A robot fault diagnosis system

Abstract

The present invention provides a robot fault diagnosis system, comprising: a data acquisition device disposed on the robot body for acquiring the current and position signals of the device, and acquiring the acceleration and pose of the end effector; for receiving the data signals acquired by the data acquisition device and outputting them to the data transmission device; the terminal device includes a fault type judgment module equipped with a supervised multi-classification machine learning model, and the terminal device is used to receive the data fed back by the data transmission device to obtain the characteristic vibration signal of the robot, perform multi-feature extraction of the robot fault characteristics on the obtained characteristic vibration signal to obtain high-dimensional original data characteristics; perform feature selection on the high-dimensional original data characteristics; reduce the high-dimensional original data characteristics to low-dimensional characteristics; and perform classification processing on the low-dimensional characteristics to obtain a fault diagnosis result. The present invention improves the accuracy of fault diagnosis and solves the problems of difficult state monitoring, slow response, high omission rate or false alarm rate existing in the existing devices.

Description

Technical Field

[0001] The present invention relates to the field of industrial robot status monitoring and fault diagnosis, and specifically, to a robot fault diagnosis system. Background Art

[0002] Industrial robots are widely used in industrial automation production. Failures caused by wear and looseness due to long-term operation of the robot or sudden events will result in the malfunction of the robot, thereby reducing industrial production efficiency and affecting the enterprise's revenue. Therefore, accurate and effective fault detection of the robot is of great significance.

[0003] After retrieval, it is found that the Chinese patent with the publication number CN107984471A provides a method and device for determining the working state of a robot and the robot. The method includes: obtaining the monitoring data of the robot through a working state monitoring thread; determining the working state of the robot according to the obtained monitoring data. The detection indicators it targets are limited to the magnitude of joint current and collision conditions, and the communication method is limited to CAN bus communication, with poor generalization effect and high cost.

[0004] The Chinese patent with the publication number CN111152219A provides a method for monitoring the state of a robot and a monitoring device for the state of a robot. It obtains the charging-on-pile motion information of the robot; compares the charging-on-pile motion information with the reference charging-on-pile motion information to determine the motion state monitoring information of the robot; determines the current state information of the robot based on the motion state monitoring information. This solution is limited to the charging behavior of the robot, making it difficult to monitor the entire life cycle of the robot and difficult to meet actual needs.

[0005] The Chinese patent with the publication number CN112192564A provides a method, device, equipment, and storage medium for remote control of a robot. The method establishes a wireless network connection between the master control device and the teaching pendant corresponding to the robot; receives the remote synchronization data from the teaching pendant; displays the remote synchronization image corresponding to the remote synchronization data in a preset display interface; monitors the state of the robot and performs remote control based on the remote synchronization image. The communication method of this solution is limited to 5G communication, with a short communication distance and high equipment cost, and it is not an optimal choice.

[0006] At the present stage, the robot fault diagnosis system often has difficulties in state monitoring, slow response, and high false alarm rate or missed alarm rate. Summary of the Invention

[0007] Aiming at the defects in the prior art, the purpose of the present invention is to provide a robot fault diagnosis system.

[0008] To solve the above problems, the present invention provides a robot fault diagnosis system, which is arranged on the robot body. The robot body includes a robotic arm and a driving part. An execution part is provided at the end of the robotic arm. The driving part provides power for the robotic arm joints to drive the robotic arm to move and drive the execution part to move synchronously, including:

[0009] A data acquisition device, which is used to acquire the current and position signals of the driving part of the robot; and to acquire the acceleration and pose of the execution part of the robot;

[0010] A data transmission device, which is used to receive the data signals acquired by the data acquisition device and output them to an external terminal device;

[0011] A terminal device, which includes a fault type judgment module equipped with a machine learning model for supervised multi-classification. The terminal device is used to receive the data fed back by the data transmission device to obtain the representative vibration signal of the robot, and use the machine learning model for supervised multi-classification to perform multi-feature extraction of the robot fault characteristics on the obtained representative vibration signal of the robot to obtain high-dimensional original data features; perform feature selection on the high-dimensional original data features, eliminate secondary features, and retain the main high-dimensional original data features that contribute to the model; reduce the retained main high-dimensional original data features to low-dimensional features; perform classification processing on the low-dimensional features to obtain a fault diagnosis result.

[0012] Preferably, the robotic arm includes:

[0013] A support base, which plays a supporting role;

[0014] A base arranged on the support base, and the base is provided with a receiving cavity;

[0015] A first robotic arm arranged on the base, and one end of the first robotic arm is connected to the base to form a first rotating pair;

[0016] A second robotic arm arranged above the first robotic arm, and one end of the second robotic arm is connected to the other end of the first robotic arm to form a second rotating pair;

[0017] A third robotic arm arranged at the other end of the second robotic arm, and the third robotic arm is arranged vertically along the second robotic arm and connected to it to form a third rotating pair, and the third robotic arm can move in the vertical direction.

[0018] Preferably, a vibration damping component for reducing vibration is provided between the base and the support base.

[0019] Preferably, the driving part includes:

[0020] A first joint drive motor connected to one end of the first robotic arm, the first joint drive motor being configured to drive the first rotating pair to rotate and drive the first robotic arm to rotate;

[0021] A second joint drive motor connected to the other end of the second robotic arm, the second joint drive motor being configured to drive the second rotating pair to rotate and drive the first robotic arm to rotate;

[0022] A third joint drive motor connected to the third robotic arm through a first transmission part, the third joint drive motor being configured to drive the third rotating pair to rotate and drive the third robotic arm to rotate;

[0023] A fourth joint drive motor connected to the third robotic arm through a second transmission part, the fourth joint drive motor being configured to drive a ball screw and drive the third robotic arm to move in the vertical direction;

[0024] And the rotation axes of the first joint drive motor, the second joint drive motor, the third joint drive motor, and the fourth joint drive motor are arranged parallel to each other.

[0025] Preferably, the first transmission part is a first synchronous belt, and the tension of the first synchronous belt is adjusted through the flange position of the third joint drive motor;

[0026] The second transmission part is a second synchronous belt, and the tension of the second synchronous belt is adjusted through the flange position of the fourth joint drive motor.

[0027] Preferably, the first joint drive motor is arranged in the accommodation cavity of the base;

[0028] The second joint drive motor, the third joint drive motor, and the fourth joint drive motor are all arranged in the second robotic arm.

[0029] Preferably, the data acquisition device includes:

[0030] A first sensor arranged on the first joint drive motor, the first sensor being configured to measure the current signal and position signal of the first joint motor sensor;

[0031] A second sensor arranged on the second joint drive motor, the second sensor being configured to measure the current signal and position signal of the second joint drive motor;

[0032] A third sensor arranged on the third joint drive motor, the third sensor being configured to measure the current signal and position signal of the third joint drive motor;

[0033] The fourth sensor disposed on the fourth joint drive motor, and the fourth sensor is used to measure the current signal and position signal of the fourth joint drive motor;

[0034] The end sensor disposed on the end effector is used to measure the acceleration and pose of the end effector of the robot body.

[0035] Preferably, a bracket for fixing the end sensor is provided on the end effector.

[0036] Preferably, the data transmission device includes: a wired data transmission module and / or a wireless data transmission module.

[0037] Preferably, the transmission mode of the wired data transmission module adopts serial communication based on the RS485 interface;

[0038] The transmission mode of the wireless data transmission module adopts the TCP protocol based on the Socket interface.

[0039] Compared with the prior art, the present invention has at least one of the following beneficial effects:

[0040] In the above system of the present invention, by setting up a data acquisition device to collect the current signal and position signal of the joint motor and the acceleration and pose of the end effector, the fault state of the robot can be detected, remote real-time monitoring can be realized, and there is no need for manual observation, which greatly improves the nursing efficiency of the robot, reduces the maintenance cost of the robot. At the same time, a supervised multi-class machine learning model is used to extract high-dimensional raw data features, feature selection, feature fusion and classification processing from the obtained vibration signals of the robot. Better classification effects can be achieved with less computing power, which improves the accuracy of fault diagnosis, reduces the cost of the fault diagnosis system, and solves the problems of difficult state monitoring, slow response, high false alarm rate or missed alarm rate in existing robot fault diagnosis devices.

[0041] In the above system of the present invention, by arranging multiple sensors, the dimension of data acquisition is increased, and the accuracy of fault diagnosis is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] By reading the following detailed description of the non-limiting embodiments with reference to the accompanying drawings, other features, objects and advantages of the present invention will become more apparent:

[0043] Figure 1 is a robot fault diagnosis system according to a preferred embodiment of the present invention;

[0044] Figure 2 is a partially enlarged schematic view of the second robotic arm according to a preferred embodiment of the present invention;

[0045] The markings in the figure are respectively represented as: support base 1, base 2, first robotic arm 3, second robotic arm 4, third robotic arm 5, end effector 6, bracket 7, first joint drive motor 8, second joint drive motor 9, third joint drive motor 10, fourth joint drive motor 11, first sensor 12, second sensor 13, third sensor 14, fourth sensor 15, end sensor 16. Detailed implementation manners

[0046] The present invention will be described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made. These all belong to the protection scope of the present invention.

[0047] Referring to Figure 1 As shown, it is a robot fault diagnosis system of a preferred embodiment of the present invention, which is arranged on the robot body and includes: a data acquisition device, a data transmission device, and a terminal device.

[0048] The robot body includes a robotic arm and a drive part. An end effector 6 is provided at the end of the robotic arm. The drive part is used to provide power for the robotic arm joints to drive the movement of the robotic arm, and the robotic arm drives the end effector 6 to move synchronously.

[0049] The data acquisition device is fixedly connected to the robot body. The data acquisition device is used to collect the current and position signals of the drive part; and is used to collect the acceleration and pose of the end effector 6.

[0050] The data transmission device is used to receive the data signals collected by the data acquisition device and output them to an external terminal device to provide the data required by the machine learning model.

[0051] The terminal device includes a fault type judgment module equipped with a supervised multi-classification machine learning model. The terminal device is used to receive the data fed back by the data transmission device to obtain the characteristic vibration signal of the robot, and use the supervised multi-classification machine learning model to perform multi-feature extraction of the robot fault characteristics on the obtained characteristic vibration signal of the robot to obtain high-dimensional original data characteristics; perform feature selection on the high-dimensional original data characteristics, eliminate secondary features, and retain the main high-dimensional original data characteristics that contribute to the model; reduce the retained main high-dimensional original data characteristics to low-dimensional features; perform classification processing on the low-dimensional features to obtain a fault diagnosis result.

[0052] As a preferred method, the supervised multi-classification machine learning model includes:

[0053] Existing algorithms are used to extract high-dimensional raw data features: specifically, for the robot's characterizing vibration signal obtained by the data acquisition device, three methods, namely time domain, frequency domain, and time-frequency domain, are used to extract multi-feature features of the robot's fault features for the robot's characterizing vibration signal to obtain high-dimensional raw data features; through the above-mentioned extraction of high-dimensional raw data features, the interference of redundant information and noise in the original signal is reduced, and in view of the non-stationarity of the fault signal, its time-frequency distribution can be better analyzed, thereby more accurately evaluating the operating status of the robot.

[0054] Use existing algorithms to select features from the extracted high-dimensional raw data features: Specifically, use XGBoost embedded feature selection to evaluate the importance of features to the classification model, delete minor features that do not contribute much to the model but consume computing resources and cause overfitting, retain the main features that contribute to the model, and obtain the data features after feature selection. Through the above feature selection process, the meaningless noise in the data set is reasonably removed, and the loss of effective information is avoided as much as possible, thereby reducing the dimension of the data set to avoid overfitting caused by the dimensional disaster, and reduce the difficulty of the learning task, saving computing power costs while improving the real-time performance of the system.

[0055] Use existing algorithms to perform feature fusion on the selected data: Specifically, the UMAP algorithm is used to reduce the high-dimensional features to higher-quality low-dimensional features through low-dimensional projection based on fuzzy topological structure on the basis of existing feature selection, and obtain the data after feature fusion. Through the above feature fusion processing, the data set is further reduced to three or even two dimensions, thereby realizing feature space visualization.

[0056] Use existing algorithms to implement classifiers: Specifically, for the data features after feature fusion, a classifier based on the XGBoost algorithm is used to obtain fault diagnosis results. By using a classifier based on the XGBoost algorithm, better accuracy, stability and speed can be obtained compared to other classifiers.

[0057] In other preferred embodiments, refer to Figure 1 As shown, the robotic arm includes: a support seat 1, a base 2, a first robotic arm 3, a second robotic arm 4 and a third robotic arm 5.

[0058] The support base 1 supports the upper components.

[0059] The base 2 is provided with a receiving cavity. The base 2 is mounted on the support base 1 and fixed by bolts. As a preferred embodiment, a vibration reduction component for reducing vibration is provided between the base 2 and the support base 1.

[0060] The first mechanical arm 3 is installed on the base 2, and one end of the first mechanical arm 3 is connected to the base 2 through a bearing to form a first rotation pair.

[0061] The second robotic arm 4 is located above the first robotic arm 3. One end of the second robotic arm 4 is connected to the other end of the first robotic arm 3 through a bearing to form a second rotating pair. The other end of the second robotic arm 4 is provided with a vertical space for installing the second robotic arm 4.

[0062] The third robotic arm 5 is installed in the vertical space of the second robotic arm 4. The third robotic arm 5 is arranged vertically along the second robotic arm 4 and is connected to it through a bearing to form a third rotating pair. Moreover, the third robotic arm 5 can move vertically, enabling the third robotic arm 5 and the second robotic arm 4 to perform relative translational motion along a fixed axis in addition to relative rotational motion. The lower end of the third robotic arm 5 is fixed to the end effector 6 through bolts, so that the third robotic arm 5 can drive the end effector 6 to move synchronously.

[0063] The support base 1, the base 2, the first robotic arm 3, and the second robotic arm 4 are arranged parallel to each other and are orthogonal to the third robotic arm 55; the rotation axes of the first robotic arm 3, the second robotic arm 4, and the third robotic arm 55 are arranged parallel to each other.

[0064] In the preferred embodiments of other parts, the driving part includes: a first joint driving motor 8, a second joint driving motor 9, a third joint driving motor 10, and a fourth joint driving motor 11.

[0065] The first joint driving motor 8 is connected to one end of the first robotic arm 3 through a bearing. The first joint driving motor 8 is used to drive the first rotating pair to rotate, driving the first robotic arm 3 to rotate.

[0066] The second joint driving motor 9 is connected to the other end of the second robotic arm 4 through a bearing. The second joint driving motor 9 is used to drive the second rotating pair to rotate, driving the first robotic arm 3 to rotate;

[0067] The third joint driving motor 10 is connected to the third robotic arm 5 through a first transmission part. The third joint driving motor 10 is used to drive the third rotating pair to rotate, driving the second robotic arm 4 to rotate; As a preferred method, the first transmission part is a first synchronous belt. One end of the first synchronous belt is connected to the third joint driving motor 10 and used as the driving end, and the other end is connected to the third robotic arm 5 and used as the driven end. The tightness of the first synchronous belt depends on the distance between the third joint driving motor 10 and the third robotic arm 5. By adjusting the flange position of the third joint driving motor 10, this distance is adjusted to achieve a suitable tightness and ensure the transmission quality.

[0068] The fourth joint drive motor 11 is connected to the third robotic arm 5 through a second transmission part. The fourth joint drive motor 11 is used to drive a ball screw to drive the third robotic arm 5 to move in the vertical direction. As a preferred mode, the second transmission part is a second synchronous belt. One end of the second synchronous belt is connected to the third joint drive motor 11 and used as the driving end, and the other end is connected to the ball screw and used as the driven end. The tightness of the first synchronous belt depends on the distance between the third joint drive motor 11 and the third robotic arm 5. By adjusting the flange position of the fourth joint drive motor 11, this distance is adjusted to achieve a proper tightness and ensure the transmission quality.

[0069] Moreover, the rotating shafts of the first joint drive motor 8, the second joint drive motor 9, the third joint drive motor 10, and the fourth joint drive motor 11 are arranged parallel to each other.

[0070] As a preferred mode, the above-mentioned first joint drive motor 8 is embedded in the accommodation cavity of the base and fixed to the base. Refer to Figure 2 As shown, the second joint drive motor 9 is embedded in the second robotic arm and fixed by four flange screws. The third joint drive motor 10 can be fixed to the motor support embedded and fixed in the second robotic arm by four flange screws. The fourth joint drive motor 11 is fixed to the motor support embedded and fixed in the second robotic arm by four flange screws. The second robotic arm is provided with a space for accommodating the second joint drive motor 9, the third joint drive motor 10, the fourth joint drive motor 11, and the motor support.

[0071] In other preferred embodiments of other parts, the data acquisition device includes: a first sensor 12, a second sensor 13, a third sensor 14, a fourth sensor 15, and a terminal sensor 16.

[0072] The first sensor 12 is fixed to the first joint drive motor 8 by bolts and is coaxially matched with the rotating shaft of the first joint drive motor 8, and is used to measure the current signal and position signal of the first joint motor sensor.

[0073] The second sensor 13 is fixed to the second joint drive motor 9 by bolts and is coaxially matched with the rotating shaft of the second joint drive motor 9, and is used to measure the current signal and position signal of the second joint drive motor 9.

[0074] The third sensor 14 is fixed to the third joint drive motor 10 by bolts and is coaxially matched with the rotating shaft of the third joint drive motor 10, and is used to measure the current signal and position signal of the third joint drive motor 10.

[0075] The fourth sensor 15 is fixed to the fourth joint drive motor 11 by bolts and is coaxially fitted with the rotating shaft of the fourth joint drive motor 11, and is used to measure the current signal and position signal of the fourth joint drive motor 11.

[0076] The end sensor 16 is fixed to the end effector by bolts and is used to measure the acceleration and pose of the end effector of the robot body. As a preferred method, a bracket 7 for fixing the end sensor 16 is provided on the end effector, and the bracket 7 and the end sensor 16 are fixedly connected to the end effector 6 by bolts and planar contact.

[0077] In specific implementation, for the measurement of the position signal of the above joint drive motor, an AVS58 series absolute position encoder of Pepperl+Fuchs GmbH can be used. The maximum resolution is 16 bits (65536 steps per revolution), and the position positioning accuracy of 0.005 degrees can be achieved. It uses optically isolated RS422 communication, supports up to 12000 rpm at most, and the instantaneous inertia is 50 gcm2. It can work under a maximum load of 40 N axially and 110 N radially; for the measurement of the current signal of the above joint drive motor, a CSNX series current sensor of Honeywell can be used. The rated current is 25 A, and it can withstand a current of ±56 A at most. It can respond within 0.2 microseconds, and the error is controlled within 0.24%. For the measurement of the acceleration and pose signals of the above end sensor, an MPU6050 chip of TDK can be used. Combining a three-axis gyroscope and a three-axis accelerometer, noise is reduced through a sensor fusion algorithm, and data is transmitted through I2C communication. The full scale of the gyroscope is ±500° / sec, the accuracy is 65.5 LSB / ° / sec, the rated noise is 0.005 mbps / rtHz, and the maximum range of the three-axis accelerometer is ±4g, and the accuracy is 8192 LSB / g.

[0078] In other preferred embodiments, the data transmission device includes: a wired data transmission module and / or a wireless data transmission module. As a preferred method, the transmission method of the wired data transmission module uses serial communication based on the RS485 interface; the frame format and baud rate of the serial communication are set as: baud rate 115200 Hz, stop bit 1 bit, data bit 8 bits, and no parity bit.

[0079] The transmission method of the wireless data transmission module uses the TCP protocol based on the Socket interface for transmission. The address family and data transmission module of the TCP protocol are set as: the address family is the IPv4 network protocol, and the data transmission module is a reliable stream-oriented service.

[0080] The working process of the above robot fault diagnosis system is as follows:

[0081] A first sensor, a second sensor, a third sensor, a fourth sensor and an end sensor are respectively arranged on the first joint drive motor, the second joint drive motor, the third joint drive motor, the fourth joint drive motor and the end effector to obtain sensor data under the normal operating state of the robot;

[0082] The data transmission device transmits the sensor data to an external terminal computer through a data transmission protocol;

[0083] The computer uses a supervised multi-class machine learning model carried thereon to comprehensively judge the collected signals to obtain the fault type. As a preferred method, high-dimensional time-domain and frequency-domain features are extracted. After XGBoost embedded feature selection and UMAP feature fusion, an XGBoost classifier is used for fault diagnosis, and the current signal, position signal, and acceleration and pose signal collected by the sensor are comprehensively judged to determine the fault type.

[0084] The fault states of the robot include: loose base screws, loose base flange, loose second motor flange screws, loose third motor flange screws, loose or tight synchronous belt caused by abnormal position of the third motor flange, loose fourth motor flange screws, loose or tight synchronous belt caused by abnormal position of the fourth motor flange, and any combination of one or more of them.

[0085] Regarding fault mode recognition, for example, during the previous training dataset marking process, data collection of three working states, namely normal, loose base screws, and loose base flange, was carried out and the above-mentioned supervised multi-class machine learning model was trained. During a subsequent actual test, data was collected and input into the above-mentioned supervised multi-class machine learning model, and then it could be determined which of the above three working states its working state belonged to.

[0086] As shown by experiments, referring to Figure 2 as shown, loose base screws occur at the loose base screw occurrence point Ⅰ; loose chassis flange occurs at the loose chassis flange occurrence point Ⅱ; loose second motor flange screws occur at the loose second motor flange screw occurrence point Ⅲ; loose third motor flange screws occur at the loose third motor flange screw occurrence point Ⅳ; abnormal position of the third motor flange occurs at the abnormal position of the third motor flange occurrence point Ⅴ; loose fourth motor flange screws occur at the loose fourth motor flange screw occurrence point Ⅵ; abnormal position of the fourth motor flange occurs at the abnormal position of the fourth motor flange occurrence point Ⅶ.

[0087] Through the above various structural optimization designs, the present invention has the characteristics of multiple data collection dimensions and remote real-time monitoring.

[0088] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various modifications or variations within the scope of the claims, which does not affect the essence of the present invention.

Claims

1. A robot fault diagnosis system, characterized in that, It is arranged on the robot body, the robot body includes a robotic arm and a driving part, an execution part is arranged at the end of the robotic arm, and the driving part provides power for the robotic arm joints to drive the robotic arm to move and drive the execution part to move synchronously, including: A data acquisition device, which is used to acquire the current and position signals of the driving part of the robot, and to acquire the acceleration and pose of the execution part of the robot; A data transmission device, which is used to receive the data signals acquired by the data acquisition device and output them to an external terminal device; A terminal device, the terminal device includes a fault type judgment module equipped with a machine learning model for supervised multi-classification. The terminal device is used to receive the data fed back by the data transmission device to obtain the characteristic vibration signal of the robot, and use the machine learning model for supervised multi-classification to perform multi-feature extraction of the characteristic vibration signal of the obtained robot to obtain high-dimensional original data features; perform feature selection on the high-dimensional original data features, eliminate secondary features, and retain the main high-dimensional original data features that contribute to the model; reduce the retained main high-dimensional original data features to low-dimensional features; perform classification processing on the low-dimensional features to obtain a fault diagnosis result; among them, extract high-dimensional time-domain and frequency-domain features, after XGBoost embedded feature selection and UMAP feature fusion, use the XGBoost classifier for fault diagnosis, comprehensively judge the current signal, position signal, and acceleration and pose signals collected by the data acquisition device, so as to judge the fault type.

2. The robot fault diagnosis system according to claim 1, characterized in that, The robotic arm includes: A support base; A base arranged on the support base, and the base is provided with a receiving cavity; A first robotic arm arranged on the base, one end of the first robotic arm is connected to the base to form a first rotating pair; A second robotic arm arranged above the first robotic arm, one end of the second robotic arm is connected to the other end of the first robotic arm to form a second rotating pair; A third robotic arm arranged at the other end of the second robotic arm, the third robotic arm is arranged vertically along the second robotic arm and is connected to it to form a third rotating pair, and the third robotic arm can move in the vertical direction.

3. According to the robot fault diagnosis system described in claim 2, characterized in that A vibration damping component for reducing vibration is arranged between the base and the support base.

4. A robot fault diagnosis system according to claim 2, wherein, The driving part includes: A first joint driving motor connected to one end of the first robotic arm, the first joint driving motor is used to drive the first rotating pair to rotate and drive the first robotic arm to rotate; A second joint driving motor connected to the other end of the first robotic arm, the second joint driving motor is used to drive the second rotating pair to rotate and drive the second robotic arm to rotate; A third joint driving motor connected to the third robotic arm through a first transmission part, the third joint driving motor is used to drive the third rotating pair to rotate and drive the third robotic arm to rotate; A fourth joint driving motor connected to the third robotic arm through a second transmission part, the fourth joint driving motor is used to drive the ball screw to drive the third robotic arm to move in the vertical direction; Moreover, the rotation axes of the first joint drive motor, the second joint drive motor, the third joint drive motor, and the fourth joint drive motor are arranged parallel to each other.

5. A robot fault diagnosis system according to claim 4, wherein The first transmission part is a first synchronous belt, and the tension of the first synchronous belt is adjusted through the flange position of the third joint drive motor; The second transmission part is a second synchronous belt, and the tension of the second synchronous belt is adjusted through the flange position of the fourth joint drive motor.

6. A robot fault diagnosis system according to claim 4, wherein The first joint drive motor is arranged in the accommodation cavity of the base; The second joint drive motor, the third joint drive motor, and the fourth joint drive motor are all arranged in the second robotic arm.

7. A robot fault diagnosis system according to claim 4, characterized in that, The data acquisition device includes: A first sensor arranged on the first joint drive motor, and the first sensor is used to measure the current signal and position signal of the first joint drive motor; A second sensor arranged on the second joint drive motor, and the second sensor is used to measure the current signal and position signal of the second joint drive motor; A third sensor arranged on the third joint drive motor, and the third sensor is used to measure the current signal and position signal of the third joint drive motor; A fourth sensor arranged on the fourth joint drive motor, and the fourth sensor is used to measure the current signal and position signal of the fourth joint drive motor; A terminal sensor arranged on the end effector, which is used to measure the acceleration and pose of the end effector of the robot body.

8. A robot fault diagnosis system according to claim 7, characterized in that, A bracket for fixing the terminal sensor is provided on the end effector.

9. A robot fault diagnosis system according to claim 1, wherein, The data transmission device includes: a wired data transmission module and / or a wireless data transmission module.

10. A robot fault diagnosis system according to claim 9, wherein The transmission mode of the wired data transmission module adopts serial communication based on the RS485 interface; The transmission mode of the wireless data transmission module adopts the TCP protocol based on the Socket interface.

Images