A track topology mapping, positioning and prediction control method and system based on electrical load fingerprints

Abstract

The application provides a track topology mapping, positioning and predictive control method and system based on electrical load fingerprints, which comprises the following steps: acquiring motor armature voltage signals and motor armature current signals of a motor driving device in real time; estimating the real-time rotating speed of the motor according to the motor armature voltage signals and the motor armature current signals, and time-integrating the real-time rotating speed to obtain real-time virtual mileage; mapping the motor armature current signals in the time domain to the virtual mileage coordinate system to form the motor armature current signals in the space domain, and taking the motor armature current signals in the space domain as the current load characteristic signals; performing sequence matching on the current load characteristic signals and the reference load fingerprint atlas to determine the absolute position of the motor driving device on the track; acquiring the road section characteristics within the predetermined virtual mileage ahead from the reference load fingerprint atlas according to the absolute position; and generating predictive safety instructions if the road section characteristics indicate that the area ahead is a high-load risk area. The application enhances the position sensing and intelligent control functions.

Description

Technical Field

[0001] This application relates to the field of automation control technology, specifically to a method and system for track topology mapping, positioning and predictive control based on electrical load fingerprinting. Background Technology

[0002] In applications such as rail transit models (e.g., racing cars), industrial rail transport systems, and automated guided vehicles (AGVs), vehicles or equipment are typically confined to fixed physical tracks. To achieve intelligent control, such as automatically decelerating before curves, precisely stopping at specific stations, or achieving energy-saving operation, the system needs to obtain the vehicle's precise position information on the track in real time and understand the layout characteristics of the track ahead (e.g., straight sections, curves, gradients).

[0003] Currently, the mainstream technical solutions for achieving the aforementioned positional awareness rely on adding external sensors, such as photoelectric encoders, machine vision cameras, or LiDAR, to vehicles or tracks. However, these solutions have a significant drawback: to achieve positioning and awareness, additional dedicated sensor hardware must be introduced, which increases the overall system cost, complexity, and maintenance difficulty. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this application provides a method and system for track topology mapping, positioning and predictive control based on electrical load fingerprinting.

[0005] The specific technical solution is as follows:

[0006] In a first aspect, this application provides a method for track topology mapping, positioning and predictive control based on electrical load fingerprint, which is applied to motor drive equipment running on a fixed track. The method is executed based on a pre-established reference load fingerprint map of the track, which is a map representing the load characteristics and position mapping relationship of the track throughout the entire track in a virtual mileage coordinate system.

[0007] The method includes:

[0008] Real-time acquisition of the motor armature voltage signal and armature current signal of the motor drive device;

[0009] The real-time speed of the motor is estimated based on the armature voltage signal and the armature current signal, and the real-time virtual mileage is obtained by time integration of the real-time speed.

[0010] Based on the real-time virtual mileage, the armature current signal in the time domain is mapped to the virtual mileage coordinate system to form the armature current signal in the spatial domain, which is then used as the current load characteristic signal.

[0011] The current load characteristic signal is sequentially matched with the reference load fingerprint map to determine the absolute position of the motor drive device on the track.

[0012] Based on the absolute position, road segment features within a predetermined virtual mileage ahead are obtained from the baseline load fingerprint map; if the road segment features indicate that the area ahead is a high load risk area, a predictive safety instruction is generated to intervene in the original control instructions of the motor drive equipment.

[0013] In one embodiment, the baseline load fingerprint is pre-established through the following steps:

[0014] Control the motor drive device to complete at least one full revolution along the fixed track;

[0015] During operation, the motor armature voltage signal and armature current signal are acquired in real time, and the real-time speed of the motor is estimated based on the armature voltage signal and armature current signal.

[0016] The real-time rotational speed is integrated over time to obtain the virtual mileage, and the armature current signal in the time domain is mapped to the virtual mileage coordinate system based on the virtual mileage to form the armature current signal in the spatial domain, which is used as the mapping load characteristic signal.

[0017] Record and process the mapping load feature signal sequence corresponding to the entire orbit, identify and label the high load feature segments and low load feature segments in the load feature signal sequence, thereby constructing the benchmark load fingerprint map that characterizes the load distribution and positional relationship throughout the orbit.

[0018] In one embodiment, estimating the real-time speed of the motor based on the armature voltage signal and the armature current signal includes:

[0019] A thermal model of the motor is established, which is used to characterize the dynamic process of motor coil temperature rise caused by Joule heating effect of armature current;

[0020] A back EMF observer is constructed based on the electrical equations of the motor, wherein the electrical equations at least include the relationship between armature voltage, armature current, armature resistance and back EMF.

[0021] Based on the armature current signal acquired in real time, the current temperature of the motor coil is estimated in real time through the thermal model, and the armature resistance parameter value in the back EMF observer is compensated in real time according to the current temperature.

[0022] The armature voltage signal, the armature current signal, and the compensated armature resistance parameter value are obtained in real time and input into the back electromotive force observer to calculate the instantaneous value of the back electromotive force of the motor.

[0023] The real-time rotational speed is calculated based on the instantaneous value of the back electromotive force and the back electromotive force constant of the motor.

[0024] In one embodiment, the current load feature signal is sequentially matched with the reference load fingerprint spectrum, and the matching algorithm is implemented using a sliding window algorithm.

[0025] The sliding window matching algorithm includes:

[0026] Obtain a segment of the current load characteristic signal with a fixed virtual mileage length as the current matching segment;

[0027] The current matching segment is slid along the virtual mileage coordinates of the baseline load fingerprint map, and the similarity between the current matching segment and the corresponding segment of the baseline load fingerprint map is calculated at each sliding position.

[0028] The position in the benchmark load fingerprint spectrum corresponding to the extreme value of the similarity is determined as the absolute position of the motor drive device.

[0029] In one embodiment, the similarity is calculated using at least one of the following algorithms: cross-correlation algorithm, dynamic time warping algorithm, or minimum absolute difference algorithm.

[0030] In one embodiment, the step of generating predictive safety instructions to intervene in the original control instructions of the motor drive device includes:

[0031] Define a look-ahead buffer associated with the absolute position, and dynamically calculate the safety instruction limit based on the road segment characteristics within the look-ahead buffer;

[0032] Obtain the original control command of the motor drive device, and compare the original control command with the safety command limit; wherein, the original control command is used to set the desired speed or desired torque of the motor;

[0033] If the original control command exceeds the safety command limit, the safety command limit will be output as the final command; otherwise, the original control command will be output as the final command.

[0034] In one embodiment, the original control command originates from a higher-level control source with a command transmission delay;

[0035] The method further includes:

[0036] Monitor the instruction transmission delay of the upper-level control source;

[0037] When the instruction transmission delay exceeds a preset delay threshold, and the road segment characteristics indicate that the area ahead is a high-load risk area, a safety arbitration is performed;

[0038] The safety arbitration includes: temporarily blocking or attenuating the original control commands from the upper-level control source according to a preset ratio, and replacing them with autonomously generated target safety commands for enabling the motor drive equipment to safely pass through the risk area.

[0039] In one embodiment, the upper-level control source with inherent delay characteristics is one of a brain-computer interface, a cloud-based remote control system, or a network teleoperation system.

[0040] In one embodiment, the method further includes performing the following operations periodically or according to preset triggering conditions during the operation of the motor drive device:

[0041] Based on the absolute position determined by the sequence matching, obtain the spectral segment corresponding to the current load characteristic signal;

[0042] Using the current load characteristic signal, the corresponding spectrum segment in the benchmark load fingerprint spectrum is updated.

[0043] The data update process includes incremental updates or weighted smoothing.

[0044] Secondly, this application provides a track topology mapping, positioning, and predictive control system based on electrical load fingerprinting, applied to motor-driven equipment running on a fixed track, the system comprising:

[0045] The signal acquisition module is used to acquire the motor armature voltage signal and armature current signal of the motor drive device in real time;

[0046] A data processing module, connected to the signal acquisition module, is used to estimate the real-time speed of the motor based on the armature voltage signal and armature current signal, and to perform time integration on the real-time speed to obtain the real-time virtual mileage; based on the real-time virtual mileage, the armature current signal in the time domain is mapped to the virtual mileage coordinate system to form a spatial armature current signal, which is used as the current load characteristic signal; the current load characteristic signal is matched sequentially with a reference load fingerprint map to determine the absolute position of the motor drive device on the track; based on the absolute position, the road segment characteristics within a predetermined virtual mileage ahead are obtained from the reference load fingerprint map; if the road segment characteristics indicate a high-load risk area ahead, a predictive safety command is generated to intervene in the original control command of the motor drive device;

[0047] The instruction arbitration module, connected to the data processing module, is used to receive the predictive safety instruction and the original control instruction from the motor drive device, arbitrate the original control instruction according to the predictive safety instruction, and output the final control instruction to the driver of the motor drive device.

[0048] This application has at least the following beneficial effects:

[0049] This application proposes an innovative sensorless positioning and predictive control method. By analyzing the armature voltage and current signals generated during the operation of the motor-driven equipment, the rotational speed is estimated and integrated to construct a virtual mileage coordinate system. This converts the speed-coupled time-domain current signal into spatial load characteristics directly related to position. By matching the real-time acquired load characteristics with a pre-established track reference fingerprint map, precise absolute positioning can be achieved. Furthermore, based on the positioning results, features of the road segment ahead can be proactively retrieved. When a high-risk area is predicted, safety commands are actively generated to intervene in control.

[0050] This application utilizes the armature voltage and current signals inevitably generated during the operation of the motor drive equipment as an information source. Through signal processing and matching algorithms, it achieves track topology mapping, real-time precise vehicle positioning, and predictive safety control of road conditions ahead without requiring any additional dedicated external position sensors (such as encoders, cameras, or lidar). This not only directly eliminates external sensor hardware and its associated circuits, fundamentally reducing the system's material costs, power consumption, and physical complexity, but also improves reliability and environmental adaptability by reducing dependence on external environments (such as light and sign cleanliness) and additional mechanical structures.

[0051] Furthermore, this application constructs a virtual odometer coordinate system and load fingerprint map decoupled from speed through a "time-space transformation," transforming the positioning basis from a time-domain waveform easily affected by vehicle speed to a spatial feature map related only to the physical characteristics of the track. This innovative method ensures stable and accurate absolute position positioning through highly robust sequence matching, even under variable speed operating conditions such as vehicle acceleration and deceleration, overcoming the shortcomings of insufficient accuracy in simple time-domain analysis.

[0052] Building upon this foundation, this application extends beyond passive positioning to proactive safety prediction and intervention. Based on the real-time matched absolute position, the system can proactively retrieve load characteristics from the map within a predetermined distance ahead, intelligently predicting high-load-risk areas such as sharp bends, and automatically generating safety commands to proactively intervene in the original control commands. Especially in high-latency control scenarios (such as brain-computer interfaces and remote control), it can effectively compensate for command transmission delays, implementing deceleration or speed limits before potential dangers occur, thereby greatly improving operational safety and avoiding the risk of loss of control due to information lag or unknown road conditions, achieving a balance between intelligence and safety.

[0053] In summary, this application replaces the traditional hardware sensor path with a purely electrical software algorithm path, which enhances position perception and intelligent control functions while achieving multiple beneficial effects such as cost reduction, improved reliability, accurate positioning, and predictive safety control. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a schematic flowchart of a track topology mapping, positioning, and predictive control method based on electrical load fingerprinting provided in this embodiment;

[0056] Figure 2 The waveform of the load current is shown from the time domain perspective.

[0057] Figure 3 The waveform of the load current is shown from a spatial perspective.

[0058] Figure 4 This is a schematic diagram of a track topology mapping, positioning and prediction control system module based on electrical load fingerprinting provided in this embodiment.

[0059] Figure label:

[0060] 1-Signal acquisition module; 2-Data processing module; 3-Command arbitration module. Detailed Implementation

[0061] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0062] In the description of this application, it should be noted that the terms "vertical", "up", "down", "horizontal", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0063] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0064] Example 1

[0065] This embodiment provides a method for track topology mapping, positioning and predictive control based on electrical load fingerprinting, which is applied to motor-driven equipment (such as rail transit models (such as slot cars) or vehicles in industrial conveying systems) running on fixed tracks.

[0066] This method is based on a pre-established and stored track reference load fingerprint map, which defines the load distribution along the entire track in a "virtual mileage coordinate system." The virtual mileage coordinate system is a map representing the relationship between the load characteristics and position along the entire track. Specifically, it is an abstract position coordinate axis proportional to physical distance, constructed by estimating the rotational speed using the back electromotive force of an integrating motor. Its unit can be considered as a "virtual meter" or "virtual mileage unit." The establishment of this coordinate system essentially involves mapping the time-domain electrical response of the motor caused by load changes and coupled with speed (such as...). Figure 2 As shown), decouple and reproject onto a coordinate axis that is only related to spatial location (e.g. Figure 3 (As shown).

[0067] like Figure 1 As shown, this method includes the following steps:

[0068] S100. Real-time acquisition of armature voltage signal of motor in motor drive device. With armature current signal .

[0069] S200, based on armature voltage signal With armature current signal Estimate the real-time speed of the motor And perform time integration on the real-time rotational speed. The initial value of the integral can be set to zero. This refers to real-time virtual mileage, which represents the cumulative distance traveled by the vehicle in the virtual mileage coordinate system.

[0070] S300: Based on real-time virtual mileage, the armature current signal in the time domain is mapped to the virtual mileage coordinate system to form the armature current signal in the spatial domain, which is then used as the current load characteristic signal.

[0071] Among them, the time-domain current sequence collected in S100 Virtual odometer function generated using S200 Through resampling or mapping relationships Transformed into a signal with virtual mileage s as the independent variable. This refers to the armature current signal in the spatial domain. This signal serves as the current load characteristic signal used for matching. Because... Proportional to the physical distance, the characteristics of the signal (such as peak value, valley value, waveform profile) mainly depend on the physical load characteristics of the track (such as curvature, gradient, friction coefficient), and are basically decoupled from the instantaneous speed of the vehicle passing through the section.

[0072] S400. Perform sequence matching between the current load characteristic signal and the reference load fingerprint map to determine the absolute position of the motor drive equipment on the track. .

[0073] S500, based on absolute position The system extracts road segment features within a predetermined virtual mileage ahead (e.g., the estimated travel distance in the next second) from the baseline load fingerprint map. If the road segment features indicate a high-load risk area ahead (such as sharp bends or steep slopes), a predictive safety command is generated to intervene in the original control commands of the motor drive equipment. The road segment features include high / low load area information marked in the map.

[0074] This embodiment proposes an innovative sensorless positioning and predictive control method. By analyzing the armature voltage and current signals generated during the operation of the motor drive equipment, the rotational speed is estimated and integrated to construct a virtual mileage coordinate system. This converts the speed-coupled time-domain current signal into spatial load characteristics directly related to position. By matching the real-time acquired load characteristics with a pre-established track reference fingerprint map, precise absolute positioning can be achieved. Furthermore, based on the positioning results, features of the road segment ahead can be proactively retrieved. When a high-risk area is predicted, safety commands are actively generated to intervene in control.

[0075] This embodiment utilizes the armature voltage and current signals inevitably generated during the operation of the motor drive equipment as an information source. Through signal processing and matching algorithms, it achieves track topology mapping, real-time precise vehicle positioning, and predictive safety control of road conditions ahead without requiring any additional dedicated external position sensors (such as encoders, cameras, or lidar). This not only directly eliminates external sensor hardware and its associated circuits, fundamentally reducing the system's material costs, power consumption, and physical complexity, but also improves reliability and environmental adaptability by reducing dependence on external environments (such as light and sign cleanliness) and additional mechanical structures.

[0076] Furthermore, this embodiment constructs a virtual mileage coordinate system and load fingerprint map decoupled from speed through "time-space transformation," transforming the positioning basis from a time-domain waveform easily affected by vehicle speed to a spatial feature map related only to the physical characteristics of the track. This innovative method ensures stable and accurate absolute position positioning through highly robust sequence matching, even under variable speed operating conditions such as vehicle acceleration and deceleration, overcoming the shortcomings of insufficient accuracy in simple time-domain analysis.

[0077] Building upon this foundation, this embodiment extends beyond passive positioning to proactive safety prediction and intervention. Based on the real-time matched absolute position, the system can proactively retrieve load characteristics from the map within a predetermined distance ahead, intelligently predicting high-load-risk areas such as sharp bends, and automatically generating safety commands to proactively intervene in the original control commands. Especially in high-latency control scenarios (such as brain-computer interfaces and remote teleoperation), it can effectively compensate for command transmission delays, implementing deceleration or speed limits before potential dangers occur, thereby greatly improving operational safety and avoiding the risk of loss of control due to information lag or unknown road conditions, achieving a balance between intelligence and safety.

[0078] In summary, this embodiment replaces the traditional hardware sensor path with a purely electrical software algorithm path, which enhances position perception and intelligent control functions while achieving multiple beneficial effects such as reduced costs, improved reliability, accurate positioning, and predictive safety control.

[0079] In one embodiment, a baseline load fingerprint is pre-established through the following steps, which are performed upon initial use or track change:

[0080] Control the motor drive equipment to run at least one full revolution along a fixed track, preferably multiple revolutions (e.g., 3-5 revolutions) to collect data for averaging and suppress random noise.

[0081] During operation, the motor armature voltage signal is acquired in real time. With armature current signal And based on armature voltage signal With armature current signal Estimate the real-time speed of the motor .

[0082] For real-time rotational speed Perform time integration Earn virtual miles and based on virtual mileage armature current signal in the time domain Mapped to a virtual odometer coordinate system, this forms a spatial armature current signal. And serve as a characteristic signal for mapping load.

[0083] Record and process the mapping load characteristic signal sequence corresponding to the entire orbit, identify and label the high load characteristic segments and low load characteristic segments in the load characteristic signal sequence, thereby constructing a benchmark load fingerprint map that characterizes the load distribution and positional relationship throughout the entire orbit.

[0084] This involves identifying and labeling high-load and low-load characteristic segments in the load characteristic signal sequence to construct a benchmark load fingerprint map characterizing the load distribution and positional relationship throughout the orbit, including:

[0085] Load characteristic signals in the reference load fingerprint spectrum Perform smoothing processing. For example, use moving average or low-pass filtering to remove high-frequency noise.

[0086] Set load threshold and The load characteristic signal The amplitude remained higher than Continuous sections are marked as high-load characteristic sections (corresponding to curves, uphill sections, etc.); sections with amplitudes consistently below [a certain value] are [marked as high-load characteristic sections]. The sections are marked as low-load characteristic sections (corresponding to straight sections). Optionally, an adaptive peak detection algorithm can be used to automatically identify characteristic peaks.

[0087] After labeling The signal sequence, the start / end virtual mileage coordinates of each characteristic segment, and the corresponding risk level are associated and stored as a structured data file or database to form a benchmark load fingerprint map.

[0088] In one embodiment, estimating the real-time speed of the motor based on the armature voltage signal and the armature current signal includes:

[0089] A thermal model of the motor is established to characterize the dynamic process of temperature rise in the motor coil caused by the Joule heating effect of the armature current.

[0090] A back EMF observer is constructed based on the electrical equations of the motor. The electrical equations must include at least the relationship between armature voltage, armature current, armature resistance and back EMF.

[0091] Based on real-time acquired armature current signal The current temperature of the motor coil is estimated in real time using a thermal model. The armature resistance parameter value in the back EMF observer is compensated in real time based on the current temperature; specifically, the compensated armature resistance value is calculated using the following formula. :

[0092]

[0093] in, For the motor at the reference temperature The nominal internal resistance at (e.g., 25°C) This refers to the temperature coefficient of resistance of the motor coil material. This step eliminates the drift of internal parameters caused by the heat generated during long-term operation of the motor, thereby ensuring the long-term accuracy of back EMF observation and virtual mileage integration throughout the entire process and avoiding cumulative errors.

[0094] The armature voltage signal acquired in real time Armature current signal and the compensated armature resistance parameter value Input the back EMF observer to calculate the instantaneous value of the motor's back EMF. Specifically, the instantaneous value of the back electromotive force is calculated using the following formula. :

[0095]

[0096] Based on the instantaneous value of the back electromotive force With the back electromotive force constant of the motor The real-time rotational speed was calculated. The calculation formula is as follows:

[0097]

[0098] In one embodiment, the current load feature signal is sequentially matched with the reference load fingerprint spectrum, and the sliding window algorithm is used for matching.

[0099] The sliding window matching algorithm includes: acquiring a current load feature signal segment with a fixed virtual mileage length (e.g., corresponding to a physical distance of 0.2 meters) as the current matching segment; sliding the current matching segment along the virtual mileage coordinates of the benchmark load fingerprint map, and calculating the similarity between the current matching segment and the corresponding segment of the benchmark load fingerprint map at each sliding position; and determining the virtual mileage coordinates of the extreme point on the benchmark load fingerprint map when the similarity reaches an extreme value. This determines the absolute position of the motor-driven device.

[0100] In one embodiment, similarity is calculated using at least one of the following algorithms: cross-correlation, dynamic time warping, or minimum absolute difference (SAD). For example, the cross-correlation algorithm is used to calculate the correlation coefficient between two signals, with the maximum value corresponding to the best match. Alternatively, the minimum absolute difference (SAD) algorithm is used to calculate the sum of the absolute values ​​of the differences between corresponding points in the two signals, with the minimum value corresponding to the best match. The dynamic time warping (DTW) algorithm can also be used, which is suitable for situations where the two signals may have slight nonlinear deformations and exhibits strong robustness.

[0101] In one embodiment, the step of generating predictive safety instructions to intervene in the original control instructions of the motor drive device includes:

[0102] Define a look-ahead buffer associated with absolute position, and dynamically calculate safety command limits based on road segment characteristics within the look-ahead buffer; where safety command limits include, for example, maximum permissible speed or maximum permissible torque.

[0103] The system acquires the original control commands for the motor-driven device (from the user's controller, automatic cruise algorithm, etc.) and compares them with safety command limits. The original control commands are used to set the desired speed or torque of the motor.

[0104] If the original control command exceeds the safety command limit, the safety command limit will be output as the final command; otherwise, the original control command will be output as the final command.

[0105] This embodiment transforms perception and positioning information into proactive safety control to prevent speeding hazards caused by operator error or unknown road conditions, ensuring that vehicles operate within safe speed limits throughout the entire track.

[0106] In one embodiment, the original control commands originate from a higher-level control source that has a command transmission delay. For example, the higher-level control source with inherent delay characteristics is one of a brain-computer interface, a cloud-based remote control system, or a network teleoperation system.

[0107] The method also includes: monitoring the command transmission delay of the host control source. When instruction transmission is delayed If the delay exceeds a preset threshold (e.g., 500ms) and the road segment characteristics indicate that the area ahead is a high-load risk area, a safety arbitration will be performed.

[0108] Specifically, safety arbitration includes: temporarily shielding or attenuating the original control commands from the higher-level control source by a preset ratio, and replacing them with autonomously generated target safety commands to ensure the safe passage of the motor-driven equipment through the risk area. These target safety commands are then directly output to the motor driver as the final control commands, thereby completely offsetting any dangers that may arise from control delays. Specifically, the target safety command may be, for example, a fixed low-speed command or a smooth deceleration curve.

[0109] This embodiment compensates for the safety hazards caused by instruction lag in high-latency control scenarios (such as brain-computer interface racing cars and remote teleoperation). Through local, real-time predictive intervention, it ensures vehicle safety and greatly improves the usability and user experience of such applications.

[0110] In one embodiment, the method further includes: during the operation of the motor drive device, periodically (e.g., every N revolutions) or according to preset triggering conditions (e.g., the matching confidence level is continuously lower than a preset threshold) performing the following operations:

[0111] Based on the absolute position determined by S400 sequence matching, obtain the spectral segment corresponding to the current load characteristic signal;

[0112] By utilizing the current load characteristic signal, the corresponding spectral segment in the reference load fingerprint spectrum is updated to adapt the reference load fingerprint spectrum to long-term changes in orbital state, environment, or equipment parameters, thereby maintaining the accuracy of positioning and prediction.

[0113] This embodiment enables the system to adapt to slow changes in track conditions (such as wear and dust accumulation), slight degradation of motor performance, or changes in ambient temperature, allowing the reference fingerprint map to be updated in stages, maintaining long-term positioning and control accuracy, and reducing the need for maintenance and recalibration.

[0114] For example, data update processing can be weighted smoothing, as shown in the following formula:

[0115]

[0116] in, Let f(x) be the forgetting factor, satisfying the following relationship: .

[0117] For example, data update processing can also be incremental, where the old data is replaced by averaging multiple samples of real-time data.

[0118] Example 2

[0119] like Figure 4 As shown, this embodiment provides a track topology mapping, positioning, and predictive control system based on electrical load fingerprinting, applied to motor-driven equipment running on a fixed track. The system includes:

[0120] Signal acquisition module 1 is used to acquire the armature voltage signal and armature current signal of the motor drive device in real time;

[0121] Data processing module 2, connected to the signal acquisition module, is used to estimate the real-time speed of the motor based on the armature voltage and armature current signals, and to perform time integration on the real-time speed to obtain the real-time virtual mileage. Based on the real-time virtual mileage, the armature current signal in the time domain is mapped to the virtual mileage coordinate system to form the armature current signal in the spatial domain, which is used as the current load characteristic signal. The current load characteristic signal is matched with the reference load fingerprint map to determine the absolute position of the motor drive equipment on the track. Based on the absolute position, the road segment characteristics within the predetermined virtual mileage ahead are obtained from the reference load fingerprint map. If the road segment characteristics indicate that the area ahead is a high-load risk area, a predictive safety command is generated to intervene in the original control command of the motor drive equipment.

[0122] Command arbitration module 3, connected to the data processing module, is used to receive predictive safety commands and raw control commands from the motor drive device, arbitrate the raw control commands according to the predictive safety commands, and output the final control commands to the driver of the motor drive device.

[0123] It should be understood that each unit in the track topology mapping, positioning, and predictive control system disclosed in this embodiment is specifically used to execute any one of the track topology mapping, positioning, and predictive control methods in Embodiment 1 above. The specific execution content of each unit is consistent with the steps in any one of the track topology mapping, positioning, and predictive control methods in Embodiment 1 above. Please refer to the above track topology mapping, positioning, and predictive control methods to determine the specific execution content of each unit. For the sake of brevity, it will not be elaborated here.

[0124] Those skilled in the art will understand that the modules or steps described above in this application can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computing device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0125] Note that the above description is merely a preferred embodiment and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this application, and the scope of this application is determined by the scope of the appended claims.

[0126] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for track topology mapping, positioning, and predictive control based on electrical load fingerprinting, characterized in that, The method is applied to motor-driven equipment running on a fixed track. It is executed based on a pre-established reference load fingerprint map of the track, which is a map representing the load characteristics and position mapping relationship of the track throughout the entire track in a virtual mileage coordinate system. The method includes: Real-time acquisition of the motor armature voltage signal and armature current signal of the motor drive device; The real-time speed of the motor is estimated based on the armature voltage signal and the armature current signal, and the real-time virtual mileage is obtained by time integration of the real-time speed. Based on the real-time virtual mileage, the armature current signal in the time domain is mapped to the virtual mileage coordinate system to form the armature current signal in the spatial domain, which is then used as the current load characteristic signal. The current load characteristic signal is sequentially matched with the reference load fingerprint map to determine the absolute position of the motor drive device on the track. Based on the absolute position, road segment features within a predetermined virtual mileage ahead are obtained from the baseline load fingerprint map; if the road segment features indicate that the area ahead is a high load risk area, a predictive safety instruction is generated to intervene in the original control instructions of the motor drive device. Estimating the real-time speed of the motor based on the armature voltage and armature current signals includes: establishing a thermal model of the motor, which characterizes the dynamic process of motor coil temperature rise caused by the Joule heating effect of armature current; constructing a back electromotive force (EMF) observer based on the motor's electrical equations, which at least include the interrelationships between armature voltage, armature current, armature resistance, and back EMF; estimating the current temperature of the motor coil in real time using the thermal model based on the real-time acquired armature current signal, and compensating the armature resistance parameter value in the back EMF observer in real time according to the current temperature; inputting the real-time acquired armature voltage signal, armature current signal, and compensated armature resistance parameter value into the back EMF observer to calculate the instantaneous value of the motor's back EMF; and calculating the real-time speed based on the instantaneous value of the back EMF and the back EMF constant of the motor. The current load feature signal is sequentially matched with the reference load fingerprint map using a sliding window matching algorithm. The sliding window matching algorithm includes: acquiring a segment of the current load feature signal with a fixed virtual mileage length as the current matching segment; sliding the current matching segment along the virtual mileage coordinates of the reference load fingerprint map, and calculating the similarity between the current matching segment and the corresponding segment of the reference load fingerprint map at each sliding position; and determining the position in the reference load fingerprint map corresponding to the extreme value of the similarity as the absolute position of the motor drive device.

2. The track topology mapping, positioning, and predictive control method based on electrical load fingerprinting according to claim 1, characterized in that, The baseline load fingerprint is pre-established through the following steps: Control the motor drive device to complete at least one full revolution along the fixed track; During operation, the motor armature voltage signal and armature current signal are acquired in real time, and the real-time speed of the motor is estimated based on the armature voltage signal and armature current signal. The real-time rotational speed is integrated over time to obtain the virtual mileage, and the armature current signal in the time domain is mapped to the virtual mileage coordinate system based on the virtual mileage to form the armature current signal in the spatial domain, which is used as the mapping load characteristic signal. Record and process the mapping load feature signal sequence corresponding to the entire orbit, identify and label the high load feature segments and low load feature segments in the load feature signal sequence, thereby constructing the benchmark load fingerprint map that characterizes the load distribution and positional relationship throughout the orbit.

3. The track topology mapping, positioning, and predictive control method based on electrical load fingerprinting according to claim 1, characterized in that, The similarity is calculated using at least one of the following algorithms: cross-correlation algorithm, dynamic time warping algorithm, or minimum absolute difference algorithm.

4. The track topology mapping, positioning, and predictive control method based on electrical load fingerprinting according to claim 1, characterized in that, The step of generating predictive safety instructions to intervene in the original control instructions of the motor drive device includes: Define a look-ahead buffer associated with the absolute position, and dynamically calculate the safety instruction limit based on the road segment characteristics within the look-ahead buffer; Obtain the original control command of the motor drive device, and compare the original control command with the safety command limit; wherein, the original control command is used to set the desired speed or desired torque of the motor; If the original control command exceeds the safety command limit, the safety command limit will be output as the final command; otherwise, the original control command will be output as the final command.

5. The method for track topology mapping, positioning, and predictive control based on electrical load fingerprinting according to claim 1, characterized in that, The original control commands originate from a higher-level control source with a command transmission delay; The method further includes: Monitor the instruction transmission delay of the upper-level control source; When the instruction transmission delay exceeds a preset delay threshold, and the road segment characteristics indicate that the area ahead is a high-load risk area, a safety arbitration is performed; The safety arbitration includes: temporarily blocking or attenuating the original control commands from the upper-level control source according to a preset ratio, and replacing them with autonomously generated target safety commands for enabling the motor drive equipment to safely pass through the risk area.

6. The track topology mapping, positioning, and predictive control method based on electrical load fingerprinting according to claim 5, characterized in that, The upper-level control source with instruction transmission delay is one of a brain-computer interface, a cloud-based remote control system, or a network teleoperation system.

7. The track topology mapping, positioning, and predictive control method based on electrical load fingerprinting according to claim 1, characterized in that, The method further includes performing the following operations periodically or according to preset triggering conditions during the operation of the motor drive device: Based on the absolute position determined by the sequence matching, obtain the spectral segment corresponding to the current load characteristic signal; Using the current load characteristic signal, the corresponding spectrum segment in the benchmark load fingerprint spectrum is updated. The data update process includes incremental updates or weighted smoothing.

8. A track topology mapping, positioning, and predictive control system based on electrical load fingerprinting, characterized in that, A system for use with motor-driven equipment that runs on a fixed track, the system comprising: The signal acquisition module is used to acquire the motor armature voltage signal and armature current signal of the motor drive device in real time; A data processing module, connected to the signal acquisition module, is used to estimate the real-time speed of the motor based on the armature voltage signal and armature current signal, and to perform time integration on the real-time speed to obtain the real-time virtual mileage; based on the real-time virtual mileage, the armature current signal in the time domain is mapped to the virtual mileage coordinate system to form a spatial armature current signal, which is used as the current load characteristic signal; the current load characteristic signal is matched sequentially with a reference load fingerprint map to determine the absolute position of the motor drive device on the track; based on the absolute position, the road segment characteristics within a predetermined virtual mileage ahead are obtained from the reference load fingerprint map; if the road segment characteristics indicate a high-load risk area ahead, a predictive safety command is generated to intervene in the original control command of the motor drive device; The instruction arbitration module, connected to the data processing module, is used to receive the predictive safety instruction and the original control instruction from the motor drive device, arbitrate the original control instruction according to the predictive safety instruction, and output the final control instruction to the driver of the motor drive device. The method for estimating the real-time speed of the motor based on the armature voltage and armature current signals includes: establishing a thermal model of the motor, which characterizes the dynamic process of temperature rise in the motor coil caused by the Joule heating effect of the armature current; constructing a back electromotive force (EMF) observer based on the motor's electrical equations, which at least include the interrelationships between armature voltage, armature current, armature resistance, and back EMF; estimating the current temperature of the motor coil in real time using the thermal model based on the real-time acquired armature current signal, and compensating the armature resistance parameter value in the back EMF observer in real time based on the current temperature; inputting the real-time acquired armature voltage signal, armature current signal, and compensated armature resistance parameter value into the back EMF observer to calculate the instantaneous value of the back EMF of the motor; and calculating the real-time speed based on the instantaneous value of the back EMF and the back EMF constant of the motor. Specifically, the current load feature signal is sequentially matched with the reference load fingerprint map using a sliding window matching algorithm. The sliding window matching algorithm includes: acquiring a segment of the current load feature signal with a fixed virtual mileage length as the current matching segment; sliding the current matching segment along the virtual mileage coordinates of the reference load fingerprint map, and calculating the similarity between the current matching segment and the corresponding segment of the reference load fingerprint map at each sliding position; and determining the position in the reference load fingerprint map corresponding to the extreme value of the similarity as the absolute position of the motor drive device.

Images