Motion object driven positioning error compensation method and system

Abstract

The application provides a motion object driving positioning error compensation method and system, and relates to the technical field of data processing. The method obtains the expected output angle of each execution structure of a motion object, motor angle feedback information of each execution structure driving unit and execution structure running state information, then updates the return stroke hysteresis state machine corresponding to each execution structure based on the motor angle feedback information and the execution structure running state information, obtains the return stroke state information of each execution structure, and determines the compensated motor reference angle of each execution structure according to the expected output angle and the return stroke state information, so as to compensate the positioning error caused by the return stroke gap and hysteresis nonlinearity in the transmission chain of the execution structure through the compensated motor reference angle, effectively reduce the motion object driving positioning error caused by the return stroke gap and hysteresis nonlinearity, and further improve the positioning accuracy, repeat positioning accuracy and trajectory continuity of each execution structure and the end effector of the motion object.

Description

Technical Field

[0001] This invention relates to data processing technology, and more particularly to a method and system for compensating for positioning errors driven by a moving object. Background Technology

[0002] Existing industrial robots and collaborative robots widely use servo motors in conjunction with transmission mechanisms such as gear reducers, harmonic reducers or planetary reducers to form multiple joint drive units.

[0003] Typically, motor angle feedback information is obtained through an encoder installed on the motor side. Based on this, servo position loops, velocity loops, and current loops are constructed with the motor angle as the controlled variable to achieve position control and trajectory tracking of each joint of the robot. However, due to unavoidable mechanical factors such as reducer meshing backlash, elastic torsion of shafts and couplings, and differences between static and dynamic friction, joint transmission chains generally exhibit directionally correlated backlash and hysteresis nonlinearity. Summary of the Invention

[0004] This application provides a method and system for compensating for motion object driven positioning errors, which reduces robot driven positioning errors caused by backlash and hysteresis nonlinearity, thereby improving the positioning accuracy, repeatability, and trajectory continuity of robot joints and end effectors.

[0005] In a first aspect, this application provides a method for compensating for positioning errors driven by a moving object, applied to a moving object control system including multiple actuator drive units; the method includes:

[0006] Obtain the expected output angle of each execution structure of the moving object;

[0007] Obtain the motor angle feedback information and execution structure operating status information of each execution structure drive unit;

[0008] Based on the motor angle feedback information and the operating status information of the execution structure, the return hysteresis state machine corresponding to each execution structure is updated to obtain the return status information of each execution structure.

[0009] Based on the desired output angle and the return state information, the compensated motor reference angle of each execution structure is determined.

[0010] Secondly, this application provides a motion object control system, comprising:

[0011] The angle acquisition module is used to obtain the expected output angle of each execution structure of the moving object;

[0012] The status acquisition module is used to acquire the motor angle feedback information and the operating status information of each execution structure drive unit;

[0013] The information processing module is used to update the return hysteresis state machine corresponding to each execution structure based on the motor angle feedback information and the execution structure operation status information, so as to obtain the return status information of each execution structure.

[0014] The motor control module is used to determine the compensated motor reference angle for each actuator based on the desired output angle and the return state information.

[0015] Thirdly, this application provides an electronic device, comprising:

[0016] Processor; and,

[0017] Memory for storing the executable instructions of the processor;

[0018] The processor is configured to perform any of the possible methods described in the first aspect by executing the executable instructions.

[0019] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement any of the possible methods described in the first aspect.

[0020] The motion object driving positioning error compensation method and system provided in this application obtains the expected output angle of each joint of the robot, as well as the motor angle feedback information and joint running state information of each joint driving unit. Then, based on the motor angle feedback information and joint running state information, the return hysteresis state machine corresponding to each joint is updated to obtain the return state information of each joint. Then, based on the expected output angle and return state information, the compensated motor reference angle of each joint is determined. Thus, the positioning error caused by the return backlash and hysteresis nonlinearity in the joint transmission chain is compensated by the compensated motor reference angle, so as to effectively reduce the robot driving positioning error caused by the return backlash and hysteresis nonlinearity, thereby improving the positioning accuracy, repeatability, and trajectory continuity of each joint of the robot and the end effector. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0022] Figure 1 This is a schematic flowchart illustrating a motion object-driven positioning error compensation method according to an example embodiment of this application;

[0023] Figure 2 yes Figure 1 A flowchart illustrating the specific implementation of S130 in the illustrated embodiment;

[0024] Figure 3 This is a flowchart illustrating a motion object-driven positioning error compensation method according to another example embodiment of this application;

[0025] Figure 4 This is a schematic diagram of the structure of a motion object control system according to an example embodiment of this application;

[0026] Figure 5 This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application.

[0027] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] Figure 1 This is a flowchart illustrating a motion object-driven positioning error compensation method according to an example embodiment of this application. Figure 1 As shown, the method provided in this embodiment includes:

[0030] S110, Obtain the expected output angle of each execution structure of the moving object.

[0031] In this step, the desired output angles of each joint of the robot are obtained. The desired output angles are the joint output axis reference angles calculated by the kinematics solving module based on the target end pose or target trajectory.

[0032] Specifically, based on the target end-effector pose or trajectory output by the external task planning module, the corresponding end-effector pose command information can be obtained. This command information includes at least one or both of the displacement and attitude components of the end-effector pose. Then, based on the end-effector pose command information and preset robot structural parameters, kinematic equations describing the mapping relationship between the end-effector pose and the output axis angles of each robot joint are constructed. Next, the end-effector pose command information is substituted into the kinematic equations, and the kinematics solving module is called to numerically solve the equations, obtaining the reference angles of each joint output axis corresponding to the end-effector pose command information, which are then used as the desired output angles of each robot joint.

[0033] S120. Obtain motor angle feedback information and execution structure operation status information of each execution structure drive unit.

[0034] In this step, the motor angle feedback information and joint operating status information of each joint drive unit are obtained. The joint operating status information includes at least the direction of motor angle change within adjacent control cycles.

[0035] Specifically, the motor shaft angle measurement value can be obtained by an angle detection device installed on the motor shaft of each joint drive unit. The angle detection device includes at least one of an absolute encoder, an incremental encoder, or a rotary transformer, and the motor shaft angle measurement value serves as motor angle feedback information.

[0036] Then, the motor angle feedback information is periodically sampled according to the preset control cycle, and the motor angle feedback information of the current control cycle and the motor angle feedback information of the previous control cycle are recorded respectively.

[0037] Then, based on the difference between the motor angle feedback information of the current control cycle and the motor angle feedback information of the previous control cycle, the amount of motor angle change in the current control cycle is determined, and the direction of motor angle change is determined based on the sign of the amount of motor angle change.

[0038] Finally, the direction of the motor angle change and the corresponding amount of the motor angle change are stored and updated as part of the joint operating status information, so that the joint operating status information includes at least the direction of the motor angle change in adjacent control cycles.

[0039] S130. Based on the motor angle feedback information and the operating status information of the execution structure, update the return hysteresis state machine corresponding to each execution structure to obtain the return status information of each execution structure.

[0040] In this step, based on the motor angle feedback information and joint operating status information, the return hysteresis state machine corresponding to each joint is updated to obtain the return status information of each joint. The return status information includes at least the meshing state and the equivalent clearance.

[0041] Optionally, the aforementioned multiple joint drive units include joint drive units employing at least one of the transmission mechanisms of gear reducers, harmonic reducers, or planetary reducers, and the parameter structure of the return hysteresis state machine is configured according to the type of transmission mechanism.

[0042] Specifically, when the joint drive unit uses a gear reducer as the transmission mechanism, a first set of parameters is configured for the return hysteresis state machine of the corresponding joint, including the forward clearance threshold, the reverse clearance threshold, the equivalent angle parameter of the tooth flank clearance, and the meshing stiffness parameter of the gear pair. The update step size of the equivalent clearance occupancy and the meshing state switching conditions are determined according to the first set of parameters.

[0043] When the joint drive unit uses a harmonic reducer as the transmission mechanism, a second set of parameters is configured for the return hysteresis state machine of the corresponding joint, including positive hysteresis offset parameters, reverse hysteresis offset parameters, flexible wheel torsional stiffness parameters, and periodic comprehensive error correction parameters. The update method of the virtual output angle in the positive meshing state and the reverse meshing state is determined according to the second set of parameters.

[0044] When the joint drive unit uses a planetary reducer as the transmission mechanism, a third set of parameters is configured for the return hysteresis state machine of the corresponding joint, including the planetary gear system comprehensive return clearance parameters, the carrier support stiffness parameters, and the multi-stage meshing error distribution coefficient. The mapping relationship between the equivalent clearance occupancy and the virtual output angle under the clearance transition state is determined based on the third set of parameters.

[0045] Furthermore, when the joint drive unit employs two or more transmission mechanisms, such as gear reducers, harmonic reducers, or planetary reducers, a set of parameters matching the transmission mechanism type is configured for each corresponding joint drive unit, so that the return hysteresis state machine can characterize the corresponding return backlash and hysteresis nonlinear characteristics under different types of transmission mechanisms.

[0046] Figure 2 yes Figure 1 The illustrated embodiment shows a flowchart illustrating the specific implementation of S130. For example... Figure 2 As shown, in one possible implementation of S130 above, it may include:

[0047] S131, maintain a return hysteresis state machine for each joint.

[0048] In this step, a return hysteresis state machine is maintained for each joint. The states of the return hysteresis state machine include at least one of the following: forward engagement state, reverse engagement state, and clearance transition state.

[0049] Specifically, each joint of the robot can be assigned an independent return hysteresis state machine data structure. The return hysteresis state machine data structure stores at least one or more of the following: the current engagement state identifier, the equivalent clearance occupancy, the motor angle feedback information of the previous control cycle, and the direction of the motor angle change in the previous control cycle.

[0050] Then, when the robot is powered on and initialized or the control system is reset, the engagement state flag of the return hysteresis state machine corresponding to each joint is initialized to the preset initial engagement state, and the equivalent clearance occupancy is initialized to zero or the preset initial clearance occupancy.

[0051] Next, within each control cycle, based on the motor angle feedback information and joint operating status information, the meshing status identifier, equivalent clearance occupancy, and relevant information from the previous control cycle in the return hysteresis state machine data structure corresponding to each joint are updated, so that the return hysteresis state machine can continuously characterize the return status information of each joint in the current control cycle.

[0052] S132. Determine whether the current joint has changed its direction of movement.

[0053] In this step, based on the sign of the motor angle change in the current control cycle, the sign of the motor angle change in the previous control cycle, and the preset equivalent clearance parameter, it is determined whether the joint has changed its direction of motion.

[0054] Specifically, within each control cycle, the motor angle change in the current control cycle can be determined based on the difference between the motor angle feedback information of the current control cycle and the motor angle feedback information of the previous control cycle, and the motor angle change or direction of change stored in the previous control cycle can be extracted.

[0055] Then, the sign of the change in motor angle in the current control cycle is compared with the sign of the change in motor angle in the previous control cycle. When the signs are opposite and the absolute value of the change in motor angle in the current control cycle is greater than the preset minimum effective change threshold, it is determined that the current joint has changed its movement direction.

[0056] When it is determined that the current joint has changed direction of motion, the reference direction of the equivalent clearance occupancy is reset or updated in combination with the preset equivalent clearance parameter corresponding to the joint, so as to be used for subsequent clearance transition state determination and compensation.

[0057] It is worth noting that in high-bandwidth servo systems such as industrial robots and collaborative robots, due to quantization errors and noise in the motor encoder, the position and velocity loops will produce minute correction movements near zero speed to suppress interference and improve tracking accuracy. These micro-vibrations often manifest as frequent flipping of the sign of the motor angle change within adjacent control cycles. If the motion direction switch is triggered immediately based solely on the sign change of a single control cycle, the return hysteresis state machine will frequently jitter between the engagement state and the gap transition state, causing the gap compensation to be incorrectly released or rebuilt. This significantly amplifies the drive positioning error and induces jitter and instability in the end-effector trajectory under low-speed, small-displacement conditions.

[0058] One possible solution is that when the sign of the motor angle change in the current control cycle is detected to be opposite to that in the previous control cycle, the return hysteresis state machine is not immediately switched from the forward engagement state or the reverse engagement state to the gap transition state. Instead, the current joint is marked as a candidate state for motion direction switching. Then, in the candidate state for motion direction switching, the absolute value of the motor angle change with the motion direction confirmed in the previous control cycle is accumulated to obtain the cumulative reverse angle change, and the corresponding control cycle duration is accumulated to obtain the cumulative reverse duration.

[0059] When the cumulative change in the reverse angle is greater than or equal to the preset minimum effective reverse angle threshold, and the cumulative reverse duration is greater than or equal to the preset minimum effective reverse duration threshold, it is determined that the current joint has undergone an effective motion direction switch, and the return hysteresis state machine is switched from the forward engagement state or the reverse engagement state to the gap transition state. The minimum effective reverse angle threshold is configured based on the preset equivalent gap parameter of the corresponding joint and the resolution of the motor angle feedback information, ensuring that the minimum effective reverse angle threshold is not less than a certain number of encoder minimum resolution units, and is a predetermined proportion of the estimated return gap angle of the joint under the current drivetrain configuration. The minimum effective reverse duration threshold is configured based on the control cycle of the control system and the servo bandwidth of the target joint, ensuring that the minimum effective reverse duration threshold corresponds to no less than a preset number of continuous control cycles, to avoid instantaneous sign flipping caused by encoder quantization jitter and closed-loop control micro-vibration being misjudged as a motion direction switch.

[0060] When the sign of the motor angle change is detected to be consistent with the motion direction confirmed in the previous control cycle during the motion direction switching candidate state, and the cumulative reverse angle change and / or cumulative reverse duration have not reached the corresponding threshold, the motion direction switching candidate state is cancelled and the original engagement state remains unchanged.

[0061] In the above implementation, on the one hand, by not immediately changing the state of the hysteresis state machine when the angle change symbol is detected to reverse, but by switching the candidate state of the motion direction, the direction switching determination is extended from an instantaneous symbol event of a single control cycle to a physical quantity determination process that accumulates over a time window.

[0062] On the other hand, by accumulating the absolute value of the motor angle change that is opposite to the original confirmed direction of motion, a cumulative reverse angle change variable is constructed, and at the same time, the corresponding control cycle duration is accumulated to construct a cumulative reverse duration variable, thereby forming a two-dimensional judgment mechanism of amplitude threshold and time threshold.

[0063] By setting a minimum effective reverse angle threshold and a minimum effective reverse duration threshold, the real direction switching is modeled as a disturbance that must cross a minimum displacement related to the equivalent gap in the angle space and last for at least a few control cycles in time, in order to distinguish it from noise-type symbol flipping that only occurs in a few sampling cycles and whose amplitude is usually close to the encoder's resolution limit.

[0064] During actual operation, if the cumulative amount of reverse motion does not reach one of these two thresholds, the system reverts from the candidate state and maintains the original meshing state, avoiding erroneous switching of the return hysteresis state machine. Only when the reverse motion meets the conditions in both amplitude and duration is it considered that the return clearance in the original direction within the transmission chain has been substantially released, and it is necessary to enter the clearance transition state to establish clearance occupancy in the new direction.

[0065] This effectively filters out short-term, small-amplitude sign flips caused by encoder noise, quantization errors, and small closed-loop corrections, reduces the probability of false triggering of the return hysteresis state machine, and ensures that the release and reconstruction of the return gap are only executed when there is a physically meaningful change in the direction of motion. In turn, it effectively suppresses compensation jitter and mechanical vibration under low-speed, small-amplitude motion conditions, and effectively improves the stability of robot joint drive positioning and end-effector positioning accuracy.

[0066] In another possible solution, the desired velocity information or desired output angle increment information of each joint can be acquired, and the desired motion direction of the joint can be determined based on the sign of the desired velocity information or the sign of the desired output angle increment. Then, the current joint is marked as a motion direction switching candidate state only when the desired motion direction of the joint is opposite to the motion direction confirmed in the previous control cycle, and the sign of the motor angle change is consistent with the desired motion direction. The cumulative reverse angle change and the cumulative reverse duration are then determined. When the desired motion direction has not reversed or the sign of the motor angle change is not consistent with the desired motion direction, even if a reverse change in the sign of the motor angle change is detected in an adjacent control cycle, the motion direction switching candidate state is not triggered to avoid micro-vibrations in direction caused by minor corrections in the position loop or velocity loop being identified as valid motion direction switching.

[0067] The above solution firstly sets a direction determination dead zone for the motor angle change in adjacent control cycles. If the absolute value is less than the dead zone threshold, it is considered a zero step, thus performing preliminary filtering of the encoder's high-frequency micro-jitter at the signal level. Simultaneously, direction hysteresis thresholds are configured for switching from forward engagement to reverse engagement and from reverse engagement to forward engagement, respectively. This gives direction determination a hysteresis characteristic; only when the cumulative reverse angle change exceeds the hysteresis threshold is the candidate state for motion direction switching allowed, thus suppressing the sensitive behavior of switching even with small reverse movements.

[0068] Secondly, by constructing candidate states for motion direction switching, the absolute value of the motor angle change opposite to the previously confirmed direction and the corresponding duration are continuously accumulated in this state. This achieves dual-threshold integral judgment of the angle and time of the reverse motion, preventing brief reverse displacements from being mistakenly identified as valid direction switching. Simultaneously, the sign of the desired speed or desired output angle increment is used as the desired motion direction. Entering the candidate state is only allowed when the desired direction matches the actual reverse change direction, ensuring that the direction switching logic is consistent with the upper-level control command logic.

[0069] Finally, by monitoring the changing trend of the equivalent gap occupancy in real time during the candidate state, it is determined whether it continues to decrease in the reverse direction and approaches zero. This trend is logically combined with the angle and time determination results. Only when the mechanical equivalent variable, geometric variable and time variable all meet the preset conditions, that is, when the gap occupancy, reverse angle and duration all meet the preset conditions, is it confirmed that a real return gap release and motion direction switch has occurred, thereby triggering the return hysteresis state machine transition.

[0070] In another possible solution, a dead-zone threshold can be set for the angle change in adjacent control cycles calculated from the motor angle feedback information. When the absolute value of the motor angle change in adjacent control cycles is less than the preset direction determination dead-zone threshold, the angle change is considered zero and is not used to update the sign of the motor angle change, thus preventing the motion direction switching determination from being triggered. Then, independent direction hysteresis thresholds are set for the motion direction switching corresponding to the transition from forward engagement to reverse engagement and from reverse engagement to forward engagement. Under a certain confirmed motion direction, the motion direction switching candidate state is only allowed when the cumulative angle change opposite to the confirmed motion direction exceeds the corresponding direction hysteresis threshold. This forms a determination interval with hysteresis characteristics during the direction determination process, suppressing small-amplitude direction reversals caused by encoder noise, quantization error, and servo control micro-vibration.

[0071] On the one hand, at the signal processing layer, by setting a preset direction determination dead zone threshold for the motor angle change in adjacent control cycles, all angle changes with absolute values ​​below the threshold are uniformly regarded as zero changes and the angle change sign is not updated. This is equivalent to applying a first-order amplitude filter to the encoder angle increment signal, so that the small-amplitude angle swing caused by the encoder's quantization unit level jitter and closed-loop micro-vibration will not enter the motion direction switching determination logic link.

[0072] On the other hand, different directional hysteresis thresholds are preset for the two types of directional switching paths: switching from forward meshing to reverse meshing and switching from reverse meshing to forward meshing. When the joint is in a certain confirmed movement direction, the absolute value of the motor angle change opposite to the confirmed movement direction is continuously accumulated. Only when the accumulated reverse angle change exceeds the corresponding directional hysteresis threshold is it considered that the joint has physically generated a sufficiently large reverse movement, allowing it to enter the movement direction switching candidate state.

[0073] By determining the cumulative reverse angle and hysteresis threshold, a hysteresis loop is formed in the direction determination space: for a joint to move from the forward direction to be determined as a reverse movement, it needs to cross a reverse angle threshold higher than zero, and vice versa, thus effectively avoiding frequent direction reversals near small back-and-forth swings.

[0074] S133. When a change in motion direction is detected, the return hysteresis state machine is switched from the forward meshing state or the reverse meshing state to the gap transition state.

[0075] Specifically, when it is determined that the joint has changed its direction of motion based on the sign of the change in motor angle within the current control cycle and the previous control cycle, and the return hysteresis state machine corresponding to the current joint is in a forward engagement state or a reverse engagement state, the engagement state identifier of the return hysteresis state machine is updated to a gap transition state.

[0076] While updating the engagement state identifier to the clearance transition state, the direction of the motor angle change within the current control cycle is recorded as the new target engagement direction, and the sign or reference direction of the equivalent clearance occupancy is reset according to the target engagement direction.

[0077] During the gap transition state, the cumulative value of the motor angle change in multiple subsequent control cycles is continuously monitored and compared with the preset positive gap threshold or negative gap threshold to determine when gap compensation is completed and trigger the switch from the gap transition state to the positive engagement state or the negative engagement state.

[0078] S134. Update the equivalent clearance occupancy based on the change in motor angle.

[0079] Specifically, in each control cycle when the return hysteresis state machine is in the intermittent transition state, the difference between the motor angle feedback information of the current control cycle and the motor angle feedback information of the previous control cycle can be obtained as the motor angle change in the current control cycle.

[0080] Then, based on the sign of the motor angle change in the current control cycle, it is compared with the target meshing direction recorded in the return hysteresis state machine. When the two directions are consistent, the motor angle change is accumulated to the current equivalent clearance occupancy by a preset proportional coefficient. When the two directions are inconsistent or the absolute value of the motor angle change is less than the preset minimum effective change threshold, the equivalent clearance occupancy remains unchanged or is gradually attenuated and updated according to the preset attenuation coefficient.

[0081] Finally, during the process of updating the equivalent gap occupancy, the updated equivalent gap occupancy is stored in the return hysteresis state machine data structure of the corresponding joint for subsequent comparison with the forward gap threshold or the reverse gap threshold.

[0082] S135. Switch the return hysteresis state machine from the gap transition state to the forward meshing state or the reverse meshing state.

[0083] When the equivalent clearance occupancy reaches the preset forward clearance threshold or reverse clearance threshold, the return hysteresis state machine will switch from the clearance transition state to the forward engagement state or the reverse engagement state.

[0084] Specifically, in each control cycle during the clearance transition state, the absolute value of the current equivalent clearance occupancy is compared with the clearance threshold corresponding to the current target meshing direction. When the absolute value of the equivalent clearance occupancy is greater than or equal to the forward clearance threshold or the reverse clearance threshold, it is determined that the return clearance compensation in the corresponding direction has been completed.

[0085] Furthermore, when it is determined that the return backlash compensation in the corresponding direction has been completed, the meshing state identifier of the return hysteresis state machine is updated from the backlash transition state to the forward meshing state or the reverse meshing state that is consistent with the current target meshing direction.

[0086] Then, while updating the meshing state identifier to either forward meshing or reverse meshing, the current equivalent clearance occupancy is locked to the target equivalent clearance angle corresponding to the target meshing direction, or adjusted to the target equivalent clearance angle according to a preset convergence strategy, to ensure the continuity and stability of the virtual output angle and subsequent compensation amount after entering the forward meshing or reverse meshing state.

[0087] Furthermore, the update of the return hysteresis state machine corresponding to each joint in S130 can also be achieved by maintaining a virtual output angle for each joint. The virtual output angle is the reducer output shaft angle determined based on motor angle feedback information, equivalent clearance occupancy, and transmission chain elastic torsional parameters. Specifically, in the forward engagement state, the virtual output angle is updated based on the motor angle feedback information and a preset forward hysteresis offset parameter. In the reverse engagement state, the virtual output angle is updated based on the motor angle feedback information and a preset reverse hysteresis offset parameter. In the clearance transition state, the virtual output angle is updated based on the motor angle change and the transmission chain elastic torsional stiffness parameter.

[0088] Specifically, to maintain the virtual output angle for each joint, a virtual output angle storage unit can be allocated to each joint of the robot. This storage unit records the current virtual output angle value associated with the corresponding joint and the virtual output angle value from the previous control cycle within each control cycle. Then, during robot power-on initialization or control system reset, the virtual output angle corresponding to each joint is initialized to the initial virtual output angle value corresponding to the mechanical zero position of the reducer output shaft, based on the mechanical zero position of each joint, the encoder zero position, and preset alignment compensation parameters. Next, within each control cycle, the virtual output angle corresponding to each joint is updated based on the motor angle feedback information of the current control cycle, the equivalent clearance occupancy, and the pre-calibrated transmission chain elastic torsion parameters, so that the virtual output angle can approximately represent the actual rotation angle of the reducer output shaft.

[0089] Then, to determine the output angle, the motor angle feedback information corresponding to each joint drive unit is obtained, and the motor angle feedback information is converted into the ideal rigid output angle of the reducer output shaft according to the reducer transmission ratio. Next, based on the equivalent clearance occupancy recorded in the return hysteresis state machine within the current control cycle, the clearance angle that has been compensated or not yet compensated in the current meshing direction is determined, and the clearance angle is used as one of the correction terms between the virtual output angle and the ideal rigid output angle. Finally, based on the elastic torsional parameters corresponding to each joint transmission chain and the current motor output torque or estimated torque, the elastic torsional angle compensation amount caused by torsional deformation is determined, and the elastic torsional angle compensation amount is combined with the clearance angle to correct the ideal rigid output angle of the reducer output shaft, thus obtaining the virtual output angle of each joint.

[0090] For the first operating condition, when the return hysteresis state machine corresponding to a certain joint is in the forward engagement state, the motor angle feedback information of the current control cycle of that joint and the forward hysteresis offset parameter corresponding to that joint are acquired. Then, based on the reducer transmission ratio, the motor angle feedback information is converted into the theoretical value of the reducer output shaft angle, and the forward hysteresis offset parameter corresponding to the current forward engagement direction is subtracted from this value to compensate for the angle offset caused by tooth surface elastic contact and micro-engagement hysteresis. Next, the reducer output shaft angle corrected by the forward hysteresis offset parameter is combined with the current equivalent clearance occupancy and transmission chain elastic torsion parameters to update the virtual output angle value in the virtual output angle storage unit of the corresponding joint, so that the virtual output angle can stably reflect the actual position of the reducer output shaft under the forward engagement condition.

[0091] For the second operating condition, when the return hysteresis state machine corresponding to a certain joint is in reverse meshing, the motor angle feedback information of the current control cycle of that joint and the corresponding reverse hysteresis offset parameter can be obtained. Next, based on the reducer transmission ratio, the motor angle feedback information is converted into the theoretical value of the reducer output shaft angle, and the reverse hysteresis offset parameter corresponding to the current reverse meshing direction is subtracted to compensate for the reverse hysteresis angle error caused by changes in the reverse meshing tooth surface contact conditions and material internal friction. Then, the reducer output shaft angle corrected by the reverse hysteresis offset parameter, combined with the current equivalent clearance occupancy and the transmission chain elastic torsion parameter, is used to update the virtual output angle value in the virtual output angle storage unit of the corresponding joint, so that the virtual output angle can continuously and stably represent the actual position of the reducer output shaft under reverse meshing conditions.

[0092] For the third operating condition, when the return hysteresis state machine corresponding to a certain joint is in a gap transition state, the difference between the motor angle feedback information of the current control cycle and the motor angle feedback information of the previous control cycle can be obtained as the motor angle change in the current control cycle. Then, according to the preset transmission chain elastic torsional stiffness parameters, the motor angle change is divided into a free travel component for overcoming the return gap and an effective drive component for generating elastic torsion. When the equivalent gap occupancy has not yet reached the gap threshold in the corresponding direction, most of the motor angle change is used to update the equivalent gap occupancy, and only the effective drive component calculated based on the elastic torsional stiffness is converted into the reducer output shaft angle increment. Furthermore, based on the above conversion, the reducer output shaft angle increment corresponding to the effective drive component is superimposed on the virtual output angle of the previous control cycle to obtain the virtual output angle of the current control cycle. This ensures that the change in the virtual output angle in the gap transition state simultaneously reflects the elastic torsional response and the characteristics of the incompletely closed return gap.

[0093] S140. Based on the desired output angle and return status information, determine the compensated motor reference angle for each actuator.

[0094] In this step, based on the desired output angle and return status information, the compensated motor reference angle or the increment of the compensated motor reference angle for each joint is determined. The compensated motor reference angle is used to compensate for the positioning error caused by the return backlash and hysteresis nonlinearity in the joint drive train.

[0095] First, for joints in positive engagement, the corresponding motor reference angle is determined based on the desired output angle and the positive hysteresis offset parameter.

[0096] Specifically, the desired output angle can be multiplied by a preset joint reduction ratio to obtain the corresponding theoretical motor reference angle. Then, a positive hysteresis offset parameter corresponding to the current positive engagement state of the joint is superimposed on this theoretical motor reference angle to obtain the motor reference angle after considering positive hysteresis compensation. Finally, the positive hysteresis offset parameter is kept constant throughout multiple control cycles to prevent high-frequency position oscillations caused by hysteresis compensation jitter.

[0097] Secondly, for joints in a reverse engagement state, the corresponding motor reference angle is determined based on the desired output angle and the reverse hysteresis offset parameter. The desired output angle can then be multiplied by a preset joint reduction ratio to obtain the corresponding theoretical motor reference angle. Next, the reverse hysteresis offset parameter corresponding to the current joint reverse engagement state is superimposed on the theoretical motor reference angle to obtain the motor reference angle after considering reverse hysteresis compensation. Then, if the joint is detected to be continuously in a reverse engagement state and the direction of motor angle change does not change, the reverse hysteresis offset parameter is kept constant to ensure the stability and continuity of reverse hysteresis compensation.

[0098] Furthermore, for joints in a transitional state with gaps, when the joint drive train has backlash and hysteresis nonlinearity, a period of idle travel often occurs between the motor shaft and the load-side output shaft during the joint's direction of motion change. If, upon detecting a direction change or reaching a certain preset condition, all gap compensation angles are superimposed onto the motor reference angle at once, thereby instantaneously completing gap compensation within a single control cycle, although this instantaneous compensation method can theoretically quickly fill the backlash, in actual servo systems it is extremely prone to causing abrupt changes in the motor reference angle, resulting in spikes in the motor drive control output of the servo driver's position loop, which in turn causes severe acceleration / deceleration, vibration, and impact of the joint.

[0099] Especially in trajectory tracking scenarios involving multi-joint coordinated motion, such sudden angle changes can disrupt the continuity of trajectory velocity and acceleration, causing jitter, trajectory overshoot, and additional load impacts on the mechanical structure of the robot's end effector, affecting the robot's stability and overall lifespan under high-precision trajectory tracking and high-frequency direction switching conditions.

[0100] To address this, the clearance compensation increment can be gradually determined over multiple control cycles based on the preset clearance angle, the current equivalent clearance occupancy, and the preset number of transition allocation cycles, in order to generate the compensated motor reference angle.

[0101] In one specific implementation, a target equivalent clearance angle can be selected from preset forward and reverse clearance angles based on the current direction of joint movement. The difference between the target equivalent clearance angle and the current equivalent clearance occupancy is then calculated to obtain the remaining clearance angle to be compensated. Next, the remaining clearance angle to be compensated is divided by a preset number of transition allocation cycles to obtain the clearance compensation angle increment for each control cycle. Subsequently, in multiple consecutive control cycles, the clearance compensation angle increment is added to the motor reference angle cycle by cycle, thereby forming a compensated motor reference angle trajectory that gradually fills the return clearance.

[0102] In another specific implementation, the difference between the target equivalent clearance angle corresponding to the current direction of joint movement and the current equivalent clearance occupancy can be divided into segments according to a preset number of cycles to obtain the clearance compensation angle allocation for each control cycle. Then, during the clearance transition state, the clearance compensation angle allocation is superimposed on the motor reference angle to smoothly compensate for the return clearance over multiple control cycles. Finally, after the return hysteresis state machine switches from the clearance transition state to the forward engagement state or the reverse engagement state, the clearance compensation angle allocation is reduced to zero, thereby realizing the multi-cycle progressive application of return clearance compensation on the time axis. This allows the compensated motor reference angle to gradually approach the compensation completion value with small, equal, or settable increments during the clearance compensation phase, thereby reducing the change amplitude of the motor reference angle within a single control cycle, reducing the instantaneous changes in the position control error signal and the servo output torque spikes, effectively suppressing the mechanical shock and vibration of the joint during the direction switching phase, and ensuring the smoothness and tracking accuracy of the end trajectory. This effectively solves the technical problems of large impact, strong vibration, and uneven trajectory caused by the one-time clearance compensation method.

[0103] It is worth noting that in the two specific implementation methods mentioned above, the first method obtains a fixed gap compensation angle increment in each control cycle by directly dividing the remaining gap angle to be compensated by the preset number of transition allocation cycles. This increment is then added in equal amounts over multiple subsequent control cycles to form a linear and uniform gap compensation rhythm. This method is simple to implement and has low computational overhead, making it easy to deploy quickly in servo control systems with high real-time requirements and fixed control cycles. However, its compensation angle increment is a constant value in the time dimension, and it cannot adaptively adjust the compensation amount for each cycle according to the dynamic response characteristics of the system, load changes, or operating modes. Therefore, under certain operating conditions with low stiffness or large load inertia changes, it may be difficult to achieve an optimal trade-off between compensation smoothness and compensation completion time in a timely manner.

[0104] In contrast, another implementation method divides the difference between the target equivalent clearance angle corresponding to the current direction of joint movement and the current equivalent clearance occupancy into segments according to a preset number of cycles to obtain the clearance compensation angle allocation for each control cycle. This segmented allocation process allows for flexible adjustment of the compensation angle allocation for different control cycles based on the state evolution of the return hysteresis state machine, the actual dynamic response characteristics of the joint, and the stability requirements of the control system. This better suppresses vibration and impact during the compensation process while ensuring multi-cycle progressive compensation, improving the system's robustness and trajectory tracking quality under complex operating conditions. Furthermore, in this implementation method, when the return hysteresis state machine detects a switch from the clearance transition state to the forward engagement state or the reverse engagement state, it can reduce the clearance compensation angle allocation to zero to achieve tight coupling between the compensation logic and the state machine switching, avoiding redundant or under-compensation issues caused by state switching.

[0105] Therefore, the first implementation method is suitable for application scenarios where system parameters are relatively fixed and operating conditions do not change significantly. The second implementation method, however, is more suitable for robot-driven positioning error compensation systems that require high-precision trajectory tracking and adaptation to complex operating conditions.

[0106] In this embodiment, the desired output angle of each joint of the robot, as well as the motor angle feedback information and joint running status information of each joint drive unit, are obtained. Then, based on the motor angle feedback information and joint running status information, the return hysteresis state machine corresponding to each joint is updated to obtain the return status information of each joint. Then, based on the desired output angle and return status information, the compensated motor reference angle of each joint is determined. In this way, the positioning error caused by the return backlash and hysteresis nonlinearity in the joint transmission chain is compensated by the compensated motor reference angle, so as to effectively reduce the robot drive positioning error caused by the return backlash and hysteresis nonlinearity, thereby improving the positioning accuracy, repeatability, and trajectory continuity of each joint of the robot and the end effector.

[0107] It is worth noting that, in the above embodiments, on the one hand, by utilizing the change in motor angle feedback and its sign within each control cycle, a return hysteresis state machine corresponding one-to-one with each joint is constructed and maintained. The meshing state of each joint is determined as three discrete states: forward meshing state, reverse meshing state, and gap transition state. Furthermore, the equivalent gap quantity, an internal continuous state variable, is used in the state machine to characterize the current degree of occupancy of the return gap. This allows for explicit modeling and real-time tracking of direction-dependent and path-dependent return gap and hysteresis behavior at the control algorithm level.

[0108] On the other hand, the desired output angle output by the kinematics solution module is regarded as the ideal target of the reducer output shaft. Based on the meshing state and equivalent clearance in the current return state information, positive hysteresis offset parameters and reverse hysteresis offset parameters that match the transmission mechanism type and load characteristics are introduced in the positive meshing state and the reverse meshing state, respectively. By superimposing the corresponding hysteresis offset onto the desired output angle, the compensated reference angle on the motor side is obtained, thereby realizing quantitative compensation for hysteresis offset.

[0109] After detecting a change in the motor's direction of motion and the return hysteresis state transitioning into a gap transition state, the gap compensation angle that has not yet been filled is segmented and allocated over several control cycles based on the target gap angle corresponding to the current direction of motion, the current equivalent gap occupancy, and the preset number of transition allocation cycles. This results in a cycle-by-cycle increase in the motor's reference angle after compensation, allowing the motor to move a small angle ahead in subsequent control cycles to smoothly fill the return gap.

[0110] This enables real-time modeling and feedforward compensation of backlash and hysteresis nonlinearity in the joint transmission chain, making the actual angular response of the reducer output shaft closer to the ideal output under a backlash-free, linear rigid transmission model. This effectively reduces robot drive positioning errors caused by backlash and hysteresis nonlinearity.

[0111] Figure 3 This is a schematic flowchart illustrating a motion object-driven positioning error compensation method according to another example embodiment of this application. For example... Figure 3 As shown, the method provided in this embodiment includes:

[0112] S210, Obtain the desired output angle of each joint of the robot.

[0113] In this step, the desired output angles of each joint of the robot are obtained. The desired output angles are the joint output axis reference angles calculated by the kinematics solving module based on the target end pose or target trajectory.

[0114] S220: Obtain motor angle feedback information and joint operating status information for each joint drive unit.

[0115] In this step, the motor angle feedback information and joint operating status information of each joint drive unit are obtained. The joint operating status information includes at least the direction of motor angle change within adjacent control cycles.

[0116] S230. Based on the motor angle feedback information and joint running status information, update the return hysteresis state machine corresponding to each joint to obtain the return status information of each joint.

[0117] In this step, based on the motor angle feedback information and joint operating status information, the return hysteresis state machine corresponding to each joint is updated to obtain the return status information of each joint. The return status information includes at least the meshing state and the equivalent clearance.

[0118] S240. Based on the desired output angle and return status information, determine the compensated motor reference angle for each joint.

[0119] In this step, based on the desired output angle and return status information, the compensated motor reference angle or the increment of the compensated motor reference angle for each joint is determined. The compensated motor reference angle is used to compensate for the positioning error caused by the return backlash and hysteresis nonlinearity in the joint drive train.

[0120] S250: Based on the compensated motor reference angle and motor angle feedback information, determine the position control error signal of each joint, and control the movement of each joint drive unit of the robot according to the position control error signal.

[0121] In this step, the position control error signal of each joint can be determined based on the compensated motor reference angle and motor angle feedback information, and the movement of each joint drive unit of the robot can be controlled according to the position control error signal to achieve robot drive positioning error compensation.

[0122] Specifically, the position error signal can be determined as the difference between the compensated motor reference angle and the motor angle feedback information. Then, based on the position error signal and combined with preset position loop control parameters, velocity loop control parameters, and current loop control parameters, a motor drive control quantity is generated. Finally, the motor drive control quantity is output to the servo driver of each joint drive unit so that the robot joints track the compensated motor reference angle.

[0123] In the position loop, the target motor speed command or intermediate speed reference for each joint is calculated based on the position error signal and the preset position loop proportional gain and integral parameters.

[0124] In the speed loop, the target electromagnetic torque command or intermediate torque reference value for each joint is calculated based on the difference between the target motor speed command and the actual motor speed calculated from the motor angle feedback information, as well as the preset speed loop proportional gain and integral parameters.

[0125] In the current loop, the motor phase current command or voltage command of each joint drive unit is calculated based on the difference between the target electromagnetic torque command and the current motor current feedback information, as well as the preset current loop proportional gain and integral parameters, and is used as the motor drive control output.

[0126] The motor drive control quantities, calculated step-by-step from the position loop, speed loop, and current loop, are sent to the corresponding servo drivers via fieldbus or real-time communication bus. The servo drivers, based on the received motor drive control quantities, control the inverter's PWM modulation mode and output duty cycle to drive the motor and generate the corresponding electromagnetic torque and speed. Then, during the above driving process, combined with real-time acquired motor angle and current feedback information, the motor's motion state is adjusted in a closed loop, causing the actual motor angle of each joint to gradually approach and track the compensated motor reference angle. This achieves robot drive positioning error compensation while compensating for backlash and hysteresis nonlinearity.

[0127] Figure 4 This is a schematic diagram of the structure of a moving object control system according to an example embodiment of this application. For example... Figure 4 As shown, the motion object control system 300 provided in this embodiment includes:

[0128] Angle acquisition module 310 is used to acquire the expected output angles of each execution structure of the moving object;

[0129] The status acquisition module 320 is used to acquire the motor angle feedback information and the operating status information of each execution structure drive unit;

[0130] Information processing module 330 is used to update the return hysteresis state machine corresponding to each execution structure based on the motor angle feedback information and the execution structure operation status information, so as to obtain the return status information of each execution structure;

[0131] The motor control module 340 is used to determine the compensated motor reference angle of each actuator based on the desired output angle and the return state information.

[0132] Figure 5 This is a schematic diagram of the structure of an electronic device according to an example embodiment of this application. For example... Figure 5 As shown, the electronic device 400 provided in this embodiment includes: a processor 401 and a memory 402; wherein:

[0133] Memory 402 is used to store computer programs, and the memory may also be flash memory.

[0134] Processor 401 is used to execute the execution instructions stored in the memory to implement the various steps in the above method. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0135] Alternatively, the memory 402 can be either standalone or integrated with the processor 401.

[0136] When the memory 402 is a device independent of the processor 401, the electronic device 400 may further include:

[0137] Bus 403 is used to connect the memory 402 and the processor 401.

[0138] This embodiment also provides a readable storage medium storing a computer program, which, when executed by at least one processor of an electronic device, enables the electronic device to perform the methods provided in the various embodiments described above.

[0139] This embodiment also provides a program product including a computer program stored in a readable storage medium. At least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the methods provided in the various embodiments described above.

[0140] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0141] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for compensating for positioning errors driven by a moving object, characterized in that, A method applied to a motion object control system comprising multiple actuator drive units; the method includes: Obtain the expected output angle of each execution structure of the moving object; Obtain the motor angle feedback information and execution structure operating status information of each execution structure drive unit; Based on the motor angle feedback information and the operating status information of the execution structure, the return hysteresis state machine corresponding to each execution structure is updated to obtain the return status information of each execution structure. Based on the desired output angle and the return state information, the compensated motor reference angle of each execution structure is determined.

2. The method for compensating for positioning errors driven by a moving object according to claim 1, characterized in that, After determining the compensated motor reference angles for each actuator, the process further includes: Based on the compensated motor reference angle and the motor angle feedback information, the position control error signal of each actuator is determined, and the driving unit of each actuator of the moving object is controlled to move according to the position control error signal.

3. The method for compensating for positioning errors driven by a moving object according to claim 2, characterized in that, The step of determining the position control error signal of each actuator based on the compensated motor reference angle and the motor angle feedback information, and controlling the action of each actuator drive unit of the moving object according to the position control error signal, includes: The difference between the compensated motor reference angle and the motor angle feedback information is determined to be the position error signal. Based on the position error signal, and combined with preset position loop control parameters, speed loop control parameters and current loop control parameters, a motor drive control quantity is generated; The motor drive control quantity is output to the servo driver of each execution structure drive unit so that the moving object execution structure tracks the compensated motor reference angle.

4. The method for compensating for positioning errors driven by a moving object according to claim 1, characterized in that, The step of updating the return hysteresis state machine corresponding to each execution structure based on the motor angle feedback information and the execution structure operating state information to obtain the return state information of each execution structure includes: Maintain a return hysteresis state machine for each execution structure; Based on the sign of the motor angle change in the current control cycle, the sign of the motor angle change in the previous control cycle, and the preset equivalent gap parameter, determine whether the current execution structure has changed its motion direction. When a change in motion direction is detected, the return hysteresis state machine is switched from the forward engagement state or the reverse engagement state to the gap transition state.

5. The method for compensating for positioning errors driven by a moving object according to claim 4, characterized in that, After switching the return hysteresis state machine from the forward engagement state or the reverse engagement state to the gap transition state, the method further includes: The equivalent clearance occupancy is updated based on the change in motor angle, and when the equivalent clearance occupancy reaches the preset forward clearance threshold or reverse clearance threshold, the return hysteresis state machine is switched from the clearance transition state to the forward engagement state or the reverse engagement state.

6. The method for compensating for positioning errors driven by a moving object according to claim 4, characterized in that, The update of the return hysteresis state machine corresponding to each execution structure also includes: A virtual output angle is maintained for each joint. The virtual output angle is the reducer output shaft angle determined based on motor angle feedback information, equivalent clearance occupancy, and transmission chain elastic torsion parameters. In the positive engagement state, the virtual output angle is updated based on the motor angle feedback information and the preset positive hysteresis offset parameter; In the reverse engagement state, the virtual output angle is updated based on the motor angle feedback information and the preset reverse hysteresis offset parameter; During the transition state of the gap, the virtual output angle is updated according to the change in motor angle and the elastic torsional stiffness parameter of the transmission chain.

7. The method for compensating for positioning errors driven by a moving object according to claim 1, characterized in that, The step of determining the compensated motor reference angle for each execution structure based on the desired output angle and the return state information includes: For the actuator in the positive engagement state, the corresponding motor reference angle is determined according to the desired output angle and the positive hysteresis offset parameter; For the actuator in the reverse engagement state, the corresponding motor reference angle is determined according to the desired output angle and the reverse hysteresis offset parameter. For the execution structure in the gap transition state, the gap compensation increment in multiple control cycles is determined based on the preset gap angle, the current equivalent gap occupancy, and the preset number of transition allocation cycles, so as to generate the compensated motor reference angle.

8. The method for compensating for positioning errors driven by a moving object according to claim 7, characterized in that, For the execution structure in a gap transition state, based on the preset gap angle, the current equivalent gap occupancy, and the preset number of transition allocation cycles, the gap compensation increment within multiple control cycles is determined to generate the compensated motor reference angle, including: The difference between the target equivalent gap angle corresponding to the current direction of motion of the executing structure and the current equivalent gap occupancy is divided into segments according to the preset number of cycles to obtain the gap compensation angle allocation amount for each control cycle. During the gap transition state, the gap compensation angle allocation is superimposed on the motor reference angle to smoothly fill the return gap over multiple control cycles. After the return hysteresis state machine switches from the gap transition state to the forward engagement state or the reverse engagement state, the gap compensation angle allocation is reduced to zero.

9. The method for compensating for positioning errors driven by a moving object according to claim 1, characterized in that, The plurality of actuator drive units include actuator drive units that employ at least one of the following transmission mechanisms: gear reducer, harmonic reducer, or planetary reducer.

10. A control system for a moving object, characterized in that, include: The angle acquisition module is used to obtain the expected output angle of each execution structure of the moving object; The status acquisition module is used to acquire the motor angle feedback information and the operating status information of each execution structure drive unit; The information processing module is used to update the return hysteresis state machine corresponding to each execution structure based on the motor angle feedback information and the execution structure operation status information, so as to obtain the return status information of each execution structure. The motor control module is used to determine the compensated motor reference angle for each actuator based on the desired output angle and the return state information.

Images