A Risk Warning Method and System for Six-Direction Crawling Racks Based on Multi-Path Distributed Drive

By utilizing the motion of active drive units to generate mechanical excitation in a six-way crawling rack with multi-channel distributed drive, and combining induced voltage and rotor micro-motion state detection, the problem of difficulty in detecting faults in inactive drive units is solved, and accurate risk warning of standby drive units is achieved.

CN123105138APending Publication Date: 2026-11-17JIANGSU BAISHUN INTELLIGENT LOGISTICS EQUIP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611635602.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-22
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the existing technology, the non-activated drive units in the six-way crawling rack with multi-path distributed drive lack continuous motion excitation, making it difficult to detect faults such as mechanical drag, insufficient brake release or insufficient locking in a timely manner, which affects the safe operation of the robot.

Method used

By using the movement of the active drive unit to generate mechanical excitation for the standby drive unit, and combining the induced voltage response and rotor micro-motion state detection, abnormal mechanical response of the standby drive unit is identified. Interference is eliminated by using the PWM switching state of the active shaft, thereby improving detection accuracy.

Benefits of technology

It enables online status monitoring of non-current motion drive units, avoiding omissions in detection that only target the current active drive unit, and improving the accuracy and timeliness of abnormal identification of standby drive units.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123105138A_ABST
    Figure CN123105138A_ABST
Patent Text Reader

Abstract

This invention discloses a risk warning method and system for a six-way crawling rack based on multi-path distributed drive, which relates to the technical field of intelligent warehouse racks and their motion equipment. By utilizing the mechanical excitation formed by the motion of the active drive unit to the standby drive unit, and combining the rotor micro-motion and induced voltage response of the standby drive unit for state detection, it solves the problem that faults such as mechanical drag, insufficient brake release, or insufficient locking caused by the lack of continuous motion excitation of the inactive drive unit are difficult to detect in a timely manner.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of intelligent warehouse automated storage and retrieval systems and their motion equipment. More specifically, this invention relates to a risk warning method and system for a six-way crawling rack based on multi-path distributed drive. Background Technology

[0002] Six-directional crawling robots are used in automated storage and retrieval systems (AS / RS) and can move in six directions (up, down, left, right, forward, and backward) on the outer track of the rack. To adapt to the multi-directional movement requirements, these robots adopt a distributed drive architecture, meaning that the crawling robot's drive system contains multiple physically independent drive units, each corresponding to one degree of freedom. The controller selectively activates the corresponding drive unit according to the current movement mode, while inactive drive units are in standby or locked states.

[0003] In existing technologies, drive status monitoring typically focuses on the currently active drive unit, determining its operational status by collecting signals such as current, rotational speed, and temperature. For inactive drive units, lacking continuous excitation during motion because they are not involved in driving the current direction, faults such as mechanical drag, insufficient brake release, or inadequate locking are difficult to detect from the current motion response. These faults accumulate during standby and may only be discovered when the drive unit is activated and participates in motion. By this time, the fault has developed to the point of affecting robot operational safety, potentially leading to track switching failures or robot instability.

[0004] To address the above problems, this invention proposes a solution. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a risk warning method and system for a six-way crawling rack based on multi-path distributed drive. By utilizing the mechanical excitation generated by the motion of the active drive unit to the standby drive unit, and combining the rotor micro-motion and induced voltage response of the standby drive unit for state detection, the method solves the problem that faults such as mechanical drag, insufficient brake release, or insufficient locking caused by the lack of continuous motion excitation of the inactive drive unit are difficult to detect in a timely manner.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A risk warning method for a six-way crawling rack based on multi-path distributed drive includes the following steps: determining the active shaft and the standby shaft; acquiring the induced voltage of the standby shaft winding during the movement of the active shaft; synchronously acquiring the PWM switching state of the active shaft, determining the switching interference reference and low-interference sampling window; using the PWM switching state of the active shaft to eliminate related switching interference in the induced voltage of the standby shaft, and extracting the induced voltage response of the standby shaft in the low-interference sampling window; synchronously acquiring the micro-motion angular velocity of the standby shaft rotor; determining the expected induced voltage response range based on the correspondence between the micro-motion angular velocity of the standby shaft and the normal induced voltage response; comparing the actual induced voltage response with the expected induced voltage response, and identifying mechanical response anomalies of the standby shaft based on the deviation between the two.

[0007] The risk warning system for a six-way crawling rack based on multi-channel distributed drive includes a shaft system determination module, a sensing and acquisition module, an interference synchronization module, an interference suppression module, a micro-motion acquisition module, an expectation determination module, and an anomaly identification module. The shaft system determination module identifies the active shaft and the standby shaft. The sensing and acquisition module acquires the induced voltage of the standby shaft winding during the active shaft's movement. The interference synchronization module synchronously acquires the PWM switching state of the active shaft, determines the switching interference reference, and establishes a low-interference sampling window. The interference suppression module uses the active shaft PWM switching state to eliminate related switching interference in the standby shaft's induced voltage and extracts the standby shaft's induced voltage response within the low-interference sampling window. The micro-motion acquisition module synchronously acquires the micro-motion angular velocity of the standby shaft rotor. The expectation determination module determines the expected induced voltage response range based on the correspondence between the standby shaft's micro-motion angular velocity and the normal induced voltage response. The anomaly identification module compares the actual induced voltage response with the expected induced voltage response and identifies mechanical response anomalies in the standby shaft based on their deviation.

[0008] The technical effects and advantages of the risk warning method and system for a six-way crawling rack based on multi-path distributed drive of the present invention are as follows: 1. This invention utilizes the mechanical motion transmitted from the active drive unit to the standby drive unit via the robot body during the active drive unit's movement to generate micro-motions in the rotor of the standby drive unit and form an induced voltage. By identifying mechanical response anomalies through the correspondence between the micro-motion state of the standby drive unit and the induced voltage response, online status monitoring of non-current motion drive units can be achieved, avoiding the omission of potential mechanical anomalies in the standby drive unit due to only detecting the current active drive unit.

[0009] 2. This invention synchronously acquires the PWM switching state of the active drive unit and uses a switching interference reference sequence to cancel out related interference in the induced voltage of the standby drive unit. At the same time, it extracts the response in a low-interference sampling window after the switching action, thereby improving the recognition accuracy of the weak induced voltage response of the standby drive unit and reducing the interference of the active drive unit's switching action on the standby drive unit's state detection. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the risk warning method for a six-way crawling rack based on multi-path distributed drive according to the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0012] This invention targets a six-directional crawling robot with a multi-path distributed drive structure. When the robot moves in one direction, only the drive unit corresponding to the current direction of movement is actively activated, while other drive units remain in a non-driven state. Since there are still mechanical connections between each drive unit and the robot body and the shelf track, the vibrations and mechanical movements generated during the movement of the active drive units are transmitted through the robot body to the non-driven drive units, causing the rotor in the non-driven drive units to undergo follow-up micro-motions and forming an induced voltage corresponding to these rotor micro-motions in its windings. Therefore, by utilizing the mechanical excitation generated by the standby drive units during the movement of the active drive units, and by collecting the induced voltage of the standby drive units and simultaneously acquiring their rotor micro-motion states, a correspondence between rotor micro-motions and induced voltage responses can be established. Furthermore, based on the deviation between the actual induced voltage response and the normal correspondence, abnormal mechanical responses of the standby drive units can be identified, enabling monitoring and risk warning of potential mechanical abnormalities in non-currently moving drive units.

[0013] Example 1, Figure 1 The present invention provides a risk warning method for a six-directional crawling rack based on multi-path distributed drive, comprising the following steps: S1: When the crawling robot moves along the current direction of motion, the drive unit corresponding to the current direction of motion is determined as the active axis, and the drive unit that is in a non-drive state and mechanically coupled to the robot body outside the current direction of motion is determined as the standby axis.

[0014] In this embodiment, when the crawling robot moves along the current direction of motion, the drive unit corresponding to the current direction of motion is determined as the active axis, and the drive units outside the current direction of motion that are in a non-driven state and mechanically coupled to the robot body are determined as standby axes. Specifically: Receive motion control commands from the crawling robot and parse the motion axis identifiers from the motion control commands.

[0015] According to the preset correspondence table between drive units and motion axes, the drive unit corresponding to the motion axis identifier is determined as the active axis.

[0016] Obtain the drive enable command, motor current sampling value, brake control command status, and clutch control command status of the other drive units besides the drive shaft from the drive status register.

[0017] For each of the remaining drive units, if its drive enable command is low and its motor current sample value is zero, then the drive unit is determined to be in a non-drive state.

[0018] For each drive unit in a non-drive state, if its clutch control command state is engaged and its brake control command state is released, it is determined that the drive unit maintains mechanical coupling with the robot body at the control level.

[0019] The drive unit that simultaneously satisfies the conditions of being in a non-driven state and maintaining mechanical coupling with the robot body at the control level is defined as the standby axis, and the active axis identifier and the standby axis identifier list are output.

[0020] In this embodiment, the motion axis identifier is the encoded value used in the motion control command to indicate the direction of target motion. This encoded value is generated by the upper-level scheduling system according to the task path and sent to the crawling robot controller. The controller parses the encoded value from the data frame of the motion control command. The correspondence between the encoded value and the motion direction is specified by the motion control protocol of the crawling robot. For example, the encoded value 01 corresponds to the forward and backward direction, the encoded value 02 corresponds to the left and right direction, and the encoded value 03 corresponds to the up and down direction.

[0021] In this embodiment, the correspondence table between drive units and motion axes is a mapping table pre-stored in the crawling robot controller. This mapping table is generated before the crawling robot leaves the factory based on the mechanical structure calibration of its drive system. Each row of the mapping table records the identifier of a drive unit and its corresponding motion axis identifier. When the controller parses the motion axis identifier, it uses the motion axis identifier as an index to query the mapping table and determines the queried drive unit identifier as the active axis identifier.

[0022] In this embodiment, the drive enable instruction is a control signal issued by the controller to the driver. This control signal is used to control whether the driver outputs drive current. When the drive enable instruction is high, the driver is allowed to output drive current. When the drive enable instruction is low, the driver is prohibited from outputting drive current. The controller reads the current level state of the drive enable instruction from the driver status register.

[0023] In this embodiment, the motor current sampling value is a digital quantity obtained by the current sampling circuit inside the driver after sampling the motor winding current. This digital quantity is stored in the driver status register. The controller reads this digital quantity from the driver status register. When the digital quantity is zero, it indicates that the driver is not outputting drive current to the motor winding.

[0024] In this embodiment, the brake control command status is the current status of the control command issued by the controller to the brake coil of the standby shaft motor. This control command status is stored in the driver status register. The controller reads the control command status from the driver status register. When the control command status is "release", it means that the controller has issued a release command to the brake coil. When the control command status is "lock", it means that the controller has issued a lock command to the brake coil.

[0025] In this embodiment, the clutch control command status is the current status of the control command issued by the controller to the standby shaft clutch actuator. This control command status is stored in the driver status register. The controller reads the control command status from the driver status register. When the control command status is engaged, it means that the controller has issued an engagement command to the clutch. When the control command status is disengaged, it means that the controller has issued a disengagement command to the clutch.

[0026] In this embodiment, the non-driving state is the working state of the drive unit when the drive enable command of the drive unit is low and the motor current sampling value is zero. In this state, the drive unit does not output drive current to the motor windings and the motor does not generate active torque.

[0027] In this embodiment, the mechanical coupling between the drive unit and the robot body at the control level is defined as the drive unit's clutch control command state being engaged and the brake control command state being released. Under this connection, the controller has issued control commands to the clutch and brake to mechanically connect the drive unit to the robot body. However, this connection only represents the state at the control command level and does not represent the actual physical state of the clutch and brake.

[0028] In this embodiment, the standby axis identifier list is a list generated by the controller by arranging the identifiers of drive units that simultaneously meet the conditions of being in a non-drive state and maintaining mechanical coupling with the robot body at the control level in a preset order. This list is stored in the controller's memory for subsequent steps to read each standby axis identifier in sequence.

[0029] It should be noted that when a crawling robot contains more than three drive units, there may be multiple drive units that meet the non-drive state and maintain mechanical coupling with the robot body at the control level among the drive units other than the active axis. In this case, the standby axis identifier list contains multiple standby axis identifiers. Subsequent steps will perform induced voltage acquisition and induced voltage response extraction on each standby axis identifier in the list.

[0030] It should be noted that when a motion axis corresponds to multiple drive units in the mapping table between drive units and motion axes, the controller will identify all the drive units corresponding to the motion axis identifier as active axes. The number of active axes is determined according to the number of drive units corresponding to the motion axis identifier in the mapping table.

[0031] In a specific embodiment, the above steps determine the standby axis identifier by reading the status signal at the control command level from the driver status register. The motivation is to provide a detection target for subsequent back EMF detection, so that subsequent steps can collect the induced voltage of the standby axis during the movement of the active axis, thereby completing the screening of the standby axis using the existing control signal without adding additional sensors.

[0032] S2: The standby shaft motor winding is put into a non-drive voltage detection state. During the movement of the drive shaft, the induced voltage generated by the standby shaft motor winding in this non-drive voltage detection state is collected.

[0033] In this embodiment, the step of putting the standby shaft motor winding into a non-drive voltage detection state, and collecting the induced voltage generated by the standby shaft motor winding in this non-drive voltage detection state during the movement of the drive shaft, specifically involves: According to the standby axis identifier list, a standby axis identifier is read sequentially from the standby axis identifier list.

[0034] According to the standby axis identifier, a winding floating command is sent to the corresponding driver. The winding floating command is used to control the driver to block the gate drive signals of all switching transistors of its inverter, so that an open circuit is formed between the three-phase windings of the standby axis motor and the DC bus of the driver.

[0035] The gate block confirmation signal is read from the status register of the driver. If the gate block confirmation signal is valid, it is determined that the standby shaft motor winding has entered the non-drive voltage detection state.

[0036] During the movement of the active axis, the movement of the active axis is transmitted to the standby axis through the robot body, causing the rotor of the standby axis to make micro-motions. The micro-motions of the standby axis rotor induce voltage in its motor windings.

[0037] The voltage signal between any two phase winding terminals of the standby shaft motor is acquired by a voltage acquisition circuit at a set sampling frequency. The two input terminals of the voltage acquisition circuit are respectively connected to the terminals of the two phase windings of the standby shaft motor. The voltage signal is the induced voltage generated by the micro-motion of the standby shaft rotor.

[0038] The collected voltage signals are stored in the data buffer corresponding to the standby axis identifier in the order of sampling time, forming the induced voltage sequence of the standby axis.

[0039] Return to the standby axis identifier list, read the next standby axis identifier, and repeat the above steps until all standby axis identifiers in the standby axis identifier list have completed the acquisition of the induced voltage sequence.

[0040] The following is a feasible arithmetic expression for determining the sampling frequency: ; ; In the formula, This is the highest frequency of the standby axis induced voltage. This represents the number of pole pairs of the standby axis motor. This is the upper limit of the preset micro-motion angular velocity range. Set a multiplier, with a value ranging from 2 to 10. The sampling frequency.

[0041] It should be noted that, This is the upper limit of the mechanical angular velocity of the standby axis rotor, obtained by taking the absolute value of the derivative of the rotor angular displacement signal output by the standby axis encoder with respect to time, and then taking the maximum value; p is read from the nameplate parameters or factory calibration data of the standby axis motor; the value of k is determined according to the signal-to-noise ratio of the standby axis induced voltage, ranging from 2 to 10. The following are examples of feasible rules for setting the multiple: Obtain the signal-to-noise ratio of the standby axis induced voltage, compare the signal-to-noise ratio with the preset lower limit and the preset upper limit of the signal-to-noise ratio, when the signal-to-noise ratio is low... When the signal-to-noise ratio (SNR) is at its lower limit, the multiplier is set to 10. When the SNR is higher than its upper limit, the multiplier is set to 2. When the SNR is between the lower and upper limits, the multiplier is equal to 10 minus the product of 8 and a first ratio. The first ratio is the ratio of the difference between the SNR and the lower limit to the difference between the upper and lower limits. The lower limit ranges from 0 dB to 5 dB, and the upper limit ranges from 20 dB to 30 dB.

[0042] In this embodiment, the standby axis identifier list is a list generated by arranging the drive unit identifiers output in the previous step that simultaneously meet the requirements of being in a non-driven state and maintaining mechanical coupling with the robot body at the control level in a preset order. This list is stored in the memory of the crawling robot controller, and each standby axis identifier corresponds to one drive unit.

[0043] In this embodiment, the winding floating command is a control command issued by the controller to the standby axis driver. This control command is used to control the standby axis driver to block the gate drive signals of all switching transistors in its inverter, thereby creating an open circuit between the three-phase windings of the standby axis motor and the DC bus of the driver.

[0044] In this embodiment, the gate drive signal is a voltage signal output from the internal controller of the standby axis driver to the gate of the inverter switching transistor. This voltage signal is used to control the switching transistor's on and off states. When the gate drive signal is blocked, all the inverter switching transistors remain off, and no current path is formed between the standby axis motor's three-phase windings and the DC bus.

[0045] In this embodiment, the gate block confirmation signal is a status signal output by the status register of the standby axis driver after the gate drive signal is blocked. When the status signal is valid, it indicates that all switches of the inverter have been turned off. When the status signal is invalid, it indicates that at least one switch of the inverter has not been turned off.

[0046] In this embodiment, the non-drive voltage detection state is the operating state of the standby shaft motor winding when an open circuit is formed between the standby shaft motor three-phase winding and the DC bus and all the inverter switches are in the off state. In this state, the standby shaft motor winding does not receive drive current from the DC bus, and the voltage on the winding terminals is determined only by the induced voltage generated by the angular motion of the standby shaft rotor.

[0047] In this embodiment, the voltage acquisition circuit is a differential voltage measurement circuit connected between the terminals of any two phase windings of the standby shaft motor. The two input terminals of the circuit are respectively connected to the terminals of the two phase windings, and the output terminal of the circuit is connected to the input terminal of the analog-to-digital converter. The analog-to-digital converter converts the differential voltage into a digital quantity and stores it in the data buffer.

[0048] In this embodiment, the sampling frequency is the number of times per second that the voltage acquisition circuit collects the voltage between the terminals of the two-phase windings of the standby shaft motor during the movement of the drive shaft. The sampling frequency is determined by the following steps: obtaining the preset micro-motion angular velocity range of the standby shaft rotor, obtaining the number of pole pairs of the standby shaft motor, multiplying the upper limit of the preset micro-motion angular velocity range by the number of pole pairs to obtain the highest frequency of the standby shaft induced voltage, and setting the sampling frequency to a set multiple of the highest frequency, wherein the value of the set multiple ranges from 2 to 10.

[0049] In this embodiment, the induced voltage sequence is a data sequence formed by arranging the digital quantities of the voltage between the terminals of the two-phase windings of the standby axis motor, which are collected by the voltage acquisition circuit in the order of sampling time during the movement of the active axis, in chronological order. This data sequence is stored in the data buffer corresponding to the standby axis identifier.

[0050] In this embodiment, the micro-motion is the angular motion generated by the mechanical disturbance transmitted by the robot body during the movement of the standby axis rotor on the active axis. The angular displacement amplitude of the angular motion is within a preset micro-motion range, which is determined by the mechanical coupling gap between the standby axis and the robot body.

[0051] In this embodiment, mechanical coupling is a mechanical relationship in which the drive wheel or transmission mechanism of the standby axis is in contact or connected with the robot body. Under this mechanical relationship, the track joint impact and frame vibration generated by the movement of the drive axis are transmitted to the drive wheel or transmission mechanism of the standby axis through the robot body, thereby driving the standby axis rotor to generate angular motion.

[0052] It should be noted that the prerequisite for the standby axis rotor to generate micro-motion is that the drive wheel or transmission mechanism of the standby axis maintains mechanical coupling with the robot body. When the active axis moves, the impact generated by the track joint and the frame vibration are transmitted to the drive wheel or transmission mechanism of the standby axis through the robot body, thereby causing the standby axis rotor to generate angular motion within the preset micro-motion range. If the brake of the standby axis is completely locked and the clutch is completely disengaged, the standby axis rotor will not generate angular motion, and the standby axis is not the detection object of this method.

[0053] It should be noted that when the gate block confirmation signal is invalid, the controller resends the winding floating command to the standby axis driver. If the gate block confirmation signal is still invalid after the number of consecutive retry attempts reaches the preset number, the standby axis identifier is removed from the standby axis identifier list and the detection result of the gate block failure of the standby axis driver is output. The preset number of retry attempts ranges from 2 to 5.

[0054] It should be noted that the voltage acquisition circuit acquires the voltage signal between any two phase winding terminals of the standby shaft motor. When the standby shaft motor is a three-phase motor, the voltage signal between any two phase winding terminals contains the induced voltage component generated by the micro-motion of the standby shaft rotor. Two phase winding terminals can be selected and connected to the voltage acquisition circuit.

[0055] In a specific embodiment, the above steps use a winding floating command to put the standby axis motor winding into a non-drive voltage detection state. The motivation for this is that the standby axis generates rotor micro-motion due to mechanical coupling during the movement of the drive axis. This rotor micro-motion generates an induced voltage in the standby axis motor winding. By collecting this induced voltage, the motion state information of the standby axis rotor can be obtained without applying a drive current to the standby axis motor. This provides a data source for subsequent steps to establish the correspondence between the standby axis micro-motion angular velocity and the induced voltage response.

[0056] S3: Synchronously acquire the PWM switching state of the drive shaft driver, determine the time period when the switching noise of the drive shaft driver is higher than a first threshold and the time period when the switching noise is lower than a second threshold based on the PWM switching state, determine the time period when the switching noise is lower than the second threshold as a low interference sampling window, and construct a switching interference reference sequence based on the PWM switching state.

[0057] In this embodiment, the synchronous acquisition of the PWM switching state of the drive shaft driver is performed. Based on the PWM switching state, time periods when the switching noise of the drive shaft driver is higher than a first threshold and time periods when the switching noise is lower than a second threshold are determined. The time periods when the switching noise is lower than the second threshold are defined as low-interference sampling windows. A switching interference reference sequence is constructed based on the PWM switching state, which has the following characteristics: Based on the induced voltage sequence stored in the data buffer corresponding to the standby axis identifier, the time stamp corresponding to each sample value in the induced voltage sequence is obtained.

[0058] During the operation of the drive shaft, the PWM switching state and switching edge time of each switching transistor of the drive shaft inverter are synchronously acquired through the PWM timer or switching edge capture circuit of the drive shaft driver, and the PWM switching state is time-aligned according to the time scale of the induced voltage sequence.

[0059] The PWM switching states of each switching transistor between each switching edge are arranged according to the time stamp to form a PWM state sequence. Each moment in the PWM state sequence corresponds to a switch state word composed of the PWM switching states of each switching transistor.

[0060] Based on the switching state changes of adjacent moments in the PWM state sequence, the switching edge at which the switching state changes is determined, and the moment corresponding to each switching edge is taken as the starting moment of the switching action; the time period after the switching edge until the next switching state change of the drive shaft inverter is determined as the corresponding stable period of the switching action.

[0061] For each stable period of switching action, the period from the moment after a preset stabilization time from the corresponding switch edge to the moment before the next switch state change is defined as the stable sampling interval, wherein the preset stabilization time is determined based on the transient interference attenuation time after the switch edge.

[0062] The stable sampling intervals corresponding to each switching action are arranged in time-scale order, and adjacent stable sampling intervals that do not have active shaft inverter switching actions are merged to form a low-interference sampling window.

[0063] Based on the switch state word corresponding to each moment in the PWM state sequence, the switch state word is converted into a corresponding switch interference reference value, wherein the switch interference reference value is used to characterize the switching operation state of the active shaft inverter at the corresponding moment, and the corresponding switch interference reference value changes synchronously when the switch state word changes.

[0064] The switching interference reference values ​​corresponding to each time point are arranged in the time-scale order of the induced voltage sequence to form a switching interference reference sequence.

[0065] The low-interference sampling window and the switching interference reference sequence are stored in the data buffer corresponding to the standby axis identifier.

[0066] In this embodiment, the time stamp is a time mark synchronously recorded by the internal clock of the controller when the voltage acquisition circuit acquires the voltage signal between the terminals of the two-phase windings of the standby shaft motor. Each sampling moment corresponds to a time stamp, and the resolution of the time stamp is determined by the internal clock frequency of the controller. The value range of the internal clock frequency of the controller is 10 MHz to 100 MHz.

[0067] In this embodiment, the PWM switching state is the level state of the gate drive signal of each switching transistor in the drive shaft driver inverter. This level state is output by the PWM timer or switching edge capture circuit of the drive shaft driver. A high level indicates that the switching transistor is in the on state, and a low level indicates that the switching transistor is in the off state.

[0068] In this embodiment, the switching edge moment is the moment when the PWM switching state of each switching transistor of the drive shaft inverter changes, including the rising edge moment when the PWM state changes from low level to high level and the falling edge moment when the PWM state changes from high level to low level. This moment is recorded by the PWM timer or switching edge capture circuit of the drive shaft driver.

[0069] In this embodiment, the switch status word is a binary word composed of the PWM switching status of each switch transistor of the drive shaft inverter at the same sampling time, arranged in the order of the switch transistor numbers. The number of bits in the switch status word is the total number of switches of the drive shaft inverter, and each bit corresponds to the PWM switching status of one switch transistor.

[0070] In this embodiment, the starting time of the switching action is the time corresponding to the change when the switching state changes between adjacent times in the PWM state sequence. This time is determined by the fact that the switching state words of two adjacent sampling times in the PWM state sequence are not simultaneously determined, and the later time of the two adjacent sampling times is taken as the starting time of the switching action.

[0071] In this embodiment, the stable switching period is the time from the edge of the switch to the next switch state change of the drive shaft inverter. The start time of this period is the time corresponding to the edge of the switch, and the end time is the time corresponding to the next switch state change of the drive shaft inverter.

[0072] In this embodiment, the preset stabilization time is the time required for the transient interference generated by the switching action of the drive shaft inverter to attenuate from the switching edge. The value of this preset stabilization time is determined through the following steps: obtaining the inductance value of the drive shaft motor winding, obtaining the resistance value of the drive shaft motor winding, multiplying the winding inductance value by a set coefficient and dividing by the winding resistance value to obtain the attenuation time constant, and multiplying the attenuation time constant by a set multiple to obtain the preset stabilization time. The set coefficient ranges from 1 to 3, and the set multiple ranges from 3 to 5. The following is a feasible arithmetic expression for determining the attenuation time constant: ; In the formula, The decay time constant; The setting coefficient is L; the inductance value of the drive shaft motor winding is read from the nameplate parameters or factory calibration data of the drive shaft motor; the resistance value of the drive shaft motor winding is read from the nameplate parameters or factory calibration data of the drive shaft motor.

[0073] It should be noted that the DC bus voltage value of the drive shaft driver is not included in the calculation of the decay time constant. This is because the decay process of the motor winding current during the turn-off period is determined by the winding inductance and winding resistance values. The DC bus voltage value does not directly affect the decay process of the winding current during the turn-off period. The DC bus voltage value only affects the initial value of the winding current at the start of the turn-off event and does not affect the decay time constant.

[0074] In this embodiment, the stable sampling interval is the period from the moment after a preset stabilization time from the corresponding switch edge to the moment before the next switch state change during the stable period of the switching action. During this period, the transient interference of the drive shaft inverter has been attenuated to a set level.

[0075] In this embodiment, the low-interference sampling window is a time period set formed by arranging the stable sampling intervals corresponding to each switching action in time-scale order and merging adjacent stable sampling intervals that do not have active shaft inverter switching actions.

[0076] In this embodiment, the switching interference reference value is a value that characterizes the switching operation state of the active shaft inverter at a corresponding moment. This value is obtained by converting the switch status word at that moment. When the switch status word changes, the corresponding switching interference reference value changes synchronously.

[0077] In this embodiment, the switching interference reference sequence is a data sequence formed by arranging the switching interference reference values ​​corresponding to each time point according to the time index order of the induced voltage sequence. Each switching interference reference value in the data sequence corresponds to a time index.

[0078] It should be noted that when the duration of the stable period of the switching action is less than the preset stable time, the stable sampling interval corresponding to the stable period of the switching action is empty, and the stable period of the switching action is not included in the low interference sampling window. The number of stable sampling intervals in the low interference sampling window is determined according to the actual timing of the active shaft PWM switching state.

[0079] It should be noted that the total number of switching transistors in the drive shaft inverter is determined by the topology of the drive shaft driver. When the drive shaft driver is a three-phase full-bridge topology, the total number of switching transistors is 6. When the drive shaft driver is another topology, the total number of switching transistors is determined according to the number of switching transistors in that topology.

[0080] In a specific embodiment, the above steps synchronize the active axis PWM switching state with the standby axis induced voltage acquisition in time. The motivation is to align the acquisition time of the standby axis induced voltage with the period when the switching noise of the active axis inverter is low, so as to provide a time window and reference sequence for subsequent adaptive filtering and discrete integration, and reduce the interference caused by the active axis switching action to the standby axis induced voltage.

[0081] S4: Using the switching interference reference sequence as a reference input, an adaptive filter is used to cancel the interference components related to the switching interference reference sequence in the induced voltage to obtain the induced voltage after interference removal. The induced voltage after interference removal is integrated and sampled within the low interference sampling window to obtain the standby axis induced voltage response.

[0082] In this embodiment, the step of using the switching interference reference sequence as a reference input, canceling the interference components related to the switching interference reference sequence in the induced voltage through an adaptive filter to obtain the interference-filtered induced voltage, and then integrating and sampling the interference-filtered induced voltage within the low-interference sampling window to obtain the standby axis induced voltage response, specifically: Based on the data buffer corresponding to the standby axis identifier, the induced voltage sequence, low-interference sampling window, and switching interference reference sequence are read from the data buffer.

[0083] According to the sampling time scale of the induced voltage sequence, the switching interference reference values ​​corresponding to the current time and several previous sampling times are extracted sequentially from the switching interference reference sequence to form the reference input vector for the current time.

[0084] The induced voltage sample value at the current moment is used as the desired input of the adaptive filter, and the reference input vector at the current moment is used as the reference input of the adaptive filter. The adaptive filter outputs the estimated value of the switching interference at the current moment based on the reference input vector.

[0085] The difference between the current induced voltage sample value and the estimated switching interference value is used to obtain the current induced voltage sample value after filtering out the switching interference.

[0086] The adaptive filter employs a minimum mean square error (MMS) filter, which updates the weight vector as follows: The current reference input vector and the current weight vector are multiplied to obtain the current switching interference estimate; the difference between the current induced voltage sample value and the MMS estimate is used to determine the current error signal; the error signal, the current reference input vector, and the step size factor are multiplied and added to the current weight vector to obtain the weight vector for the next time step.

[0087] The sampled values ​​of the induced voltage after filtering out the interference at each time point are arranged in time-scale order to form the induced voltage sequence after filtering out the interference.

[0088] The start and end times of the first stable sampling interval are read from the low-interference sampling window.

[0089] In the filtered-out induced voltage sequence, all filtered-out induced voltage sample values ​​with time stamps located within the first stable sampling interval are extracted.

[0090] Based on the sampling interval of the induced voltage sampled values ​​after filtering out interference, the absolute values ​​of each extracted induced voltage sampled value after filtering out interference are discretely integrated to obtain the standby axis induced voltage response sub-value corresponding to the first stable sampling interval.

[0091] The remaining stable sampling intervals in the low-interference sampling window are read sequentially, and the standby axis induced voltage response sub-values ​​corresponding to each stable sampling interval are determined in the same way.

[0092] Arrange the standby axis induced voltage response sub-values ​​corresponding to all stable sampling intervals in chronological order to form a standby axis induced voltage response sub-value sequence, and determine the standby axis induced voltage response sub-value sequence as the standby axis induced voltage response.

[0093] The standby axis sensed voltage response is stored in the data buffer corresponding to the standby axis identifier.

[0094] It should be noted that the step size factor is determined based on the correlation of the switching interference reference sequence. Specifically, the autocorrelation matrix of the reference input matrix corresponding to the switching interference reference sequence is obtained, and the largest eigenvalue of the autocorrelation matrix is ​​determined. The upper limit of the step size factor is determined based on the product of the reciprocal of the largest eigenvalue and a set coefficient. The step size factor not exceeding the upper limit is used for updating the weight vector of the minimum mean square error filter.

[0095] It should be noted that the initial value of the weight vector of the minimum mean square error filter is set to zero vector, and the length of the weight vector is determined according to the length of the reference input vector used to characterize the correlation of active axis switching interference.

[0096] The following are feasible arithmetic expressions for the adaptive filter output, weight vector update, induced voltage after interference filtering, and standby axis induced voltage response sub-value calculation: ; ; ; ; In the formula, n is the sequence number of the current sampling time, and its value is an integer greater than or equal to 1; This is the output value of the adaptive filter at the current sampling time, i.e., the estimated value of the interference components at the current sampling time; This is the weight vector at the current sampling time; The reference input vector at the current sampling time is composed of the reference values ​​at the current sampling time and the P previous sampling times in the switching interference reference sequence, where P is the length of the weight vector; the symbol · represents the vector dot product operation, i.e. and Multiply corresponding elements and then sum them; This is the desired input at the current sampling moment, i.e., the sampled value at the current sampling moment in the induced voltage sequence; The difference between the sampled value of the induced voltage at the current sampling time and the estimated value of the interference component is used as both the error signal in the weight vector update and the sampled value of the induced voltage after filtering out interference at the current sampling time. is the weight vector for the next sampling time; μ is the step size factor, ranging from 0.001 to 0.01; V(k) is the standby axis induced voltage response sub-value corresponding to the k-th stable sampling interval; k is the index of the stable sampling interval, taking the value of an integer greater than or equal to 1; This is the sequence number of the starting sampling time of the k-th stable sampling interval; Σ is the total number of sampling points in the (k-1)th stable sampling interval, which is obtained by multiplying the difference between the end time and the start time of the stable sampling interval by the sampling frequency and then rounding it down; Σ is the sum of the absolute values ​​of e(n) over all sampling times in the kth stable sampling interval.

[0097] The following is a feasible arithmetic expression for determining the upper limit of the step size factor: ; In the formula, is the step size factor; c is a set coefficient, ranging from 0.01 to 0.1; The maximum eigenvalue of the autocorrelation matrix of the switching interference reference sequence can be obtained by performing eigenvalue decomposition on the autocorrelation matrix using the Jacobi method and then taking the maximum eigenvalue.

[0098] It should be noted that the following are the feasible rules for setting the coefficient: Obtain the convergence time requirement of the switching interference component in the standby axis induced voltage sequence. The convergence time requirement is the maximum number of iterations allowed for the adaptive filter weight vector to converge from the initial value to the stable value. Multiply the maximum number of iterations by the minimum eigenvalue of the autocorrelation matrix of the switching interference reference sequence to obtain the convergence speed index. Divide the convergence speed index by the maximum eigenvalue of the autocorrelation matrix of the switching interference reference sequence to obtain the set coefficient. The value range of the set coefficient is 0.01 to 0.1.

[0099] In this embodiment, the adaptive filter is a digital filter that automatically adjusts its weight vector according to the statistical characteristics of the input signal. The filter takes the switching interference reference sequence as the reference input and the induced voltage sequence as the desired input. By iteratively updating the weight vector, the filter output value approximates the interference component in the induced voltage that is related to the switching interference reference sequence. The length of the weight vector of the adaptive filter is determined according to the length of the reference input vector used to characterize the correlation of the active shaft switching interference.

[0100] In this embodiment, the minimum mean square error filter is a specific implementation of an adaptive filter. This filter updates its weight vector based on the criterion of minimizing the expected value of the square of the difference between the sampled value of the induced voltage sequence at the current moment and the filter output value. The weight vector of the minimum mean square error filter is updated in the following way: the reference input vector at the current moment is multiplied by the weight vector at the current moment to obtain the estimated value of the switching interference at the current moment; the error signal at the current moment is determined based on the difference between the sampled value of the induced voltage at the current moment and the estimated value of the switching interference; the error signal, the reference input vector at the current moment, and the step size factor are multiplied by the weight vector at the current moment and added to the weight vector at the current moment to obtain the weight vector at the next moment.

[0101] In this embodiment, the desired input is the signal that serves as the target signal input in the adaptive filter. This signal is the induced voltage sequence stored in the data buffer corresponding to the standby axis identifier. The induced voltage sequence contains the induced voltage component generated by the micro-motion of the standby axis rotor and the interference component generated by the switching action of the active axis.

[0102] In this embodiment, the reference input is the signal used as the interference reference signal input in the adaptive filter. This signal is the switching interference reference sequence stored in the data buffer corresponding to the standby axis identifier. Each reference value in the switching interference reference sequence corresponds to a time scale, which is used to characterize the intensity of electromagnetic interference generated by the switching action of the active axis inverter corresponding to that time scale.

[0103] In this embodiment, the reference input vector is a vector composed of the switching interference reference values ​​corresponding to the current time and several previous sampling times extracted sequentially from the switching interference reference sequence according to the sampling time scale of the induced voltage sequence.

[0104] In this embodiment, the weight vector is a coefficient vector used in the adaptive filter to perform weighted summation on the reference input. The length of the weight vector is determined according to the length of the reference input vector used to characterize the correlation of active axis switching interference. The initial value of the weight vector is set to zero vector. The weight vector is updated in each iteration according to the weight vector update method.

[0105] In this embodiment, the step size factor is a coefficient that controls the update amplitude when updating the adaptive filter weight vector. The value of the step size factor is determined by the following steps: obtaining the maximum eigenvalue of the autocorrelation matrix of the switching interference reference sequence, multiplying the reciprocal of the maximum eigenvalue by a set coefficient to obtain the upper limit of the step size factor, and using the upper limit as the step size factor. The set coefficient ranges from 0.01 to 0.1. The maximum eigenvalue of the autocorrelation matrix is ​​obtained by performing eigenvalue decomposition on the autocorrelation matrix using the Jacobian method and then taking the maximum eigenvalue.

[0106] In this embodiment, the interference component estimate is the output value of the adaptive filter at the current time. This output value is equal to the product of the current weight vector and the current reference input vector. This output value is used to characterize the interference component in the induced voltage at the current time that is related to the switching interference reference sequence.

[0107] In this embodiment, the sampled value of the induced voltage after filtering out interference is the value obtained by subtracting the estimated value of the interference component at the current time from the sampled value at the current time in the induced voltage sequence. The interference component generated by the action of the drive shaft switch is canceled in this value.

[0108] In this embodiment, the interference-filtered induced voltage sequence is a data sequence formed by arranging the interference-filtered induced voltage sample values ​​at all times in the induced voltage sequence according to the time scale order.

[0109] In this embodiment, the stable sampling interval is the time period corresponding to the stable period of each switch action in the low-interference sampling window, starting from the moment after a preset stabilization time from the edge of the corresponding switch to the moment before the next switch state change. This stable sampling interval is read from the low-interference sampling window.

[0110] In this embodiment, the standby shaft induced voltage response subvalue is a value obtained by discretely integrating the absolute values ​​of all extracted induced voltage samples after filtering out interference. The discrete integration is obtained by multiplying the absolute value of each induced voltage sample after filtering out interference by the sampling interval corresponding to that sample value and then summing the results. This value is used to characterize the cumulative amount of induced voltage generated by the micro-motion of the standby shaft rotor within the stable sampling interval. The reason for taking the absolute value is that the induced voltage generated by the micro-motion of the standby shaft rotor alternates between positive and negative as the direction of rotor angular motion changes. Taking the absolute value and then summing the results can avoid the positive and negative components canceling each other out.

[0111] In this embodiment, the standby axis induced voltage response sub-value sequence is a data sequence formed by arranging the standby axis induced voltage response sub-values ​​corresponding to all stable sampling intervals in chronological order.

[0112] In this embodiment, the standby axis induced voltage response is a sequence of standby axis induced voltage response sub-values. This response is stored in the data buffer corresponding to the standby axis identifier for subsequent steps to compare with the expected induced voltage response range.

[0113] It should be noted that the induced voltage sequence may include thermal noise from the standby axis motor winding resistance, quantization noise from the voltage acquisition circuit, and interference components from mechanical vibrations transmitted by the robot body coupled to the standby axis motor winding through other paths. These interference components are unrelated to the switching interference reference sequence, and the adaptive filter does not cancel them. The interference components are partially smoothed by summing absolute values ​​in the subsequent discrete integration step.

[0114] It should be noted that the upper limit of the step size factor is determined by the maximum eigenvalue of the autocorrelation matrix of the switching interference reference sequence. When the step size factor exceeds the upper limit, the weight vector update process of the minimum mean square error filter does not converge. When the step size factor is lower than the upper limit, the weight vector update process of the minimum mean square error filter converges. The specific value of the step size factor is determined within the convergence range according to the weight vector update speed requirement.

[0115] In a specific embodiment, the above steps use an adaptive filter to cancel the interference component in the induced voltage that is related to the switching interference reference sequence, and perform discrete integration on the induced voltage after filtering out interference within a low-interference sampling window. The motivation for this is that the electromagnetic interference generated by the active shaft switching action is transmitted to the standby shaft motor winding through the DC bus and spatial coupling path. This interference component is related to the switching interference reference sequence. The adaptive filter can estimate and cancel this interference component. At the same time, the active shaft inverter switching action stops and the transient interference has attenuated within the low-interference sampling window. Performing discrete integration within this window can improve the signal-to-noise ratio of the standby shaft induced voltage response, providing a data source for subsequent steps to establish the correspondence between the standby shaft micro-motion angular velocity and the induced voltage response.

[0116] S5: Synchronously acquire the rotor angular displacement signal output by the standby shaft encoder during the movement of the drive shaft, determine the micro-motion angular displacement sequence of the standby shaft rotor based on the rotor angular displacement signal, and determine the micro-motion angular velocity sequence of the standby shaft rotor based on the micro-motion angular displacement sequence.

[0117] In this embodiment, the standby shaft encoder synchronously acquires the rotor angular displacement signal output during the movement of the drive shaft, determines the micro-motion angular displacement sequence of the standby shaft rotor based on the rotor angular displacement signal, and determines the micro-motion angular velocity sequence of the standby shaft rotor based on the micro-motion angular displacement sequence.

[0118] Based on the data buffer corresponding to the standby axis identifier, the micro-motion angular velocity sequence and the start and end times of each stable sampling interval in the low-interference sampling window are read from the data buffer.

[0119] The micro-motion angular velocity corresponding to each sampling moment is determined based on the rotor angular displacement change at adjacent sampling moments in the micro-motion angular velocity sequence, and the absolute value of the micro-motion angular velocity is taken to obtain the micro-motion angular velocity amplitude sequence.

[0120] The pre-established normal electromagnetic response relationship of the standby axis is obtained. The normal electromagnetic response relationship of the standby axis is established in the following way: During the normal operation of the robot, for the standby axis, the amplitude of the micro-motion angular velocity collected during the normal operation is divided into multiple continuous angular velocity intervals. Using each angular velocity interval as an index, the induced voltage response of the standby axis in the corresponding angular velocity interval is recorded respectively, and the corresponding normal induced voltage response range is determined according to the induced voltage response distribution in each angular velocity interval.

[0121] For each stable sampling interval in the low-interference sampling window, extract all micro-motion angular velocity amplitudes with time stamps between the start and end times of the stable sampling interval from the micro-motion angular velocity amplitude sequence.

[0122] The average value of all extracted micro-motion angular velocity values ​​is taken as the representative micro-motion angular velocity value corresponding to the stable sampling interval.

[0123] In the normal electromagnetic response relationship of the standby axis, find the angular velocity range that contains the amplitude of the micro-motion angular velocity, and read the normal induced voltage response range corresponding to the angular velocity range.

[0124] Arrange the normal induced voltage response ranges corresponding to each stable sampling interval in chronological order to form a sequence of expected induced voltage response ranges, and store the expected induced voltage response range sequence in the data buffer corresponding to the standby axis identifier.

[0125] The normal induced voltage response range can be determined based on the distribution range of the normal induced voltage response within the corresponding angular velocity range; in one embodiment, the normal induced voltage response range is determined based on the mean and standard deviation of the induced voltage response within the corresponding angular velocity range.

[0126] The following are feasible arithmetic expressions for calculating the angular displacement increment and the average angular velocity: ;

[0127] In the formula, i is the sampling time number in the original rotor angular displacement sequence, and its value is an integer greater than or equal to 1 and less than or equal to N-1; N is the total number of sampling times in the original rotor angular displacement sequence; The angular displacement value at the i-th sampling moment in the original sequence of rotor angular displacement is output and read by the standby shaft encoder; This represents the angular displacement value at the (i+1)th sampling moment in the original sequence of rotor angular displacements; This represents the angular displacement increment between the i-th sampling time and the (i+1)-th sampling time. This is the time stamp of the i-th sampling moment in the original sequence of rotor angular displacement; This is the time stamp of the (i+1)th sampling moment in the original sequence of rotor angular displacement, in seconds; It represents the average angular velocity value between the i-th sampling time and the (i+1)-th sampling time.

[0128] The following is a feasible arithmetic expression for calculating encoder interpolation: ; In the formula, This is the interpolated angular displacement output after encoder interpolation processing; The angular displacement corresponding to the original encoder signal is obtained by multiplying the original encoder pulse count by the encoder resolution. This is the subdivision factor of the encoder's internal subdivision circuit, ranging from 4 to 256, and is read from the encoder's factory calibration data.

[0129] It should be noted that in the above formula, the average angular velocity values ​​are arranged in the time scale order to form a micro-motion angular velocity sequence. Each average angular velocity value in this micro-motion angular velocity sequence corresponds to a time scale interval, and the starting time of this time scale interval is... The end time is .

[0130] The encoder interpolation value is the minimum resolution of the encoder's output angular displacement value, that is, the smallest unit of angular displacement that the encoder can distinguish. The angular displacement value read by the controller from the encoder is already a value processed by the encoder's internal subdivision circuit, and this value implicitly includes the encoder interpolation value, but the controller does not need to calculate this interpolation value separately. The purpose of the encoder interpolation value is to indicate that the smallest non-zero value of the angular displacement increment is equal to the encoder interpolation value. It is used to determine whether the encoder can distinguish when the standby shaft rotor micro-motion amplitude is too small. It belongs to the constraint condition for encoder selection or sampling parameter setting, and is not a calculation step in the main process.

[0131] In this embodiment, the rotor angular displacement signal is a pulse signal or digital signal output by the standby shaft encoder during the movement of the drive shaft, which characterizes the change in the angular position of the standby shaft rotor. The signal is synchronously output by the encoder according to the sampling frequency, and each sampling moment corresponds to an angular displacement value. The unit of the angular displacement value is degrees or radians, which is determined by the resolution of the encoder.

[0132] In this embodiment, the original rotor angular displacement sequence is a data sequence formed by arranging the collected rotor angular displacement signals according to the time scale order. Each angular displacement value in the data sequence corresponds to a time scale, and the data sequence is stored in the data buffer corresponding to the standby axis identifier.

[0133] In this embodiment, the angular displacement increment is the difference between the angular displacement values ​​of two adjacent sampling times in the original sequence of rotor angular displacement. The sign of the difference indicates the rotation direction of the standby shaft rotor between the two adjacent sampling times, and the absolute value of the difference indicates the rotation amplitude of the standby shaft rotor between the two adjacent sampling times.

[0134] In this embodiment, the average angular velocity value is obtained by dividing the increment of angular displacement by the difference in time scale between two adjacent sampling times. The unit of this value is radians per second, and this value represents the average rotational speed of the standby shaft rotor between the two adjacent sampling times.

[0135] In this embodiment, the micro-motion angular velocity sequence is a data sequence formed by arranging the average angular velocity values ​​between all two adjacent sampling times in the original rotor angular displacement sequence according to the time scale order. Each average angular velocity value in the data sequence corresponds to a time scale interval, and the data sequence is stored in the data buffer corresponding to the standby axis identifier.

[0136] In this embodiment, the micro-motion angular displacement sequence is a data sequence formed by arranging the angular displacement values ​​at each sampling time in the original rotor angular displacement sequence according to the time scale order. Each angular displacement value in the data sequence corresponds to a time scale, and the data sequence is stored in the data buffer corresponding to the standby axis identifier.

[0137] It should be noted that the standby shaft encoder is an encoder capable of outputting angular displacement values. When the micro-motion angular displacement amplitude of the standby shaft rotor is less than the angular displacement corresponding to one pulse of a normal incremental encoder, a high-resolution encoder, a magnetic encoder, an absolute encoder, or a position signal output device with interpolation processing is used. The high-resolution encoder is an incremental encoder that outputs more than a set pulse number threshold per revolution. The set pulse number threshold ranges from 4096 to 16384.

[0138] It should be noted that the time scale of each value in the micro-motion angular displacement sequence and the micro-motion angular velocity sequence is consistent with the time scale of the corresponding sampling time in the induced voltage sequence, so that each average angular velocity value in the micro-motion angular velocity sequence can correspond to the induced voltage response of the corresponding time scale interval in the induced voltage sequence in subsequent steps.

[0139] It should be noted that when the micro-motion angular displacement amplitude of the standby shaft rotor during the movement of the drive shaft is greater than the angular displacement corresponding to one encoder pulse, the difference between the angular displacement values ​​at two adjacent sampling moments in the original rotor angular displacement sequence is non-zero; when the micro-motion angular displacement amplitude of the standby shaft rotor during the movement of the drive shaft is less than the angular displacement corresponding to one encoder pulse, the difference between the angular displacement values ​​at two adjacent sampling moments in the original rotor angular displacement sequence is zero. At this time, the interpolated amount of the angular displacement is output by the encoder's interpolation processing output device. The interpolated amount is obtained by the encoder's internal subdivision circuit subdividing the encoder's original signal.

[0140] In a specific embodiment, the above steps synchronously acquire the rotor angular displacement signal output by the standby shaft encoder at the sampling frequency, and calculate the average angular velocity value based on the angular displacement increment between two adjacent sampling times. The motivation is to provide actual motion data of the standby shaft rotor for subsequent steps to establish the correspondence between the standby shaft micro-motion angular velocity and the induced voltage response. This allows subsequent steps to find the corresponding expected induced voltage response range in the normal electromagnetic response relationship of the standby shaft based on the actual micro-motion angular velocity, rather than judging whether there is a mechanical response abnormality in the standby shaft solely based on the back electromotive force itself.

[0141] S6: Based on the micro-motion angular velocity sequence, find the expected induced voltage response range corresponding to the micro-motion angular velocity sequence in the pre-established standby axis normal electromagnetic response relationship. The standby axis normal electromagnetic response relationship is the correspondence between the standby axis micro-motion angular velocity and the induced voltage response recorded during the normal operation of the robot.

[0142] In this embodiment, the step of searching for the expected induced voltage response range corresponding to the micro-motion angular velocity sequence in the pre-established normal electromagnetic response relationship of the standby axis, based on the micro-motion angular velocity sequence, is as follows: The normal electromagnetic response relationship of the standby axis is the correspondence between the micro-motion angular velocity of the standby axis and the induced voltage response recorded during normal robot operation. Specifically: Based on the data buffer corresponding to the standby axis identifier, the micro-motion angular velocity sequence and the start and end times of each stable sampling interval in the low-interference sampling window are read from the data buffer.

[0143] The pre-established normal electromagnetic response relationship of the standby axis is obtained. The normal electromagnetic response relationship of the standby axis is established in the following way: During the normal operation of the robot, for the standby axis, the mean and standard deviation of the induced voltage response of the standby axis in each angular velocity range are recorded using a set angular velocity range as an index. The product of the mean minus the standard deviation multiple and the standard deviation is used as the lower limit of the expected induced voltage response range, and the product of the mean plus the standard deviation multiple and the standard deviation is used as the upper limit of the expected induced voltage response range.

[0144] For each stable sampling interval within a low-interference sampling window, extract all average angular velocity values ​​from the micro-motion angular velocity sequence whose time scales lie between the start and end times of that stable sampling interval.

[0145] Calculate the average of all extracted average angular velocity values, and determine this average value as the representative angular velocity value for the stable sampling interval.

[0146] In the normal electromagnetic response relationship of the standby axis, find the angular velocity range to which the representative angular velocity value belongs, and read the expected induced voltage response range corresponding to the angular velocity range.

[0147] Arrange the expected induced voltage response ranges corresponding to all stable sampling intervals in chronological order to form an expected induced voltage response range sequence, and store the expected induced voltage response range sequence into the data buffer corresponding to the standby axis identifier.

[0148] It should be noted that when establishing the normal electromagnetic response relationship of the standby axis, the width of the set angular velocity range needs to be determined first. The following is a feasible arithmetic expression for determining the width of the set angular velocity range: ; In the formula, To set the width of the angular velocity range, This represents the maximum value of the standby axis micro-motion angular velocity during normal robot operation. This represents the minimum standby axis micro-motion angular velocity during normal robot operation. To define the number of intervals, Q is an integer greater than 1.

[0149] The following are feasible arithmetic expressions for determining the upper and lower limits of the j-th set angular velocity interval: ; ; In the formula, j is the index of the set angular velocity range, and its value is an integer greater than or equal to 1 and less than or equal to Q; This is the lower limit value of the j-th set angular velocity range; This is the upper limit value of the j-th set angular velocity range.

[0150] The following are feasible arithmetic expressions for determining the upper and lower limits of the expected induced voltage response range corresponding to the j-th set angular velocity interval: ; ; In the formula, This is the lower limit of the expected induced voltage response range corresponding to the j-th set angular velocity interval; This represents the upper limit of the expected induced voltage response range corresponding to the j-th set angular velocity range; This is the mean value corresponding to the j-th set angular velocity interval; Let be the standard deviation corresponding to the j-th set angular velocity interval.

[0151] It should be noted that when establishing the normal electromagnetic response relationship of the standby axis, a lower limit and an upper limit are determined for each set angular velocity range. The purpose is to provide the range boundary for the subsequent search of the angular velocity range to which the representative angular velocity value belongs. In subsequent steps, the representative angular velocity value is compared with the lower limit and upper limit of each set angular velocity range one by one. The range number that satisfies the condition that the representative angular velocity value is greater than or equal to the lower limit of the range and less than the upper limit of the range is found. Then, the expected induced voltage response range corresponding to the range number is read. Therefore, the calculation of the upper and lower limits of the set angular velocity range is a prerequisite for the subsequent search action.

[0152] In this embodiment, the normal electromagnetic response relationship of the standby axis is the correspondence between the micro-motion angular velocity amplitude of the standby axis and the induced voltage response of the standby axis, which is recorded during the normal operation of the robot. This correspondence is indexed by a set angular velocity range, and each set angular velocity range corresponds to a set of mean and standard deviation of the induced voltage response of the standby axis. This correspondence is stored in the memory of the crawling robot controller and is obtained by collecting the micro-motion angular velocity amplitude sequence of the standby axis and the induced voltage response of the standby axis and performing statistics according to the set angular velocity range during the normal operation of the robot.

[0153] In this embodiment, the set angular velocity interval is a series of consecutive intervals obtained by dividing the range of values ​​of the micro-motion angular velocity amplitude of the standby axis. Each interval has an equal width, and the interval width is determined by the following steps: obtaining the minimum and maximum values ​​of the micro-motion angular velocity amplitude of the standby axis during normal robot operation, dividing the difference between the maximum and minimum values ​​by the number of set intervals to obtain the interval width. The number of set intervals ranges from 5 to 20, and the lower limit and upper limit of each set angular velocity interval are determined by increasing sequentially according to the interval width.

[0154] In this embodiment, the mean is the arithmetic mean of the induced voltage responses of all standby axes within a certain set angular velocity range during normal robot operation. The mean is calculated by the following steps: obtaining the sub-values ​​of the induced voltage responses of all standby axes within the set angular velocity range, adding all the sub-values ​​of the induced voltage responses of all standby axes and dividing by the total number of sub-values ​​of the induced voltage responses of standby axes within the set angular velocity range to obtain the mean corresponding to the set angular velocity range.

[0155] In this embodiment, the standard deviation is a measure of the degree of deviation of the induced voltage response of all standby axes within a certain set angular velocity range from the mean value corresponding to the set angular velocity range during normal operation of the robot. The standard deviation is calculated by the following steps: obtaining the difference between the induced voltage response sub-value of each standby axis within the set angular velocity range and the mean value corresponding to the set angular velocity range; squaring each difference; summing all the squared values ​​and dividing by the total number of induced voltage response sub-values ​​of standby axes within the set angular velocity range; and taking the square root of the result to obtain the standard deviation corresponding to the set angular velocity range.

[0156] In this embodiment, the standard deviation multiple is a coefficient used to determine the upper and lower limits of the expected induced voltage response range. The value of the standard deviation multiple is determined by the following steps: obtaining the allowable false alarm rate and allowable false alarm rate of the standby axis mechanical response anomaly, finding the corresponding quantile in the standard normal distribution table according to the allowable false alarm rate and allowable false alarm rate, and determining the quantile as the standard deviation multiple. The value of the standard deviation multiple is in the range of 2 to 3.

[0157] In this embodiment, the expected range of induced voltage response is the normal range of standby axis induced voltage response within a certain set angular velocity range during normal robot operation. The lower limit of this range is equal to the product of the mean value corresponding to the set angular velocity range minus the standard deviation multiple and the standard deviation, and the upper limit of this range is equal to the product of the mean value corresponding to the set angular velocity range plus the standard deviation multiple and the standard deviation.

[0158] In this embodiment, the micro-motion angular velocity amplitude sequence is a data sequence formed by taking the absolute value of each average angular velocity value in the micro-motion angular velocity sequence and arranging them in the time-scale order.

[0159] In this embodiment, the representative angular velocity value is the arithmetic mean of all micro-motion angular velocity amplitudes within a certain stable sampling interval. The representative angular velocity value is calculated through the following steps: extracting all micro-motion angular velocity amplitudes with time scales between the start and end times of the stable sampling interval from the micro-motion angular velocity amplitude sequence; adding all the extracted micro-motion angular velocity amplitudes together and dividing by the total number of extracted micro-motion angular velocity amplitudes to obtain the representative angular velocity value of the stable sampling interval.

[0160] In this embodiment, the expected induced voltage response range sequence is a data sequence formed by arranging the expected induced voltage response ranges corresponding to all stable sampling intervals in chronological order. Each expected induced voltage response range in this data sequence corresponds to a stable sampling interval.

[0161] It should be noted that when establishing the normal electromagnetic response relationship of the standby axis, a lower limit and an upper limit are determined for each set angular velocity range. The purpose of this is to provide the range boundary for the subsequent search of the angular velocity range to which the representative angular velocity value belongs. In subsequent steps, the representative angular velocity value is compared with the lower limit and upper limit of each set angular velocity range one by one. The range number that satisfies the condition that the representative angular velocity value is greater than or equal to the lower limit of the range and less than the upper limit of the range is found. Then, the expected induced voltage response range corresponding to the range number is read. Therefore, the calculation of the upper and lower limits of the set angular velocity range is a prerequisite for the subsequent search operation.

[0162] It should be noted that when the total number of standby axis induced voltage response sub-values ​​recorded within a certain set angular velocity range is less than the set sample number threshold, the mean and standard deviation corresponding to the set angular velocity range are obtained by linear interpolation of the mean and standard deviation of two adjacent set angular velocity ranges. The set sample number threshold ranges from 10 to 30.

[0163] It should be noted that the normal electromagnetic response relationship of the standby axis is continuously updated during the normal operation of the robot. Whenever the standby axis completes the acquisition of induced voltage and micro-motion angular velocity during the movement of the active axis, the induced voltage response and micro-motion angular velocity amplitude of the standby axis acquired in this acquisition are assigned to the corresponding interval according to the set angular velocity interval, and the mean and standard deviation of the interval are recalculated.

[0164] In a specific embodiment, the above steps find the expected induced voltage response range corresponding to the micro-motion angular velocity amplitude sequence by using a pre-established normal electromagnetic response relationship of the standby axis. The motivation is that the induced voltage response of the standby axis depends not only on the micro-motion angular velocity amplitude of the standby axis rotor, but also on factors such as the magnetic flux linkage and mechanical coupling state of the standby axis motor. By recording the statistical distribution of the induced voltage response under different micro-motion angular velocity amplitudes during normal robot operation, a normal response range corresponding to the current micro-motion angular velocity amplitude can be provided for subsequent comparison steps, avoiding misjudgment caused by changes in the micro-motion angular velocity amplitude when using a fixed threshold for judgment.

[0165] S7: Compare the standby axis induced voltage response with the expected induced voltage response range. If the standby axis induced voltage response is outside the expected induced voltage response range, it is determined that there is an abnormal deviation between the electromagnetic response of the standby axis and the actual micro-motion, and the detection result of the standby axis having an abnormal mechanical response is output.

[0166] In this embodiment, the step of comparing the standby axis induced voltage response with the expected induced voltage response range, and if the standby axis induced voltage response is outside the expected induced voltage response range, then it is determined that there is an abnormal deviation between the electromagnetic response of the standby axis and the actual micro-motion, and a detection result indicating that the standby axis has an abnormal mechanical response is output. Specifically: Based on the data buffer corresponding to the standby axis identifier, read the standby axis induced voltage response sub-value sequence and the expected induced voltage response range sequence from the data buffer.

[0167] Each sub-value in the standby axis induced voltage response sub-value sequence is matched one-to-one with each range in the expected induced voltage response range sequence in chronological order to obtain multiple comparison pairs. Each comparison pair contains a standby axis induced voltage response sub-value and an expected induced voltage response range.

[0168] For each comparison pair, determine whether the standby axis induced voltage response sub-value in the comparison pair is less than the lower limit of the expected induced voltage response range in the comparison pair, or greater than the upper limit of the expected induced voltage response range in the comparison pair.

[0169] If the standby axis induced voltage response sub-value in the comparison pair is less than the lower limit of the expected induced voltage response range in the comparison pair, then the stable sampling interval corresponding to the comparison pair is determined as an abnormal sampling interval, and the abnormality type of the abnormal sampling interval is recorded as low response.

[0170] If the standby axis induced voltage response sub-value in the comparison pair is greater than the upper limit of the expected induced voltage response range in the comparison pair, then the stable sampling interval corresponding to the comparison pair is determined as the abnormal sampling interval, and the abnormal type of the abnormal sampling interval is recorded as high response.

[0171] If the standby axis induced voltage response sub-values ​​in all comparison pairs are greater than or equal to the lower limit of the corresponding expected induced voltage response range and less than or equal to the upper limit of the corresponding expected induced voltage response range, then the detection result that there is no mechanical response abnormality for the standby axis is output.

[0172] If at least one abnormal sampling interval exists, the detection result of the abnormal mechanical response of the standby axis is output, along with the standby axis identifier, the sequence number of each abnormal sampling interval, the sub-value of the induced voltage response of the standby axis corresponding to each abnormal sampling interval, the lower and upper limits of the expected induced voltage response range corresponding to each abnormal sampling interval, and the abnormality type of each abnormal sampling interval.

[0173] In this embodiment, the standby axis induced voltage response sub-value sequence is a data sequence formed by arranging the standby axis induced voltage response sub-values ​​corresponding to all stable sampling intervals in chronological order. This data sequence is stored in the data buffer corresponding to the standby axis identifier, and each sub-value in the data sequence corresponds to a stable sampling interval.

[0174] In this embodiment, the expected induced voltage response range sequence is a data sequence formed by arranging the expected induced voltage response ranges corresponding to all stable sampling intervals in chronological order. This data sequence is stored in the data buffer corresponding to the standby axis identifier. Each range in this data sequence corresponds to a stable sampling interval, and each range contains a lower limit value and an upper limit value.

[0175] In this embodiment, the comparison pair is a binary pair consisting of a sub-value in the standby axis induced voltage response sub-value sequence and a range in the expected induced voltage response range sequence at the same time sequence position. In each comparison pair, the sub-value and the range correspond to the same stable sampling interval.

[0176] In this embodiment, the abnormal sampling interval is the stable sampling interval corresponding to the standby axis induced voltage response sub-value when it is outside the corresponding expected induced voltage response range. The stable sampling interval is determined from the low interference sampling window in chronological order.

[0177] In this embodiment, the anomaly type is an identifier used to distinguish the direction in which the standby axis induced voltage response sub-value deviates from the expected induced voltage response range. The identifier includes low response and high response. Low response indicates that the standby axis induced voltage response sub-value is less than the lower limit of the corresponding expected induced voltage response range, and high response indicates that the standby axis induced voltage response sub-value is greater than the upper limit of the corresponding expected induced voltage response range.

[0178] In this embodiment, a low response refers to the abnormal type corresponding to the sub-value of the standby axis induced voltage response when it is less than the lower limit of the expected induced voltage response range. This abnormal type indicates that the induced voltage generated by the standby axis under the actual micro-motion angular velocity is lower than the statistical lower limit during normal operation.

[0179] In this embodiment, a high response is an anomaly type corresponding to the standby axis induced voltage response sub-value when it is greater than the upper limit of the expected induced voltage response range. This anomaly type indicates that the induced voltage generated by the standby axis under the actual micro-motion angular velocity is higher than the statistical upper limit during normal operation.

[0180] It should be noted that when there is at least one abnormal sampling interval, the detection result of the standby axis having abnormal mechanical response is output. The detection result may also include the number of abnormal sampling intervals. When the number of abnormal sampling intervals is greater than the set number threshold, the abnormality level of the detection result is determined to be high. When the number of abnormal sampling intervals is less than or equal to the set number threshold, the abnormality level of the detection result is determined to be low. The set number threshold ranges from 1 to 3.

[0181] It should be noted that when the abnormality type is low response, the standby shaft may have brake drag, short circuit between winding turns, or decreased flux linkage; when the abnormality type is high response, the standby shaft may have excessive brake release, low encoder reading, or loose mechanical coupling. The above fault types are only for reference information for subsequent maintenance and do not constitute the basis for judgment in this step.

[0182] In a specific embodiment, the above steps compare the standby axis induced voltage response sub-value of each stable sampling interval with the corresponding expected induced voltage response range one by one. The motivation for this is that the standby axis induced voltage response changes with the micro-motion angular velocity. If the overall average of all stable sampling intervals is used for judgment, the abnormal response of individual stable sampling intervals will be averaged by other normal responses, resulting in the abnormality not being detected. Therefore, it is necessary to compare one by one according to the stable sampling intervals so that the abnormal response of any stable sampling interval can be detected.

[0183] Example 2: This invention also includes a risk warning system for a six-way crawling rack based on multi-channel distributed drive, comprising a shaft system determination module, a sensing acquisition module, an interference synchronization module, an interference suppression module, a micro-motion acquisition module, an expectation determination module, and an anomaly identification module. The shaft system determination module is used to determine the active shaft and the standby shaft. The sensing acquisition module is used to acquire the induced voltage of the standby shaft winding during the movement of the active shaft. The interference synchronization module is used to synchronously acquire the PWM switching state of the active shaft, determine the switching interference reference and the low-interference sampling window. The interference suppression module is used to eliminate related switching interference in the induced voltage of the standby shaft using the PWM switching state of the active shaft, and extract the induced voltage response of the standby shaft in the low-interference sampling window. The micro-motion acquisition module is used to synchronously acquire the micro-motion angular velocity of the standby shaft rotor. The expectation determination module is used to determine the expected induced voltage response range based on the correspondence between the micro-motion angular velocity of the standby shaft and the normal induced voltage response. The anomaly identification module is used to compare the actual induced voltage response with the expected induced voltage response, and identify mechanical response anomalies of the standby shaft based on the deviation between the two.

Claims

1. A risk warning method for a six-way crawling rack based on multi-path distributed drive, characterized in that, Includes the following steps: Determine the drive axis and standby axis; The induced voltage of the standby shaft winding is collected during the movement of the active shaft. Synchronously acquire the PWM switching status of the active axis to determine the switching interference reference and low-interference sampling window; The relevant switching interference in the standby axis induced voltage is eliminated by utilizing the active axis PWM switching state, and the standby axis induced voltage response is extracted in the low-interference sampling window; Synchronously acquire the micro-angular velocity of the standby shaft rotor; The expected induced voltage response range is determined based on the correspondence between the standby axis micro-motion angular velocity and the normal induced voltage response; Compare the actual induced voltage response with the expected induced voltage response, and identify abnormal mechanical response of the standby axis based on the deviation between the two.

2. The method according to claim 1, characterized in that, The determination of the active axis and the standby axis specifically involves: The current direction of motion is determined according to the motion control command of the crawling robot. The drive unit corresponding to the current direction of motion is determined as the active axis, and the drive unit outside the current direction of motion, which is in a non-drive state and is mechanically coupled to the robot body, is determined as the standby axis.

3. The method according to claim 2, characterized in that, The process of collecting the induced voltage of the standby shaft winding during the movement of the drive shaft is specifically as follows: The inverter switching transistor controlling the standby axis motor is put into a blocked state, so that the standby axis motor winding is in a non-drive voltage detection state; During the movement of the active shaft, the voltage between the terminals of the two-phase windings of the standby shaft motor is collected to form a standby shaft induced voltage sequence.

4. The method according to claim 3, characterized in that, The determination of the switching interference reference and low-interference sampling window is specifically as follows: The PWM switching state and switching edge time of each switching transistor of the active shaft inverter are obtained. Based on the PWM switching state, a switching interference reference sequence corresponding to the active shaft switching action is constructed. Based on the switching edge time, a low-interference sampling window is determined after the active shaft switching action stops.

5. The method according to claim 4, characterized in that, The extraction of the standby axis induced voltage response specifically includes: Using the switching interference reference sequence as the adaptive filtering reference input, the interference components related to the active axis switching action in the standby axis induced voltage sequence are canceled, and the amplitude of the induced voltage after the interference is canceled is integrated within the low interference sampling window to form the standby axis induced voltage response sequence.

6. The method according to claim 5, characterized in that, The process of synchronously acquiring the micro-angular velocity of the standby shaft rotor specifically involves: While acquiring the standby axis induced voltage, the rotor angular displacement sequence output by the standby axis encoder is obtained. The corresponding micro-motion angular velocity is determined according to the rotor angular displacement change at adjacent sampling times. The micro-motion angular velocity is segmented according to the low-interference sampling window to form a micro-motion angular velocity response sequence corresponding to the standby axis induced voltage response sequence.

7. The method according to claim 6, characterized in that, The process of determining the expected induced voltage response range and identifying mechanical response anomalies of the standby axis specifically includes: Based on the correspondence between the micro-motion angular velocity response and the induced voltage response formed during the normal operation of the standby axis, the normal induced voltage response range corresponding to each micro-motion angular velocity response is determined. The standby axis induced voltage response sequence is compared with the corresponding normal induced voltage response range, and the mechanical response abnormality of the standby axis is determined based on the response deviation that occurs in multiple consecutive sampling intervals.

8. A system using the method as described in any one of claims 1-7, characterized in that, It includes a shaft system determination module, a sensing and acquisition module, an interference synchronization module, an interference suppression module, a micro-motion acquisition module, an expected determination module, and an anomaly identification module; Shaft system determination module, used to determine the drive shaft and standby shaft; The sensing and acquisition module is used to acquire the induced voltage of the standby shaft winding during the movement of the drive shaft; The interference synchronization module is used to synchronously acquire the PWM switching status of the active shaft, determine the switching interference reference and low interference sampling window; The interference suppression module is used to eliminate the relevant switching interference in the standby axis induced voltage by utilizing the active axis PWM switching state, and to extract the standby axis induced voltage response in the low interference sampling window. The micro-motion acquisition module is used to synchronously acquire the micro-motion angular velocity of the standby shaft rotor; The expected determination module is used to determine the expected induced voltage response range based on the correspondence between the standby axis micro-motion angular velocity and the normal induced voltage response; The anomaly identification module is used to compare the actual induced voltage response with the expected induced voltage response, and to identify mechanical response anomalies of the standby axis based on the deviation between the two.