Torque control method and device of robot joint module, computer device, storage medium and program product

Abstract

The application relates to a torque control method and device of a robot joint module, computer equipment, a storage medium and a program product. The method comprises the following steps: in response to a mechanical failure event of the robot joint module, identifying a failure level of the mechanical failure event; determining a mechanical failure coefficient of the joint module according to the failure level; the mechanical failure coefficient is used for representing the influence degree of the mechanical failure event on torque control of the joint module; determining a target torque value of the joint module according to the mechanical failure coefficient, a motor temperature value of the joint module and a joint vibration frequency of the joint module; and controlling the motor of the joint module to adjust the output torque according to the target torque value. The method can improve the safety of the robot under mechanical failure.

Description

Technical Field

[0001] This application relates to the field of robot control technology, and in particular to a torque control method, device, computer equipment, computer-readable storage medium, and computer program product for a robot joint module. Background Technology

[0002] With the development of robotics technology, legged robots, represented by quadrupedal robot dogs, are increasingly being widely used in diverse scenarios such as industrial environments and daily life. As the core motion components of a robot, the robot's joint module, when performing highly dynamic and high-load tasks such as running, jumping, climbing, and carrying, requires its internal mechanical components, such as bearings, gearboxes, and output shafts, to withstand enormous impacts and prolonged wear. Over time, these mechanical components inevitably experience performance degradation, leading to mechanical failures.

[0003] In traditional technology, the desired state of the joint module is predicted by a precise mathematical model of the robot, and the torque is controlled based on the predicted desired state. However, this method has a model mismatch problem. When the joint module has a mechanical failure, its actual physical characteristics have deviated from the original ideal model. At this time, controlling the torque based on the ideal model may not only fail to achieve precise control, but may also output incorrect control commands due to model mismatch, which will aggravate the failure and lead to lower robot safety under mechanical failure. Summary of the Invention

[0004] Therefore, it is necessary to provide a torque control method, device, computer equipment, computer-readable storage medium, and computer program product for robot joint modules that can improve the safety of robots under mechanical failures, in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a torque control method for a robot joint module, comprising:

[0006] In response to a mechanical failure event in the robot's joint module, the failure level of the mechanical failure event is identified;

[0007] Based on the fault level, the mechanical fault coefficient of the joint module is determined; the mechanical fault coefficient is used to characterize the degree of influence of the mechanical fault event on the torque control of the joint module.

[0008] The target torque value of the joint module is determined based on the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module.

[0009] Based on the target torque value, the motor of the joint module is controlled to adjust its output torque.

[0010] In one embodiment, determining the mechanical failure coefficient of the joint module based on the failure level includes:

[0011] Based on the fault level, determine the initial fault coefficient of the joint module;

[0012] The working condition of the joint module is determined based on the task execution information of the joint module; the task execution information includes at least one of the joint module's gait type, load status, and task duration.

[0013] When the operating condition meets the fault-sensitive condition, the initial fault coefficient is adjusted to obtain the mechanical fault coefficient; the mechanical fault coefficient is higher than the initial fault coefficient; the mechanical fault coefficient is negatively correlated with the target torque value of the joint module.

[0014] In one embodiment, determining the target torque value of the joint module based on the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module includes:

[0015] Mechanical risk parameters are determined based on the torque fluctuation of the joint module and the joint vibration frequency of the joint module;

[0016] The temperature risk parameter is determined based on the difference between the motor temperature value and the reference temperature value of the joint module;

[0017] The torque adjustment coefficient is determined based on the product of the mechanical risk parameter, the temperature risk parameter, and the mechanical failure coefficient.

[0018] The target torque value of the joint module is determined based on the reference torque value of the joint module and the torque adjustment coefficient.

[0019] In one embodiment, determining the target torque value of the joint module based on the reference torque value of the joint module and the torque adjustment coefficient includes:

[0020] The torque limit of the joint module is determined based on the reference torque value of the joint module and the torque adjustment coefficient.

[0021] When the joint vibration frequency of the joint module increases by a preset amount, the torque limit of the joint module is adjusted according to an adjustment ratio that matches the preset amount, so as to obtain the target torque value of the joint module.

[0022] When the motor temperature of the joint module increases by a preset amount relative to the safe temperature, the torque limit of the joint module is adjusted according to an adjustment ratio that matches the preset temperature amount to obtain the target torque value of the joint module.

[0023] In one embodiment, controlling the motor of the joint module to adjust its output torque according to the target torque value includes:

[0024] According to the gait cycle of the joint module, the motor of the joint module is controlled to linearly transition the output torque from the current torque value to the target torque value.

[0025] In one embodiment, after controlling the motor of the joint module to adjust its output torque according to the target torque value, the method further includes:

[0026] When the joint vibration frequency, the motor temperature, and the joint load of the joint module meet the preset torque recovery conditions, the motor of the joint module is controlled to gradually restore the output torque from the current torque value to the reference torque value of the joint module at preset intervals.

[0027] Secondly, this application also provides a torque control device for a robot joint module, comprising:

[0028] The identification module is used to identify the fault level of a mechanical fault event in response to a mechanical fault event in the robot's joint module.

[0029] A determination module is used to determine the mechanical failure coefficient of the joint module based on the failure level; the mechanical failure coefficient is used to characterize the degree of influence of the mechanical failure event on the torque control of the joint module;

[0030] The determining module is further configured to determine the target torque value of the joint module based on the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module.

[0031] The control module is used to control the motor of the joint module to adjust the output torque according to the target torque value.

[0032] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0033] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0034] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0035] The aforementioned torque control method, device, computer equipment, computer-readable storage medium, and computer program product for robot joint modules identify the fault level of a mechanical fault event in response to such an event. Based on the fault level, they determine the degree of impact of the mechanical fault event on the joint module. Then, based on the mechanical fault coefficient, the motor temperature of the joint module, and the joint vibration frequency, they determine the target torque value of the joint module. In this way, the severity of the mechanical fault in the joint module can be accurately quantified through the fault level. Then, by dynamically combining the mechanical fault coefficient (representing the severity of the fault), the motor temperature (representing the consequences and aggravating factors of the fault), and the joint vibration frequency (representing the direct manifestation of the fault), the target torque value is determined. This allows for precise torque control according to the target torque value when a mechanical fault exists in the robot, avoiding the worsening of the mechanical fault due to torque control errors, and improving the safety and reliability of the robot under mechanical fault conditions. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is an application environment diagram of a torque control method for a robot joint module in one embodiment.

[0038] Figure 2 This is a flowchart illustrating a torque control method for a robot joint module in one embodiment;

[0039] Figure 3 This is a logic diagram of a torque control method for a robot joint module in one embodiment;

[0040] Figure 4 This is a flowchart illustrating a torque control method for a robot joint module in another embodiment;

[0041] Figure 5 This is a structural block diagram of a torque control device for a robot joint module in one embodiment;

[0042] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] In related technologies, torque control methods employed when robots experience mechanical faults can include: passive, threshold-triggered protection methods, which are slow to react, crude in approach, and unable to prevent faults, easily leading to task interruptions or accidents; control methods relying on precise mathematical models, whose performance drops sharply after model mismatch and cannot adapt to gradual degradation of mechanical performance; isolated, pre-defined fault response strategies, which have poor generalization ability, are only applicable to specific faults, rely on traditional experience, and have high development costs; control methods based on single sensor feedback, lacking multi-dimensional information fusion, which cannot comprehensively diagnose fault types and levels, resulting in insufficient perception; methods that ignore the impact of specific task conditions on mechanical faults, which cannot identify high-risk scenarios and provide enhanced protection, resulting in a disconnect between control strategies and tasks; and simple linear derating methods, which cannot cope with the nonlinear deterioration characteristics of mechanical systems and have poor protection effects.

[0045] The torque control method for robot joint modules provided in this application can be applied to, for example... Figure 1 In the application environment shown, robot 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0046] Robot 102 may include legged robots, robot dogs, etc., and may also include any robot that uses motors, gearboxes, bearings, etc. to form joint modules and requires torque control.

[0047] Server 104 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0048] In practical applications, robot 102 responds to mechanical failure events in the robot's joint modules by identifying the failure level of the mechanical failure event; robot 102 determines the mechanical failure coefficient of the joint module based on the failure level; the mechanical failure coefficient is used to characterize the severity of the mechanical failure of the joint module; robot 102 determines the target torque value of the joint module based on the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module; robot 102 controls the motor of the joint module to adjust the output torque based on the target torque value.

[0049] For example, robot 102 may include a sensor layer (sensing system), a control layer, and an execution layer.

[0050] The sensor layer may include joint vibration sensors, high-precision encoders, torque sensors, and plantar pressure sensors. Optionally, the joint vibration sensor may include a piezoelectric accelerometer, mounted on the joint module housing, with a bandwidth covering a frequency range of 0 to 5000 Hz and a sampling rate of no less than 1 kHz, used to accurately extract joint vibration frequencies; the high-precision encoder may include a multi-turn absolute encoder with 2048 lines or higher, mounted at both ends of the motor and output shaft, indirectly yielding angular displacement error and speed difference by calculating the angle difference between the two; the torque sensor may include a strain gauge torque sensor, integrated at the gearbox output, with a range covering the robot's maximum joint torque, such as 0 to 100 N·m, used to directly measure output torque and calculate its fluctuation; the plantar pressure sensor may include a thin-film pressure sensor or a torque sensor, mounted on each foot, used to measure ground reaction force and calculate its fluctuation.

[0051] The control layer may include joint controllers (JCs) and a main control unit (MCU). Optionally, each joint can be equipped with a joint controller, which is the core execution unit of the algorithm and needs to have sufficient computing power to perform real-time vibration spectrum analysis, torque fluctuation monitoring, and local closed-loop control. The main control unit serves as the coordination center of the entire machine, used to receive fault and status information uploaded by all joint controllers, perform gait planning and coordination, and when a joint needs to de-descent, the main control unit will adjust the gait and the output of the other joints to maintain the balance of the entire machine and prevent falls.

[0052] The execution layer may include joint motors. Optionally, the joint motors may support a high-precision torque closed-loop control mode and be able to receive torque commands from the joint controller and adjust the output torque quickly and linearly. For example, the torque may change smoothly at a rate of 20 Nm per 100 milliseconds to avoid sudden torque changes that could lead to gait instability.

[0053] In one exemplary embodiment, such as Figure 2 As shown, a torque control method for a robot joint module is provided, which is applied to... Figure 1 Taking robot 102 as an example, the explanation includes:

[0054] Step S202: In response to a mechanical fault event in the robot's joint module, identify the fault level of the mechanical fault event.

[0055] Mechanical failure events can include abnormal events caused by performance degradation or damage of mechanical components during the operation of the robot joint module. For example, the mechanical components of the joint module may include bearings, gearboxes, output shafts, etc.

[0056] Among them, the fault level of a mechanical failure event can be used to characterize the severity of the mechanical failure event. For example, the fault level may include minor failure, severe failure, etc.

[0057] In practical applications, precise fault classification and quantification can be performed on typical failure modes of core mechanical components (bearings, gearboxes, output shafts) in robot joint modules. A mechanical fault event can be identified if any fault type is determined based on mechanical characteristic parameters collected by sensors (such as joint vibration frequency, bearing clearance, torque fluctuation, etc.). Furthermore, in addition to defining fault types, the fault level within the corresponding fault type can be determined based on the mechanical characteristic parameters detected by sensors and grading standards (mild / severe). This allows for accurate determination of the nature and severity of the fault, laying the foundation for subsequent precise torque control.

[0058] The typical failure modes, mechanical fault types, mechanical characteristic parameters, and fault level classification methods of the mechanical components of the robot joint module are shown in Table 1.

[0059] Table 1

[0060]

[0061] Table 1 defines quantifiable fault characteristic parameters (such as vibration frequency and torque fluctuation) and grade standards (light / heavy) for the three core components of robot joints: bearings, gearboxes, and output shafts. This enables multi-dimensional quantitative diagnosis of mechanical faults and a leap from fuzzy judgment to precise diagnosis.

[0062] Specifically, firstly, raw data can be synchronously read from multiple sensors via a joint controller (JC): time-domain acceleration signals (a) can be read from vibration sensors. x , a y , a z ); Read the position values ​​(pos) from the motor end and the output end of the encoder. motor, pos output ); Read the current output torque value (T) from the torque sensor. current ); Read the motor and gearbox temperatures (T) from the temperature sensor. motor , T gear ).

[0063] Then, signal processing and feature extraction can be performed on these raw data to obtain mechanical characteristic parameters: vibration frequency analysis can be performed by applying a Fast Fourier Transform (FFT) to the time-domain vibration signal (such as the Z-axis, i.e., axial vibration) to convert the signal from the time domain to the frequency domain. Then, specific high-frequency peaks (e.g., above 1500Hz) that are distinct from motor electromagnetic noise and gait frequency can be identified in the frequency domain spectrum. The frequency of this peak is the joint vibration frequency (f vib It can perform torque fluctuation calculation, recording a series of torque sensor readings within a fixed time window (e.g., 100 milliseconds), and obtaining the output torque fluctuation (ΔT) by calculating the standard deviation or peak-to-peak value of these values; it can also perform angular displacement error calculation, based on the position values ​​of the encoders at the motor end and the output end, theoretically, there should be a pos [value missing] when there is no fault. output =pos motor / gear-ratio (reduction ratio), and continuously calculate the error value X. error =pos output - (pos motor / gear-ratio), this error value X error The root mean square (RMS) or maximum value is the shaft angular displacement error (ε); for bearing clearance (δ) and gear meshing clearance (c), it can be estimated indirectly through prior calibration or based on vibration and torque signals, rather than direct real-time measurement.

[0064] Next, the joint controller (JC) can calculate the mechanical characteristic parameters (f) vib The fault severity rating (f, ΔT, ε) is compared with the fault severity rating criteria pre-stored in memory. For example, if f vib If the frequency is ≥2000Hz and the estimated δ≥0.3mm, the joint controller (JC) will diagnose a mechanical fault event of "severe bearing wear" in the joint module and set the fault flag and fault level.

[0065] The above process can be cycled every 1 to 10 milliseconds on the joint controller (JC). By performing high-frequency acquisition and quantization of mechanical characteristic parameters, the raw data collected by the sensors can be converted into usable mechanical characteristic parameters, thereby determining the mechanical fault events existing in the joint module and the corresponding fault levels in real time and accurately.

[0066] Step S204: Determine the mechanical failure coefficient of the joint module based on the failure level.

[0067] Among them, the mechanical failure coefficient is used to characterize the degree of impact of mechanical failure events on the joint module.

[0068] For example, if the fault level is minor, the mechanical fault coefficient is set to a first preset value, such as 0.3; if the fault level is severe, the mechanical fault coefficient is set to a second preset value, such as 0.6, where the second preset value is greater than the first preset value. A higher mechanical fault coefficient indicates a more severe mechanical fault. By setting the mechanical fault coefficient, torque can be controlled based on the severity of the mechanical fault. A higher mechanical fault coefficient results in a greater torque derating in subsequent torque calculations, thus providing stronger protection for the joint module.

[0069] By classifying and grading faults in a refined manner, the traditional approach of treating mechanical faults as a whole and uniformly reducing power is avoided, which leads to the inability to distinguish the type and severity of faults and improves the accuracy of subsequent torque control.

[0070] Step S206: Determine the target torque value of the joint module based on the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module.

[0071] Among them, the mechanical failure coefficient can be an adaptive adjustment parameter used to automatically adjust the aggressiveness of the torque control strategy according to the severity of the failure.

[0072] The motor temperature value can be obtained in real time by a temperature sensor installed near the motor windings or on the module housing, and is used to characterize the thermal state or thermal stress of the motor inside the joint module.

[0073] The joint vibration frequency can be determined by collecting the original vibration signal through a vibration sensor (such as an accelerometer) installed on the joint module shell, and then by the joint controller through signal processing algorithms (such as Fourier transform). It can be used as one of the most direct indicators reflecting the mechanical health status of the joint.

[0074] In one embodiment, a preset torque derating model can be used to map the mechanical failure coefficient, the motor temperature value of the joint module, the reference torque value of the joint module, and the joint vibration frequency of the joint module to a target torque value for the joint module. This torque derating model can be used to: determine the torque adjustment coefficient based on the mechanical failure coefficient, motor temperature value, and joint vibration frequency; and determine the target torque value of the joint module based on the reference torque value and the torque adjustment coefficient. The reference torque value can be a torque value determined based on basic torque requirements, which may include robot task requirements. Torque adjustment can be scaled based on the reference torque value, ensuring clear torque control logic. By establishing a three-dimensional mapping model of vibration-temperature-torque, the degree of failure, temperature state, and task requirements are dynamically coupled, and the optimal safe torque limit is calculated non-linearly, rather than through simple linear derating, achieving a balance between precise performance and protection.

[0075] Traditional derating strategies are typically linear, with higher temperatures resulting in lower torque and a simplistic model. This embodiment proposes a dynamic three-dimensional "vibration-temperature-torque" mapping model. This torque derating model unifies three key variables—mechanical vibration (direct manifestation of fault), temperature (fault consequences and aggravating factors), and base torque (operating condition requirements)—into a single mathematical model. Simultaneously, a mechanical fault coefficient is introduced as an adaptive factor to automatically adjust the aggressiveness of the strategy based on the severity of the fault.

[0076] Optionally, intelligent rules based on engineering experience can be embedded, such as how to derate the torque for every 500 Hz increase in vibration frequency, how to further derate the torque at high temperatures, and how to strengthen the protection of the joint module under specific working conditions, so that the torque derate process is both scientific and safe.

[0077] Step S208: Based on the target torque value, control the motor of the joint module to adjust the output torque.

[0078] In practice, the robot controls the motors of the joint module to adjust the output torque according to the target torque value. This can be done by adjusting the output torque to be lower than the target torque value or by adjusting the output torque to the target torque value.

[0079] Specifically, a new torque control command can be generated based on the target torque value and sent to the motor through the motor controller; the motor controller receives the torque control command and adjusts the motor drive current according to the target torque value in the torque control command, thereby controlling the output torque of the motor.

[0080] In the aforementioned torque control method for robot joint modules, the fault level of a mechanical fault event is identified in response to such an event. Based on the fault level, the degree of impact of the mechanical fault event on the joint module is determined. Then, based on the mechanical fault coefficient, the motor temperature of the joint module, and the joint vibration frequency, a target torque value for the joint module is determined. This allows for precise quantification of the severity of the mechanical fault in the joint module through the fault level. Then, by dynamically combining the mechanical fault coefficient (representing the severity), the motor temperature (representing the consequences and aggravating factors), and the joint vibration frequency (representing the direct manifestation of the fault), the target torque value is determined. This enables precise torque control according to the target torque value when a mechanical fault exists, preventing the mechanical fault from worsening due to torque control errors and improving the robot's safety and reliability under mechanical fault conditions.

[0081] In another embodiment, determining the mechanical failure coefficient of the joint module based on the failure level includes: determining the initial failure coefficient of the joint module based on the failure level; determining the working condition of the joint module based on the task execution information of the joint module; adjusting the initial failure coefficient to obtain the mechanical failure coefficient when the working condition meets the fault-sensitive working condition; the mechanical failure coefficient is higher than the initial failure coefficient; the mechanical failure coefficient is negatively correlated with the target torque value of the joint module.

[0082] The initial fault coefficient is a mechanical fault coefficient that the robot directly determines based on the fault level. For details, please refer to the description of the steps for determining the mechanical fault coefficient of the joint module based on the fault level.

[0083] The task execution information includes information reflecting the robot's current motion state and task status. This information includes at least one of the following: gait type of the joint modules, load status, and task duration. Gait type refers to the robot's current gait pattern. For example, gait types can include walking, trotting, and galloping. Different gait types correspond to different movement speeds and impact intensities; for example, galloping gait has a high movement speed and a high impact intensity on the joint modules. Load status refers to the current load on the joint modules, which can be characterized by joint load parameters. These parameters can include the load torque or load force currently borne by the joint modules. When the joint load parameters exceed a preset multiple of the rated value, the robot is considered to be in a heavy-load state. Task duration refers to the duration for which the robot performs a certain task or remains in a certain state. The longer the task duration, the greater the cumulative wear on the mechanical components.

[0084] In practice, the robot can match the current task execution information with preset working condition judgment conditions to identify the working condition of the joint module.

[0085] Fault-sensitive operating conditions can include specific operating conditions that are sensitive to mechanical failures, also known as sensitive operating conditions. Under fault-sensitive operating conditions, mechanical failures may deteriorate more rapidly, therefore the mechanical failure coefficient needs to be adjusted to provide stronger protection.

[0086] For example, fault-sensitive operating conditions may include heavy-load running conditions, rugged terrain climbing conditions, static weight-bearing turning conditions, and continuous obstacle-crossing conditions. Among them, heavy-load running conditions may include conditions where the robot runs at high speed under heavy load; rugged terrain climbing conditions may include conditions where the robot climbs uphill on rugged terrain; static weight-bearing turning conditions may include conditions where the robot turns at low speed under heavy load; and continuous obstacle-crossing conditions may include conditions where the robot continuously crosses obstacles.

[0087] In practice, when the operating conditions meet the criteria for fault-sensitive conditions, the initial fault coefficient can be adjusted to obtain the mechanical fault coefficient. The adjustment can be achieved by multiplying the initial fault coefficient by an adjustment factor, which can be a value greater than 1. For example, under heavy-load running conditions, the initial fault coefficient can be increased by 20% (i.e., multiplied by 1.2) to obtain the adjusted mechanical fault coefficient.

[0088] Because the mechanical failure coefficient is negatively correlated with the target torque value of the joint module—that is, the larger the mechanical failure coefficient, the smaller the target torque value—the robot can provide stronger protection under fault-sensitive conditions, preventing further deterioration of mechanical failures.

[0089] In practical applications, the differences in the performance of mechanical faults under different gait states can be analyzed using gait type (G), gait frequency (f), and joint load (F). joint Vibrational energy (E) vib The duration (t) is used to identify operating conditions that are sensitive to mechanical wear, i.e., fault-sensitive operating conditions. The classification and identification conditions of each fault-sensitive operating condition (sensitive operating condition) are shown in Table 2.

[0090] Table 2

[0091]

[0092] In practical applications, all joint controllers (JCs) will display local fault characteristics and joint loads (F). joint Vibrational energy (E) vib Data such as data is uploaded to the main controller (MCU) of the torso.

[0093] Then, the main trunk controller (MCU) can make a judgment by combining the following information: gait generator status, including the current gait type (such as Trot, Gallop, Walk); gait planner status, including stride frequency and expected joint torque; IMU data collected by the environmental perception module (to determine whether it is climbing), visual / LiDAR data (to determine whether it is crossing obstacles), etc.; and data uploaded from the joint controller (JC).

[0094] The main control unit (MCU) can perform pattern matching between all current states and a predefined library of sensitive operating conditions. For example, if the MCU detects that "gait = Gallop and joint load > 1.8 * rated load value and duration > 10 seconds," it determines that the robot is currently in a heavy-load running condition. The MCU will broadcast this condition flag to all JCs. Furthermore, the main control unit (MCU) can cycle through the above process every 100 milliseconds, performing fault-sensitive operating condition identification at a medium frequency.

[0095] By identifying fault-sensitive operating conditions, the risk of exacerbating joint mechanical failures due to the current task being performed by the robot can be identified. Furthermore, in addition to monitoring the joints themselves, the robot also analyzes overall operating condition information such as gait type, load, and duration in real time. Through correlation analysis, such as the increased bearing wear during heavy-load running conditions and the highest gearbox risk during hill climbing conditions on rugged terrain, the protection strategy for the joint module becomes more targeted and forward-looking.

[0096] The technical solution of this embodiment can not only diagnose joint faults, but also intelligently identify fault-sensitive working conditions (such as heavy-load running) that will rapidly accelerate mechanical wear by comprehensively considering information such as gait, load, and duration, making the protection strategy forward-looking; it solves the problem of insufficient torque control accuracy in traditional technologies, which are usually static or based solely on temperature, and rarely consider the specific task the robot is currently performing (such as running or climbing).

[0097] In another embodiment, determining the target torque value of the joint module based on the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module includes: determining mechanical risk parameters based on the torque fluctuation of the joint module and the joint vibration frequency of the joint module; determining temperature risk parameters based on the difference between the motor temperature value of the joint module and the reference temperature value; determining a torque adjustment coefficient based on the product of the mechanical risk parameter, the temperature risk parameter, and the mechanical failure coefficient; and determining the target torque value of the joint module based on the reference torque value of the joint module and the torque adjustment coefficient.

[0098] The torque fluctuation can be obtained by collecting torque data from a torque sensor within a certain time window and calculating its standard deviation or peak-to-peak value (the difference between the maximum and minimum values).

[0099] The mechanical risk parameter is determined by combining the torque fluctuation of the joint module with the joint vibration frequency. For example, the mechanical risk parameter can be equal to the sum of the normalized torque fluctuation and the normalized joint vibration frequency.

[0100] The reference temperature value may include a pre-set baseline temperature value, which represents the reasonable upper limit of temperature at which the motor can operate safely and for a long time, such as 85 degrees Celsius.

[0101] The temperature risk parameter is determined based on the difference between the motor temperature and the reference temperature. For example, the temperature risk parameter can be equal to the difference between the motor temperature and the reference temperature, divided by the reference temperature.

[0102] The reference torque value may include the expected torque value issued by the upper controller (such as the torso main controller MCU) to the joint controller JC according to the task currently being performed by the robot (such as running or climbing). It can be used to characterize the torque required to complete the current task under ideal conditions without mechanical failure.

[0103] The torque adjustment coefficient is determined based on the product of the mechanical risk parameter, the temperature risk parameter, and the mechanical failure coefficient. For example, the torque adjustment coefficient can be equal to 1 minus the product of the mechanical risk parameter, the temperature risk parameter, and the mechanical failure coefficient.

[0104] In one embodiment, a preset torque derating model can be used to map the mechanical failure coefficient, the motor temperature value of the joint module, and the joint vibration frequency of the joint module to the target torque value of the joint module. This torque derating model can be expressed as:

[0105] T limit =T base ×[1-k mech ×(F vib_norm +ΔT norm )×(TT ref ) / T ref ];

[0106] Among them, T limit For the target torque value ( ), that is, torque limit; T base This is the baseline torque value under fault-free conditions, matched to the current operating conditions; k mech The mechanical failure coefficient can be determined based on the failure level; for example, k is the coefficient for a minor failure level. mech The value is 0.3, and k is at the severe fault level. mech If the current operating condition is 0.6, then k can be adjusted to 0.6. mech Multiply by 1.2; F vib_normThe normalized joint vibration frequency, for example, via f vib / 2000Hz for joint vibration frequency f vib Normalization is performed; ΔT norm This refers to the normalized torque fluctuation, for example, by normalizing the torque fluctuation ΔT by ΔT / 20%; T is the motor temperature value (°C); T ref This is a reference temperature value, such as 85℃.

[0107] Specifically, the joint controller (JC) can read the local motor temperature value T; the joint controller (JC) can receive operating condition flags from the main control unit (MCU), such as heavy-load running conditions; and the joint controller (JC) can obtain the reference torque value (T) requested by the joint in the current joint module. base This is a requirement from the upper-level controller.

[0108] Then, the normalized value of the joint vibration frequency can be calculated:

[0109] F vib_norm =f vib / 2000Hz, where 2000Hz is a predefined value.

[0110] Simultaneously, the normalized value of the torque fluctuation can be calculated:

[0111] ΔT norm =ΔT / 20%, where 20% is a predefined value.

[0112] Then, the above data can be input into the torque derating model to obtain the target torque value. For example, the target torque value can be calculated based on the torque derating model by looping the calculation every 10 to 50 milliseconds on the joint controller (JC) at a medium to high frequency. This allows for the dynamic calculation of the safe upper limit of torque, i.e., the target torque value, based on the current fault and operating conditions.

[0113] This torque derating model is a temperature-torque co-derating model used under mechanical faults. It is a multi-factor coupled dynamic adaptive model that creatively integrates fault characteristics (vibration), system state (temperature), and external demand (baseline torque) into a concise and efficient mathematical model. In the torque derating model, the more severe the fault (F... vib_norm and ΔT norm The larger the temperature (T), the greater the derating; the higher the temperature (T), the greater the derating, and the steeper the derating slope at higher temperatures, through (T - T). ref ) / T ref Reflection; Basic torque requirement (T) baseThe derating is retained, and scaling is applied on this basis to ensure the clarity of the control logic; the adaptive mechanical failure coefficient k mech This is the adjustment knob of the model. Based on the fault level setting, it directly differentiates the severity of faults at the strategy level. Furthermore, under fault-sensitive conditions such as heavy-load operation, it can improve k... mech .

[0114] The technical solution of this application embodiment multiplies the mechanical risk parameters determined by vibration and torque fluctuations with the temperature risk parameters determined by the motor temperature value, and the mechanical fault coefficient representing the fault level, to jointly determine the final torque adjustment coefficient. Thus, the target torque value is determined based on the reference torque value and the torque adjustment coefficient, realizing the quantification and coupling of multi-physics risk, and improving the accuracy and reliability of torque control.

[0115] In another embodiment, determining the target torque value of the joint module based on the reference torque value and torque adjustment coefficient of the joint module includes: determining the torque limit of the joint module based on the reference torque value and torque adjustment coefficient of the joint module; adjusting the torque limit of the joint module according to an adjustment ratio matched with the preset frequency amount for each increase in the joint vibration frequency of the joint module to obtain the target torque value of the joint module; and adjusting the torque limit of the joint module according to an adjustment ratio matched with the preset temperature amount for each increase in the motor temperature value of the joint module relative to the safe temperature value to obtain the target torque value of the joint module.

[0116] The torque limit is a preliminary target torque value determined based on the reference torque value and the torque adjustment coefficient. For details, please refer to the above description of the steps to determine the target torque value of the joint module based on the reference torque value and the torque adjustment coefficient of the joint module.

[0117] In practical applications, additional key torque control methods can be added to the core torque derating model to handle extreme situations. These additional torque control methods may include: for every preset increase in joint vibration frequency, additional torque derating is performed according to an adjustment ratio matched to the preset frequency, actively avoiding resonance points, preventing avalanche-like deterioration of faults, and achieving anti-resonance; after exceeding the safe temperature value, for every preset increase in temperature, additional torque derating is performed according to an adjustment ratio matched to the preset temperature, addressing the nonlinear risks caused by high-temperature lubrication failure and achieving high-temperature protection.

[0118] For example, if the joint vibration frequency increases by 500Hz compared to the last calculation, the torque limit will be reduced by an additional 8% on top of the torque limit. That is, for every 500Hz increase in joint vibration frequency, the torque limit will be reduced by 8% to avoid resonance exacerbating wear. If the motor temperature T>100℃, an additional 5% derating will be applied for every 5℃ increase above 100℃. In other words, when the temperature exceeds 100℃, an additional 5% derating will be applied for every 5℃ increase to avoid high temperature exacerbating lubrication failure. The final target torque value can be limited to between 0 and an allowable maximum value.

[0119] The technical solution of this application embodiment dynamically determines the torque limit by comprehensively considering the mechanical failure coefficient, motor temperature value and joint vibration frequency, and makes additional adjustments to the torque according to the changes in joint vibration frequency and motor temperature value. This can cope with the nonlinear deterioration characteristics of mechanical failure, provide enhanced protection for the joint module in extreme cases, thereby effectively preventing further deterioration of mechanical failure and improving the safety of the joint module under mechanical failure.

[0120] In another embodiment, controlling the motor of the joint module to adjust the output torque according to the target torque value includes: controlling the motor of the joint module to linearly transition the output torque from the current torque value to the target torque value according to the gait cycle of the joint module.

[0121] The gait cycle can include the time required for the robot to complete a full gait cycle. For example, the gait cycle of a trot is about 0.3 seconds, and the gait cycle of a walking gait is about 0.5 seconds.

[0122] The process of linearly transitioning the output torque from the current torque value to the target torque value may include gradually changing the output torque from the current torque value to the target torque value at a constant rate of change.

[0123] For example, the robot can obtain the torque change based on the difference between the current torque value and the target torque value; divide the torque change by the number of control cycles corresponding to the gait cycle to obtain the torque change step size for each control cycle; within each control cycle, the robot reduces the output torque command value of the motor by one torque change step size until the output torque reaches the target torque value.

[0124] In practical applications, the robot's joint controller (JC) no longer responds to the original T. base The request is not based on the calculated T. limit As the new target value for torque closed-loop control, the torque derating process is not instantaneous, but rather linearly transitions from the current torque to the new T within a gait cycle (e.g., 200ms). limitThis system achieves smooth torque derating, preventing robot instability and severe vibrations caused by sudden torque changes. Simultaneously, when a joint controller (JC) begins to dedrate, the main control unit (MCU) is immediately notified. The MCU's gait coordinator then redistributes torque output from other healthy joints and may proactively change the gait (e.g., forcibly switching from a Gallop gait to a more stable Trot gait) to compensate for the performance loss of the faulty joint. This ensures the robot doesn't fall due to a single joint malfunction, achieving coordinated gait control. Therefore, the joint controllers (JCs) dedrate linearly according to the gait cycle, and the MCU synchronously adjusts the gait to ensure a smooth and safe reduction in torque while maintaining overall robot balance. It is evident that when a joint dedrates due to a fault, the MCU proactively adjusts the overall gait and outputs of other joints, forming system-level coordinated compensation. This ensures mechanical safety while maximizing the robot dog's motion stability.

[0125] The technical solution of this application embodiment avoids the impact of sudden torque changes on the mechanical system and motion stability by linearly transitioning the output torque according to the gait cycle. If the output torque drops sharply in an instant, it may cause the joint to suddenly lose support, causing the robot to become unstable or fall. The linear transition makes the torque change smooth, and the robot can adjust the gait and the output of other joints in time to maintain the overall motion balance.

[0126] In another embodiment, after controlling the motor of the joint module to adjust the output torque according to the target torque value, the method further includes: when the joint vibration frequency, motor temperature value and joint load of the joint module meet the preset torque recovery conditions, controlling the motor of the joint module to gradually restore the output torque from the current torque value to the reference torque value of the joint module at preset intervals.

[0127] The preset torque recovery condition is a set of conditions used to determine whether the mechanical fault condition has improved. The preset torque recovery condition may include joint vibration condition, temperature condition, load condition, and time condition.

[0128] In practical applications, a torque recovery verification mechanism can be introduced. Moreover, this torque recovery is not a simple reverse derating process, but requires the simultaneous fulfillment of conditions such as vibration subsidence, load reduction, and temperature drop, and must be maintained for a certain period of time. This ensures that the mechanical system truly returns to a safe operating state, avoids the oscillation and unpredictability of protection, and solves the problem of secondary impact on the mechanical system that has not yet fully cooled down due to the immediate restoration of full torque output after the fault is cleared in traditional technology.

[0129] For example, the joint controller (JC) and the trunk main controller (MCU) continuously monitor whether the following preset torque recovery conditions are met simultaneously: vibration condition, joint vibration frequency ≤1500Hz (for bearing wear); load condition, joint load ≤1.2 * rated value (usually achieved by the MCU switching to low-impact gait); temperature condition, motor temperature ≤80°C; and the above conditions must be stable and sustained for ≥3 complete gait cycles (e.g., approximately 1.5 seconds in Trot gait). After verifying the vibration, load, and temperature conditions, torque can be recovered in stages.

[0130] The preset interval may include a preset time interval or a preset number of gait cycle intervals. For example, the robot may increase the output torque every 1 to 2 gait cycles to gradually restore the output torque from the current torque value to the reference torque value of the joint module.

[0131] For example, once the preset torque recovery condition is met, the joint controller (JC) begins a step-by-step recovery process. During the recovery process, the torque does not recover to the reference torque value T all at once. base Instead of increasing the output torque by 10% every 1-2 gait cycles, the joint controller (JC) rigorously monitors fault characteristics during each increase. If any indicator (such as vibration) deteriorates again, the recovery process is immediately paused, and the torque may even be drated again, forming a closed-loop protection mechanism. Ultimately, the output torque recovers to the reference torque value T. base When the fault characteristics are equal and have not recurred, the fault tolerance mode can be exited and normal operation can be restored.

[0132] The technical solution of this application embodiment restores the torque value in stages when the joint vibration frequency, motor temperature value and joint load of the joint module meet the preset torque recovery conditions. This ensures that the performance is carefully restored only after the dangerous state is completely eliminated. By introducing a safe recovery mechanism, the safety of joint torque is improved. Torque recovery must simultaneously meet three conditions: vibration subsidence, load reduction and temperature drop, and remain stable for a period of time to avoid secondary damage caused by excessively rapid recovery, thus ensuring the high reliability of fault-tolerant operation.

[0133] The parameters appearing in the embodiments of this application are described in Tables 3 and 4.

[0134] Table 3

[0135]

[0136] Table 4

[0137]

[0138] For the convenience of those skilled in the art, Figure 3A logic diagram of a torque control method for a robot joint module is provided as an example. It can be seen that an intelligent closed-loop torque control system integrating perception, diagnosis, decision-making, execution, and verification is constructed, rather than a simple torque limitation. The entire process includes high-frequency data acquisition and feature extraction to achieve refined fault modeling; identification of fault-sensitive working conditions, linking working condition perception with fault analysis; dynamic torque limit calculation, realizing intelligent torque control through multi-physics field fusion; torque smoothing derating and gait coordination, improving the safety and stability of the derating process; and a condition-based collaborative recovery mechanism and step-by-step torque recovery, enhancing the safety of torque recovery.

[0139] The technical solution of this application embodiment, through active fault-tolerant control, enables the robot to continue performing critical tasks in a "limping" manner instead of immediately becoming paralyzed after a mechanical failure, greatly improving the survival rate and task completion rate in harsh or dangerous environments. Based on a precise diagnostic intelligent torque derating model, torque can be dynamically limited to match the "working with defects" state of mechanical components, avoiding secondary damage caused by excessive load, thereby significantly delaying the failure process of critical components such as bearings and gearboxes and extending the overall lifespan of the machine. It not only responds after a failure but also provides proactive protection by identifying "sensitive working conditions." At the same time, continuously recorded fault characteristic data provides an accurate basis for predictive maintenance, enabling a shift from "periodic inspection" to "on-demand maintenance" and reducing maintenance costs. The torque derating process is smooth and closely integrated with gait coordination control, ensuring the robot's motion balance under performance constraints and effectively preventing safety accidents such as gait swaying, instability, or even falls caused by sudden joint jamming or torque mutation.

[0140] In another embodiment, such as Figure 4 As shown, a torque control method for a robot joint module is provided, which is applied to... Figure 1 Taking robot 102 as an example, the explanation includes the following steps:

[0141] Step S402: In response to a mechanical fault event in the robot's joint module, identify the fault level of the mechanical fault event.

[0142] Step S404: Determine the initial fault coefficient of the joint module according to the fault level; determine the working condition of the joint module according to the task execution information of the joint module; if the working condition meets the fault-sensitive working condition, adjust the initial fault coefficient to obtain the mechanical fault coefficient.

[0143] The task execution information includes at least one of the following: gait type of the joint module, load status, and task duration.

[0144] Among them, the mechanical failure coefficient is higher than the initial failure coefficient; the mechanical failure coefficient is negatively correlated with the target torque value of the joint module.

[0145] Step S406: Determine the mechanical risk parameters based on the torque fluctuation of the joint module and the joint vibration frequency of the joint module; determine the temperature risk parameters based on the difference between the motor temperature value and the reference temperature value of the joint module; determine the torque adjustment coefficient based on the product of the mechanical risk parameters, the temperature risk parameters and the mechanical failure coefficient.

[0146] Step S408: Determine the torque limit of the joint module based on the reference torque value and torque adjustment coefficient of the joint module; adjust the torque limit of the joint module according to an adjustment ratio that matches the preset frequency amount for each increase in the joint vibration frequency of the joint module to obtain the target torque value of the joint module; adjust the torque limit of the joint module according to an adjustment ratio that matches the preset temperature amount for each increase in the motor temperature value of the joint module relative to the safe temperature value to obtain the target torque value of the joint module.

[0147] In step S410, according to the gait cycle of the joint module, the motor of the joint module is controlled to linearly transition the output torque from the current torque value to the target torque value.

[0148] Step S412: When the joint vibration frequency, motor temperature and joint load of the joint module meet the preset torque recovery conditions, the motor of the joint module is controlled to gradually restore the output torque from the current torque value to the reference torque value of the joint module according to the preset interval.

[0149] It should be noted that the specific limitations of the above steps can be found in the specific limitations of a torque control method for a robot joint module described above.

[0150] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0151] Based on the same inventive concept, this application also provides a torque control device for a robot joint module to implement the torque control method of the robot joint module described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of the torque control device embodiments of one or more robot joint modules provided below can be found in the limitations of the torque control method of the robot joint module above, and will not be repeated here.

[0152] In one exemplary embodiment, such as Figure 5 As shown, a torque control device for a robot joint module is provided, comprising:

[0153] The identification module 510 is used to identify the fault level of a mechanical fault event in response to a mechanical fault event in the robot's joint module.

[0154] The determination module 520 is used to determine the mechanical fault coefficient of the joint module according to the fault level; the mechanical fault coefficient is used to characterize the degree of influence of mechanical fault events on the torque control of the joint module; the target torque value of the joint module is determined based on the mechanical fault coefficient, the motor temperature value of the joint module and the joint vibration frequency of the joint module.

[0155] The control module 530 is used to control the motor of the joint module to adjust the output torque according to the target torque value.

[0156] In one embodiment, the determining module 520 is specifically configured to: determine an initial fault coefficient of the joint module based on the fault level; determine the operating condition of the joint module based on the task execution information of the joint module; the task execution information includes at least one of the gait type, load state, and task duration of the joint module; adjust the initial fault coefficient to obtain a mechanical fault coefficient when the operating condition meets the fault-sensitive operating condition; the mechanical fault coefficient is higher than the initial fault coefficient; and the mechanical fault coefficient is negatively correlated with the target torque value of the joint module.

[0157] In one embodiment, the determining module 520 is specifically configured to: determine mechanical risk parameters based on the torque fluctuation of the joint module and the joint vibration frequency of the joint module; determine temperature risk parameters based on the difference between the motor temperature value and the reference temperature value of the joint module; determine a torque adjustment coefficient based on the product of the mechanical risk parameters, the temperature risk parameters, and the mechanical failure coefficient; and determine a target torque value of the joint module based on the reference torque value of the joint module and the torque adjustment coefficient.

[0158] In one embodiment, the determining module 520 is specifically configured to determine the torque limit of the joint module based on the reference torque value of the joint module and the torque adjustment coefficient; adjust the torque limit of the joint module according to an adjustment ratio matching the preset frequency amount for each increase in the joint vibration frequency of the joint module to obtain the target torque value of the joint module; and adjust the torque limit of the joint module according to an adjustment ratio matching the preset temperature amount for each increase in the motor temperature value of the joint module relative to the safe temperature value to obtain the target torque value of the joint module.

[0159] In one embodiment, the control module 530 is specifically configured to control the motor of the joint module to linearly transition the output torque from the current torque value to the target torque value according to the gait cycle of the joint module.

[0160] In one embodiment, the control module 530 is specifically configured to, when the joint vibration frequency, the motor temperature value, and the joint load of the joint module meet the preset torque recovery conditions, control the motor of the joint module to gradually restore the output torque from the current torque value to the reference torque value of the joint module at preset intervals.

[0161] Each module in the torque control device of the aforementioned robot joint module can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0162] In one exemplary embodiment, a computer device is provided, which may be a robot, and its internal structure diagram may be as follows: Figure 6As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a torque control method for a robot joint module. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0163] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0164] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0165] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0166] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0167] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0168] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic resistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0169] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0170] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A torque control method for a robot joint module, characterized in that, The method includes: In response to a mechanical failure event in the robot's joint module, the failure level of the mechanical failure event is identified; Based on the fault level, determine the initial fault coefficient of the joint module; The working condition of the joint module is determined based on the task execution information of the joint module; the task execution information includes at least one of the joint module's gait type, load status, and task duration. When the operating condition meets the fault-sensitive condition, the initial fault coefficient is adjusted to obtain a mechanical fault coefficient; the mechanical fault coefficient is higher than the initial fault coefficient; the mechanical fault coefficient is negatively correlated with the target torque value of the joint module; the mechanical fault coefficient is used to characterize the degree of influence of the mechanical fault event on the torque control of the joint module. Mechanical risk parameters are determined based on the torque fluctuation of the joint module and the joint vibration frequency of the joint module; The temperature risk parameter is determined based on the difference between the motor temperature value and the reference temperature value of the joint module; The torque adjustment coefficient is determined based on the product of the mechanical risk parameter, the temperature risk parameter, and the mechanical failure coefficient. The torque limit of the joint module is determined based on the reference torque value of the joint module and the torque adjustment coefficient. When the joint vibration frequency of the joint module increases by a preset amount, the torque limit of the joint module is adjusted according to an adjustment ratio that matches the preset amount, so as to obtain the target torque value of the joint module. When the motor temperature of the joint module increases by a preset amount relative to the safe temperature, the torque limit of the joint module is adjusted according to an adjustment ratio that matches the preset temperature to obtain the target torque value of the joint module. According to the gait cycle of the joint module, the motor of the joint module is controlled to linearly transition the output torque from the current torque value to the target torque value.

2. The method according to claim 1, characterized in that, After controlling the motor of the joint module to adjust its output torque according to the target torque value, the method further includes: When the joint vibration frequency, the motor temperature, and the joint load of the joint module meet the preset torque recovery conditions, the motor of the joint module is controlled to gradually restore the output torque from the current torque value to the reference torque value of the joint module at preset intervals.

3. A torque control device for a robot joint module, characterized in that, The device includes: The identification module is used to identify the fault level of a mechanical fault event in response to a mechanical fault event in the robot's joint module. A determination module is used to determine an initial fault coefficient of the joint module based on the fault level; determine the operating condition of the joint module based on the task execution information of the joint module; the task execution information includes at least one of the gait type, load state, and task duration of the joint module; if the operating condition meets the fault-sensitive operating condition, adjust the initial fault coefficient to obtain a mechanical fault coefficient; the mechanical fault coefficient is higher than the initial fault coefficient; the mechanical fault coefficient is negatively correlated with the target torque value of the joint module; the mechanical fault coefficient is used to characterize the degree of influence of the mechanical fault event on the torque control of the joint module. The determining module is further configured to: determine mechanical risk parameters based on the torque fluctuation of the joint module and the joint vibration frequency of the joint module; determine temperature risk parameters based on the difference between the motor temperature value and the reference temperature value of the joint module; determine a torque adjustment coefficient based on the product of the mechanical risk parameters, the temperature risk parameters, and the mechanical failure coefficient; determine a torque limit of the joint module based on the reference torque value of the joint module and the torque adjustment coefficient; adjust the torque limit of the joint module according to an adjustment ratio matching the preset frequency amount for each increase in the joint vibration frequency of the joint module, to obtain a target torque value of the joint module; and adjust the torque limit of the joint module according to an adjustment ratio matching the preset temperature amount for each increase in the motor temperature value of the joint module relative to the safe temperature value, to obtain a target torque value of the joint module. The control module is used to control the motor of the joint module to linearly transition the output torque from the current torque value to the target torque value according to the gait cycle of the joint module.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 2.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 2.

6. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 2.

Images