of the drone.

[0041] In addition, the main control device, electromyography sensor, rotating winding mechanism 111 and drive unit 123 are all electrically connected to the power module and are powered by the power module.

[0042] Example 2

[0043] This embodiment further explains the structure of the first flexible connector 130 based on the technical solution provided in Embodiment 1:

[0044] In this embodiment, the first flexible connector 130 also includes a second connector 132, which is used to fix and support the arm assembly 120, and directly transfer the upper arm load to the shoulder support assembly 110, so as to avoid the load being borne entirely by the shoulder muscles and realize segmented load unloading.

[0045] Specifically, the first end of the second connecting member 132 is connected with the first fixing part 121, and the second end of the second connecting member 132 is connected with the shoulder support assembly 110; when the operator lifts the arm, the first connecting member 131 is tightened to provide a pulling force through the rotating winding mechanism 111, and the second connecting member 132 synchronously bears part of the upper arm gravity, thereby reducing the bearing load of the shoulder muscle; when the arm naturally droops, the second connecting member 132 remains in a relaxed state and does not limit the natural activity of the upper arm, and only provides elastic support at the moment of exertion.

[0046] Embodiment 3

[0047] This embodiment is based on the technical solution provided in Embodiment 1, and the structure of the exoskeleton body 100 is further described:

[0048] In this embodiment, the exoskeleton body 100 further comprises a waist assembly 140 adapted to the human waist and a support frame arranged on the front of the waist assembly 140; the waist assembly 140 is annular or semi-annular and is made of lightweight material, and a flexible gasket such as sponge or breathable fabric is arranged on the inner side to adapt to different waist sizes and improve the wearing comfort.

[0049] Further, the waist assembly 140 is connected with the shoulder support assembly 110 through a second flexible connecting member 150 for transmitting the load of the upper body to the waist to share the shoulder pressure.

[0050] The second flexible connecting member 150 is configured to be adjustable in length, for example, by a day buckle or the like to realize length adjustment, so as to adapt to users of different heights and shoulder widths or adjust the tightness according to the operation posture.

[0051] In a preferred embodiment of this embodiment, the first flexible connecting member 130 and the second flexible connecting member 150 are both made of nylon webbing, which can provide mechanical support while realizing lightweight and reducing the wearing burden, and does not interfere with the natural movement of the human body.

[0052] The lower end of the support frame is connected with the front of the waist assembly 140, and the upper end of the support frame is provided with a placement table adapted to the remote controller of the unmanned aerial vehicle.

[0053] The support frame is made of lightweight rigid material, for example, aluminum alloy or high-strength plastic, and the connection point of the lower end with the front of the waist assembly 140 is close to the waist position to ensure the stability of the support; the placement table of the upper end is designed to be adjustable in size to adapt to remote controllers of different models. The height of the placement table is adapted to the natural operation posture of the operator, so that when the forearm is placed thereon, the elbow is in a natural bending state, reducing the muscle load of the suspended control; in some embodiments, the height of the support frame is adjustable, and the specific adjustment mode is not limited here, and a telescopic sleeve or the like structure can be selected to realize the telescopic adjustment function.

[0054] Specifically, the embodiment transfers the weight of the remote controller from the hand to the waist assembly 140 through the support frame, uses the load-bearing capacity of the waist and torso to replace the continuous lifting of the arm, and provides a stable support surface for hand operation; on the one hand, the load of the wrist and forearm muscles (such as the radial wrist flexor and brachioradialis muscle) can be reduced, effectively relieving wrist fatigue caused by long-time holding; on the other hand, the fixed placement table makes the position of the remote controller relatively stable, reduces the risk of accidental touch caused by hand shaking during operation, and frees up some hand power, so that the operator can focus more on action control rather than load-bearing. In addition, the combination of the support frame and the waist assembly 140 can also form a complete support system with the shoulder support assembly 110 and the arm assembly 120, further optimizing the whole body load distribution and improving the practicality of the device in long-time operation scenarios.

[0055] In some embodiments, the lower end of the support frame is hinged to the front of the waist assembly 140, so that the support frame can be adjusted in angle around the hinge point, and then the inclination angle can be flexibly adjusted according to the operating posture of the operator, such as body straightening in standing posture and body leaning forward in bending posture, to ensure that the upper end placement table is always in a position convenient for hand operation, avoiding the inconvenience caused by changes in body posture.

[0056] The second flexible connecting piece 150 includes a third connecting piece 151 arranged on the front of the shoulder support assembly 110 and a fourth connecting piece arranged on the back of the shoulder support assembly 110; wherein the third connecting piece 151 serves as a front connecting structure and adopts a three-point structure to form a stable triangular support through three connecting points; the upper two ends of the three-point structure are respectively connected to the two sides of the front of the shoulder support assembly 110, corresponding to the shoulder positions of the human body, and the lower end of the three-point structure is connected to the front of the waist assembly 140, and the three-point line forms an inverted triangle, which can uniformly transmit the load on the front of the shoulder support assembly 110 to the waist.

[0057] Specifically, the third and fourth connecting pieces in front of and behind each other cooperate to form a flexible wrapping around the torso, enhancing the fit of the exoskeleton with the human body; on the one hand, this can improve the stability of the shoulder support assembly 110 and reduce left and right shaking during operation; on the other hand, by uniformly distributing the pressure, the discomfort caused by excessive local pressure on the waist can be avoided, and the tolerance for long-time wearing can be improved; in addition, the three-point structure can adapt to different body types, and by adjusting the length of the nylon webbing, it can adapt to different shoulder widths and waist and abdomen dimensions, further enhancing the universality of the exoskeleton.

[0058] Embodiment 4

[0059] Based on the technical solutions provided in Embodiment 1, the structure of the arm assembly 120 is further described in this embodiment:

[0060] In the embodiment, the first fixing part 121 comprises an upper arm module and a first arm ring connected with the upper arm module, and the second fixing part 122 comprises a forearm module and a second arm ring connected with the forearm module, and the electromyographic sensor is arranged on the inner side of the first arm ring and / or the second arm ring.

[0061] The upper arm module and the forearm module are support structures corresponding to the bone and muscle parts of the wrapped upper arm and forearm, respectively, and the first arm ring and the second arm ring are ring-shaped or semi-ring-shaped binding structures, and the inner side is provided with a flexible pad, and the tightness is adjusted by a magic tape or a buckle to fit the limb, so as to ensure that the fixing part is stably connected with the arm and does not affect the blood circulation. The electromyographic sensor is arranged on the inner side of the arm ring, which can directly contact the skin of the upper arm (such as the biceps brachii and the triceps brachii) and / or the forearm (such as the extensor carpi radialis longus and the flexor carpi ulnaris), and accurately collect the weak electrical signals generated during muscle activity.

[0062] Specifically, the embodiment can ensure stable contact between the electromyographic sensor and the skin through the close fitting of the arm ring, reduce interference in the signal collection process, and thus improve the accuracy and stability of the electromyographic signal, and provide reliable data basis for the subsequent master control device to identify the operation intention.

[0063] The upper arm module and the forearm module are configured to be adjustable in length. The upper arm module, as a structure for supporting the upper arm in the first fixing part 121, can be adjusted in length by means of a telescopic sleeve, a multi-section buckle or a sliding rail, so as to adapt to the length difference of the upper arms of different users; the forearm module, as a structure for supporting the forearm in the second fixing part 122, also adopts a similar telescopic adjustment structure to match the lengths of the forearms of different users, thereby further improving the universality of the exoskeleton.

[0064] Embodiment 5

[0065] The embodiment is based on any one of the technical solutions provided in embodiments 1 to 4, and further illustrates the design of the master control device.

[0066] Referring to Figure 2 In the embodiment, the master control device comprises:

[0067] a data receiving module for receiving electromyographic signals and joystick thrust signals;

[0068] a preprocessing module for preprocessing and feature extraction of the electromyographic signals to obtain an electromyographic feature vector, and preprocessing and feature extraction of the joystick thrust signals to obtain a thrust feature vector;

[0069] a prediction module for performing fusion processing on the electromyographic feature vector and the thrust feature vector by using a multi-modal information fusion algorithm to identify the operation intention;

[0070] A strategy generation module determines an exoskeleton control strategy based on the operation intention;

[0071] An execution module executes control based on the exoskeleton control strategy.

[0072] Thus, a complete process from signal acquisition and processing to final control is realized, so that the exoskeleton can accurately respond to the operation intention of the operator, provide more demand-oriented assistance, and further improve the stability and efficiency of the unmanned aerial vehicle control.

[0073] Embodiment 6

[0074] This embodiment further describes the specific means for realizing the functions of each functional module based on the technical solutions provided in Embodiment 5.

[0075] Regarding the preprocessing module, the preprocessing and feature extraction of the electromyographic signal include the following specific steps:

[0076] Step S01, noise reduction processing:

[0077] The moving average filtering algorithm is used to eliminate high-frequency noise.

[0078] Step S02, threshold decomposition:

[0079] The threshold is used to distinguish the motion segment (when operating) and the non-motion segment (when stationary).

[0080] Step S03, performing feature extraction on the electromyographic signal after threshold decomposition:

[0081] Time domain features are extracted, and the time domain features extracted in this embodiment include absolute mean, variance, and root mean square, which are not described in detail here.

[0082] The preprocessing and feature extraction of the joystick thrust signal include the following specific steps:

[0083] Step S04, noise reduction and signal separation processing:

[0084] The thrust signal contains operation jitter noise, and the singular value decomposition is used in this embodiment to separate the effective features and noise.

[0085] Step S05, performing feature extraction on the joystick thrust signal after noise reduction and signal separation processing:

[0086] The time domain features extracted in this embodiment include thrust peak value, thrust rate of change, thrust duration, and thrust integral, etc. The thrust peak value is the maximum value of the X direction or Y direction thrust in a single operation period, for example, a single push rod action, which is used to reflect the size of the operation force. The larger the peak value, the more intense the intention. The thrust rate of change is the rate of change of the thrust with time, that is, the first derivative of the thrust signal, which is used to reflect the speed of the operation. The larger the thrust rate of change, the more rapid adjustment is needed, for example, the urgent push when avoiding obstacles. The smaller the thrust rate of change, the more fine tuning is needed. The thrust duration is the duration of the thrust value exceeding the threshold of the rocker activation, which is used to reflect the persistence of the operation. The longer the duration, the more stable adjustment is needed, for example, maintaining the heading when cruising. The shorter the duration, the more instantaneous correction is needed, for example, eliminating small deviations. The thrust integral is the area surrounded by the rocker thrust signal and the time axis in a single operation period, which is used to reflect the force and time of the rocker thrust, and distinguish between fine tuning and coarse tuning.

[0087] Regarding the prediction module, a multi-modal information fusion algorithm is used to fuse the myoelectric feature vector and the thrust feature vector to identify the operation intention, including the following specific steps:

[0088] Step S11, signal similarity ratio calculation, to quantify the correlation between the myoelectric signal and the thrust signal.

[0089] Step S12, set a decision threshold. When the signal similarity ratio is greater than the decision threshold, use the SVM algorithm for identification. When the signal similarity ratio is less than the decision threshold, use the FCM algorithm for clustering. No specific details are given here.

[0090] Step S13, fuse the results output by SVM and FCM to obtain a fused feature vector.

[0091] Step S14, based on the fused feature vector, confirm the operation intention.

[0092] In this embodiment, the operation intention is divided into three types, namely static maintenance, fine tuning, and rapid coarse tuning, and the specific identification is as follows:

[0093] When the muscle force intention in the fused feature vector and the thrust embodiment are at a low level, the thrust peak value is small and the thrust rate of change is close to 0, it is determined as a static arm-lifting scene of the UAV cruising / hovering, and the operation intention is determined as static maintenance.

[0094] When the muscle force intention in the fused feature vector and the thrust embodiment are at a medium level, the thrust peak value is medium and the thrust rate of change is a slow change of the thrust, it is determined as a fine adjustment scene of the UAV attitude, and the operation intention is determined as fine tuning.

[0095] When the muscle force intention in the fusion feature vector and the thrust force are both high values, the thrust peak is large, and the thrust rate of change is fast, it is determined that the unmanned aerial vehicle is in an emergency obstacle avoidance or other fast operation scene, and the operation intention is determined to be fast and rough adjustment.

[0096] Regarding the strategy generation module, based on the operation intention, the exoskeleton control strategy is determined, and the details are as follows:

[0097] For static maintenance, the control strategy is specifically: the driving part 123 outputs a constant low torque to maintain the relative angle of the first fixed part 121 and the second fixed part 122 stable, reducing the load of the elbow muscle sustained force; the rotating winding mechanism 111 applies a constant pre-tightening force to the first connecting part 131, providing an upward supporting tension to the forearm through a slight tightening state, offsetting part of the forearm gravity, and reducing the static load of the shoulder lifting muscle.

[0098] For fine adjustment, the control strategy is specifically: the driving part 123 outputs a dynamically matched low torque according to the slight change of the electromyographic signal, assisting the elbow to complete small angle rotation, providing assistance without interfering with the accuracy of fine action; the rotating winding mechanism 111 dynamically releases or slightly tightens the first connecting part 131 according to the forearm movement direction, slowly releases the connecting part when the forearm is fine-tuned forward, and slightly tightens to provide auxiliary tension when the forearm is fine-tuned backward.

[0099] For fast and rough adjustment, the control strategy is specifically: the driving part 123 outputs high torque according to the strong force characteristics of the electromyographic signal, and instantaneously outputs high torque to drive the first fixed part 121 and the second fixed part 122 to quickly complete large angle relative rotation, assisting the elbow to quickly flex and extend, providing additional power for arm movement, and improving operation speed; the rotating winding mechanism 111 quickly tightens or releases the first connecting part 131 according to the movement direction, when the arm needs to be quickly lifted, the winding wheel rotates at high speed to tighten the connecting part, providing strong upward tension; when the arm needs to be quickly lowered, the winding wheel quickly releases the connecting part, reducing the action resistance, and cooperating with the fast force of the arm.

[0100] Through the above differentiated strategy, the exoskeleton can accurately match different operation intentions, reduce muscle load, and ensure the stability and flexibility of operation.

[0101] The above is only a preferred embodiment of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A UAV-operated upper-limb-assistance exoskeleton, characterized by, The exoskeleton body (100) is composed of a shoulder support assembly (110) adapted to the human shoulder and back and an arm assembly (120) adapted to the human arm, two arm assemblies (120) are arranged on the front side of the shoulder support assembly (110), and the arm assemblies (120) are connected to the shoulder support assembly (110) through first flexible connectors (130). The arm assembly (120) is provided with an electromyographic sensor, the signal output end of the electromyographic sensor is connected to the first signal input end of the master control device, so as to send the sensed electromyographic signal to the master control device, and the second signal input end of the master control device is connected to the signal output end of the unmanned aerial vehicle remote controller, so as to receive the rocker push signal sent by the unmanned aerial vehicle remote controller. The arm assembly (120) includes a first fixed part (121) adapted to the upper arm and a second fixed part (122) adapted to the forearm, the first fixed part (121) and the second fixed part (122) are connected through a driving member (123), the signal input end of the driving member (123) is connected to the first signal output end of the master control device, so as to receive the first control signal sent by the master control device and drive the first fixed part (121) and the second fixed part (122) to rotate relative to each other. The first flexible connector (130) includes a first connector (131), the shoulder support assembly (110) is provided with a rotary winding mechanism (111), the first end of the first connector (131) is connected to the second fixed part (122), and the second end of the first connector (131) is wound around the winding wheel of the rotary winding mechanism (111), the signal input end of the rotary winding mechanism (111) is connected to the second signal output end of the master control device, so as to receive the second control signal sent by the master control device and drive the first connector (131) to be wound on the winding wheel or released from the winding wheel. The master control device, the electromyographic sensor, the rotary winding mechanism (111) and the driving member (123) are electrically connected to the power module. The master control device includes: A data receiving module for receiving the electromyographic signal and the rocker push signal; A preprocessing module for preprocessing and feature extraction of the electromyographic signal to obtain an electromyographic feature vector, and preprocessing and feature extraction of the rocker push signal to obtain a push force feature vector; A prediction module for using a multi-modal information fusion algorithm to fuse and process the electromyographic feature vector and the push force feature vector to identify an operation intention; A strategy generation module for determining an exoskeleton control strategy based on the operation intention; An execution module for executing control based on the exoskeleton control strategy. The first flexible connector (130) further includes a second connector (132), the first end of the second connector (132) is connected to the first fixed part (121), and the second end of the second connector (132) is connected to the shoulder support assembly (110).

2. The upper-limb exoskeleton for assisting the operation of a UAV according to claim 1, wherein, ​ 3. The upper-limb exoskeleton for assisting the operation of a UAV according to claim 1, wherein The waist assembly (140) is connected to the shoulder support assembly (110) by a second flexible connecting member (150), and the second flexible connecting member (150) is configured to be adjustable in length.

4. The upper-limb exoskeleton for assisting the operation of a UAV according to claim 3, wherein The first flexible connecting member (130) and the second flexible connecting member (150) are both made of nylon webbing.

5. The upper-limb exoskeleton for assisting the operation of a UAV according to claim 3, wherein, The support frame is provided on the front surface of the waist assembly (140), and the lower end of the support frame is connected to the front surface of the waist assembly (140), and the upper end of the support frame is provided with a placement table adapted to the remote controller of the unmanned aerial vehicle.

6. The upper limb exoskeleton for assisting the operation of UAV according to claim 5, wherein, The lower end of the support frame is hingedly connected to the front surface of the waist assembly (140).

7. The upper limb exoskeleton for assisting the operation of UAV according to claim 4, wherein, The second flexible connecting member (150) comprises a third connecting member (151) provided on the front surface of the shoulder support assembly (110) and a fourth connecting member provided on the back surface of the shoulder support assembly (110), the third connecting member (151) is a three-point structure, the upper two ends of the three-point structure are respectively connected to the two sides of the front surface of the shoulder support assembly (110), and the lower end of the three-point structure is connected to the front surface of the waist assembly (140).

8. The upper-limb exoskeleton for assisting the operation of a UAV according to claim 1, wherein, The first fixed part (121) comprises an upper arm module and a first arm ring connected to the upper arm module, and the second fixed part (122) comprises a forearm module and a second arm ring connected to the forearm module, and the electromyographic sensor is arranged on the inner side of the first arm ring and / or the second arm ring.

9. The upper-limb exoskeleton for assisting the operation of a UAV according to claim 8, wherein, The upper arm module and the forearm module are configured to be adjustable in length.

Images