Underwater human body motion intention detection equipment

Abstract

The invention relates to the technical field of human body recognition, and discloses underwater human body motion intention detection equipment. The underwater human body motion intention detection equipment comprises a master station unit and a plurality of slave station units, the slave station units are in communication connection in sequence, and the former slave station unit and the latter slave station unit are in duplex communication; the foremost slave station unit in the slave station units is in duplex communication with the master station unit, and the slave station units are attached to all parts of a human body; the slave station unit comprises a network communication module and a sensing detection module, the sensing detection module comprises an I MU attitude detection module and a tension and pressure detection module, and the I MU attitude detection module communicates with the network communication module and is connected with the tension and pressure detection module. The system can solve the problem that the motion intention of a human body is difficult to detect underwater, a master-slave structure is designed, each slave station unit is responsible for combining the previous data connected with the slave station unit in series with the node data of the slave station unit and sending the combined data to the next node connected with the slave station unit in series, and by means of the design, new nodes can be added at any time.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an underwater human motion intention detection device. Background Art

[0002] An ocean exoskeleton robot is an innovative wearable robot device for ocean operations, which is used to enhance the diver's motion ability and operation efficiency underwater and is applicable to various underwater tasks in the fields of submarine rescue, sunken ship salvage, offshore oil and gas pipeline construction, ocean civil engineering industry, as well as ocean science and military applications.

[0003] The research and development and application of ocean exoskeleton robots have important requirements and needs in terms of social needs and technological development. The ocean is one of the most challenging engineering fields on the earth, which involves multi-disciplinary, multi-field, and multi-level technical problems and is of great significance to human technological innovation.

[0004] However, the variability and uncertainty of the ocean environment have brought great difficulties to technology research and development and application. Traditional underwater technology methods, such as remotely operated vehicles and manned submersibles, all have problems such as poor reliability, poor adaptability, and poor intelligence. The ocean exoskeleton robot combines human intelligence and the high load and motion ability of robot technology, which can improve the reliability and adaptability of underwater technology and provide a new platform for ocean technological innovation. Furthermore, when using underwater exoskeleton equipment, how to detect information such as the diver's motion state and human-machine interaction force underwater, so as to realize the perception and understanding of the diver's actions and intentions, is one of the key technical problems to be solved urgently. Summary of the Invention

[0005] Based on this, in view of the technical problem of the poor detection effect of underwater human motion intention in the prior art, an underwater human motion intention detection device is proposed. The underwater human motion intention detection device includes: a master station unit and a plurality of slave station units;

[0006] Each of the slave station units is communicatively connected in sequence, wherein, the previous slave station unit communicates with the next slave station unit in duplex;

[0007] The slave station unit at the forefront position among each of the slave station units communicates with the master station unit in duplex, and each slave station unit is attached to each part of the human body;

[0008] The slave station unit includes a network communication module and a sensing detection module, wherein, the sensing detection module includes an IMU attitude detection module and a tensile and compressive force detection module, and the IMU attitude detection module communicates with the network communication module and is connected to the tensile and compressive force detection module respectively.

[0009] The underwater human motion intention detection device proposed by the present invention includes a master station unit and multiple slave station units; each of the slave station units is communicatively connected in sequence, where the previous slave station unit communicates with the next slave station unit in duplex; the slave station unit at the frontmost position among each of the slave station units communicates with the master station unit in duplex, and each slave station unit is attached to each part of the human body; the slave station unit includes a network communication module and a sensing detection module, where the sensing detection module includes an IMU attitude detection module and a tensile and compressive force detection module, and the IMU attitude detection module communicates with the network communication module and is connected to the tensile and compressive force detection module respectively. The present invention can solve the dilemma of difficult detection of human motion intention underwater and establish an underwater human-machine interaction interface. To enhance scalability and facilitate the intention recognition requirements of various underwater operations, it is designed as a master-slave structure. Each sensing node is responsible for combining the data of the previous node in series with its own node data and sending it to the next sensing node in series. When the last node finds that there is no next node in series, it sends the data back to the previous node, and the previous node continues to send it back after receiving the data until it returns to the master station unit to complete the data transmission. Such a design can add new nodes at the middle or end of the bus at any time. Description of the Drawings

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0011] Among them:

[0012] Figure 1 It is a schematic diagram of a module of the underwater human motion intention detection device in an embodiment;

[0013] Figure 2 It is a flowchart of the underwater human motion intention detection device in an embodiment;

[0014] Figure 3 It is a schematic diagram of delay calculation of the underwater human motion intention detection device in an embodiment;

[0015] Figure 4 It is a schematic diagram of communication time synchronization of the underwater human motion intention detection device in an embodiment;

[0016] Figure 5 It is a schematic diagram of human motion intention recognition of the underwater human motion intention detection device in another embodiment. Detailed Embodiments

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs; the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above drawings are intended to cover non-exclusive inclusion. The terms "first", "second", etc. in the specification and claims of this application or the above drawings are used to distinguish different objects and not to describe a specific order.

[0018] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of this application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Please refer to Figure 1 as shown in Figure 1 a schematic diagram of a module of an underwater human motion intention detection device provided by an embodiment of the present invention. The underwater human motion intention detection device includes a master station unit and a plurality of slave station units;

[0021] Each of the slave station units is communicatively connected in sequence. Among them, the previous slave station unit communicates with the next slave station unit in duplex;

[0022] The slave station unit at the frontmost position among each of the slave station units communicates with the master station unit in duplex, and each slave station unit is attached to each part of the human body;

[0023] The slave station unit includes a network communication module and a sensing and detection module. Among them, the sensing and detection module includes an IMU attitude detection module and a tensile and compressive force detection module. The IMU attitude detection module communicates with the network communication module and is connected to the tensile and compressive force detection module respectively.

[0024] In this embodiment, refer to Figure 1, as an example, the slave units include slave unit 1, slave unit 2, and slave unit 3. The output of slave unit 1 is connected to the input of slave unit 2, the output of slave unit 2 is connected to the input of slave unit 3, and the input of slave unit 1 is connected to the output of the master unit. The master unit can use a single-chip microcomputer chip. The slave units communicate in duplex, that is, the slave units can transmit data to each other. The IMU attitude detection module is used to collect IMU data, and the tensile and compressive force detection module is used to detect pressure. The network communication module is used to obtain the data collected by the sensing detection module. The network communication module of one slave unit can transfer the data to the network communication module of the next slave unit. Each slave unit also includes a power module for supplying power to the network communication module and the sensing detection module.

[0025] As an example, the master unit and the sensing slave units are interconnected through full-duplex transmission cables. All data packets are sent from the master station to slave station 1 serially through the full-duplex data cable. After loading the data of slave station 1 into the packet, it is continued to be sent to slave station 2. After loading the packet of slave station 2, it is continued to be sent to the next slave unit until the data is returned to the master unit through the uplink after the last slave node. Therefore, when adding sensors, only the sensing unit needs to be added to the link, achieving high-efficiency scalability. The operation flow chart of the underwater human motion intention detection device of the present invention is as Figure 2 shown, and mainly consists of three parts: system initialization, sending time calibration frames, and sending periodic data acquisition frames. The process of system initialization is mainly the initialization of the functions of the master station and the self-configuration of each slave node, such as reading the configuration data of the sensing module. The main purpose of sending time calibration frames is to solve the problem of asynchronous data acquisition of sequentially connected sensors when transmitting data serially. The process of data synchronization is mainly divided into two main steps. First, after the system is powered on and started, the master station sends a time frame. After receiving this time frame, the slave station records its own local time tick count (tick) when it receives this time frame. After the master station times for 1 second, it sends out the second time frame. After receiving this time frame, the slave station records its own current local time tick count. Therefore, after the slave station subtracts its own two tick counts, the transmission delay error can be eliminated, and the difference in the counting units between the slave station and the master station can be obtained. At this time, each slave station can know the error between its own tick and the clock tick count of the master station, and the clock counting error can be realized. The overall process is as Figure 3 shown. The process of communication time synchronization is as Figure 4As shown, the master station first sends the first time frame T0. When the master station receives the time frame again after it has passed through all nodes, the time is T′0. After receiving the time frame, Slave 1 counts a fixed interval Δ according to the calibrated ticks before, and then sends its own time frame T1. Subsequent slave stations all count an interval multiplied by Δ according to their own numbers and send time frames T2, T3, etc. Eventually, the master station will receive a series of time frames T′1, T′2, T′3, etc. The master station records the local ticks when it receives the time frames. Then, the communication time from the master station to Slave 1 is T′0 - T′1 - Δ. When updating data, Slave 1 uploads the data before the time of T′0 - T′1 - Δ. Subsequent slave stations perform compensation according to the same principle. After compensation, the data received by the master station will be collected at the moment when the master station sends the time frame, achieving the data time synchronization of all sensors.

[0026] In one embodiment, the master station unit is used to obtain the IMU data transmitted from the slave station unit and perform attitude calibration based on the IMU data.

[0027] In one embodiment, the master station unit is further used to extract target feature components from the IMU data after attitude calibration and perform human motion intention recognition based on the target feature components. For example, by inputting the target feature components into an adaptive oscillator to obtain the phase and frequency, and detecting the human motion intention based on the phase and frequency.

[0028] In one embodiment, the steps for the master station unit to further extract target feature components from the IMU data after attitude calibration include:

[0029] Taking the projection angle of the human waist on the horizontal plane as the first component;

[0030] Calculating the first projection angles of the waist with the left and right legs on the coronal plane respectively, and then taking the maximum value of the two first projection angles as the second component;

[0031] Calculating the projection angle of the waist on the sagittal plane and the second projection angles of the left and right legs on the sagittal plane, and then taking the sum of the two second projection angles as the third component;

[0032] The first component, the second component, and the second component respectively enter a sliding integration window with a length of 1000 data to obtain the integral values corresponding to the first component, the second component, and the third component respectively. The component corresponding to the maximum integral value among the integral values is used as the target feature component.

[0033] As an example, referring to Figure 5, the human motion intention recognition part of the present invention is processed in the main station unit and mainly consists of three parts. The first part is attitude calibration; the second part is feature extraction; the third part is the Adaptive Oscillator (AO). The attitude calibration involves the following coordinate systems, all of which follow the right-hand rule of the Cartesian coordinate system: the motion capture system coordinate system (d), the human body local coordinate system (l), the human body limb coordinate system (b), the sensor coordinate system (s), the initial attitude coordinate system (o), and the ground coordinate system (e). The assumptions followed by the attitude calibration of the present invention are as follows: First, the origin of the human body local coordinate system coincides with the origin of the coordinate system of the waist sensor; Second, the limb coordinate system coincides with the sensor coordinate system; Third, during the attitude calibration process, the alignment procedure of the ground coordinate system is omitted. The following attitude calibrations are all based on quaternion calculations, that is, the form of the original data obtained from the IMU is a quaternion. The motion capture system, which is the visualization host computer software for the present invention to complete the verification of attitude calibration and human body modeling, does not belong to the content of the present invention. Through the following formula, based on the coordinate system transformation of quaternion operations, the attitude calibration is completed (it is a 3×4 attitude matrix, representing the coordinates of the coordinate axes of the coordinate system b in the coordinate system d). In the following formula, represents quaternion multiplication, q * represents the conjugate quaternion of the quaternion q, and the quaternion q is obtained in the IMU attitude detection module. represents the attitude transformation of the unit vector from the coordinate system j to the coordinate system i. represents the coordinates of the three axes of the human body limb coordinate system b in the motion capture coordinate system d, and W d is the representation of the attitude of each IMU after correction in the motion capture system.

[0034]

[0035] Since the determination of the coordinate system has been achieved during the attitude calibration process, when reconstructing the human body model, there is no need to use the traditional S-DH or M-DH modeling method to determine the coordinate system again. At this time, only the rotation amount and offset amount need to be concerned, and the attitude solution can be completed by applying the homogeneous transformation matrix. can be converted from the quaternion q, V represents the limb vector in the limb coordinate system, and P A represents the joint coordinate in the coordinate system A, and P B represents the joint coordinate in the coordinate system B. Thus, the attitude calibration and human body model reconstruction are completed.

[0036]

[0037] The feature extraction part proposes a principle for extracting multi-plane motion features: The combined feature extraction method of the present invention can be composed of n feature components, and the theoretical amplitude range of the i-th feature component is denoted as The actual amplitude of the i-th characteristic component is denoted as The overall theoretical amplitude range is denoted as If the indexes of each characteristic component are equivalent, that is, the theoretical amplitude ranges of each characteristic component are equivalent, then it can be considered that this feature extraction method is relatively reasonable. Thus, the index can be taken as the normalized characteristic component. The above is the reference principle, and the specific multi-motion plane feature extraction method is not unique. The specific method of the present invention is as follows: The first component is the projection angle of the waist on the horizontal plane; the second component is to first find the projection angles of the waist with the left leg and the right leg on the coronal plane respectively, and then take the maximum value among them; the third component is to first find the sum of the projection angle of the waist on the sagittal plane and the projection angles of the left and right legs on the sagittal plane. After obtaining the three characteristic components, they will respectively enter a sliding integration window with 1000 data points, compare the magnitudes of the output values of the three, and identify the plane mapped by the number of the largest one as the main motion plane (1 for the horizontal plane, 2 for the coronal plane, 3 for the sagittal plane), and take the characteristic component it represents as the input of the next part AO. The AO part can give the estimated phase, frequency and signal of the input signal.

[0038] It mainly includes a master station unit, a full-duplex transmission cable, a slave station unit (including a network communication module and a sensing detection module), a waterproof and sealed housing, and a waterproof connector. Among them, the master station unit provides the heartbeat time of the device and the time calibration function of the bus. Data is transmitted between the master station and the slave station through the full-duplex cable. First, the master station unit sends a data synchronization frame to the slave station system through the downlink cable. After receiving the master station synchronization frame, the slave station unit inserts its own system time into the data synchronization frame and then continues to send it to the next slave station unit until the last unit, and then returns it to the master station unit through the uplink cable. After the master station unit calculates the communication time of each slave station unit, it will send the compensation amount to each slave station unit in the same way, and the slave station unit will sample the sensing information according to the agreed time. After establishing the time synchronization mechanism, the master station unit will send the acquisition data frame to the first cascaded node at a fixed acquisition frequency, and collect the sensing information of all slave station units in a similar way. In terms of structure, a sealed and waterproof housing is designed, and a waterproof plug is used to connect the sensor and other sub-modules. In terms of human motion intention recognition, through three parts: attitude calibration, feature extraction and adaptive oscillator (AO), the motion recognition of multiple types of underwater motions based on multiple motion planes is completed.

[0039] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An underwater human motion intention detection device, characterized in that, The underwater human motion intention detection device includes: a master station unit and multiple slave station units; Each of the slave station units is sequentially connected for communication in order, wherein, the previous slave station unit and the next slave station unit perform duplex communication; The slave station unit at the foremost position among each of the slave station units performs duplex communication with the master station unit, and each slave station unit is attached to each part of the human body; The slave station unit includes a network communication module and a sensing detection module, wherein, the sensing detection module includes an IMU attitude detection module and a tensile and compressive force detection module, and the IMU attitude detection module is respectively in communication with the network communication module and connected to the tensile and compressive force detection module.

2. The underwater human motion intention detection device according to claim 1, characterized in that, The master station unit is used to obtain the IMU data transmitted by the slave station unit and perform attitude calibration based on the IMU data.

3. The underwater human motion intention detection device according to claim 2, characterized in that, The master station unit is further used to extract target feature components from the IMU data after attitude calibration and perform human motion intention recognition based on the target feature components.

4. The underwater human motion intention detection device according to claim 3, characterized in that, The master station unit is further used to extract target feature components from the IMU data after attitude calibration, and the steps include: Taking the projection angle of the human waist on the horizontal plane as the first component; Calculating the first projection angles of the waist with the left leg and the right leg on the coronal plane respectively, and then taking the maximum value of the two first projection angles as the second component; Calculating the projection angle of the waist on the sagittal plane and the second projection angles of the left and right legs on the sagittal plane, and then taking the sum of the two second projection angles as the third component; The first component, the second component, and the second component respectively enter a sliding integration window with a length of 1000 data to obtain the integral values corresponding to the first component, the second component, and the third component respectively, and the component corresponding to the maximum integral value among the integral values is used as the target feature component.

Images