Control method and system of intelligent railway robot

Abstract

The invention relates to the technical field of robot control, in particular to a control method and system of an intelligent railway robot. Comprising the following steps: S1, acquiring track identification data of a railway robot; the track identification data comprises single-device identification data and multi-device identification data; s2, sorting the single equipment identification data according to equipment operation turnover time to obtain a first sorting result; sorting the single device identification data according to a collaborative combination historical frequency to obtain a second sorting result; and S3, acquiring normal operation time and emergency operation time of the railway robot, and determining a first control method of the railway robot according to the first sorting result. According to the method, through double-sequence optimization scheduling, dynamic priority calculation, screening of high-frequency and high-stability combinations during faults and self-adaptive switching modes, full-process autonomous control of track damage identification is realized, and the operation efficiency of the railway robot and the reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the field of robot control technology, and in particular to a control method and system for an intelligent railway robot. Background Art

[0002] Current railway inspections mainly rely on robots controlled by humans or fixed programs, which have significant defects: traditional robot control methods are difficult to dynamically adapt to the complex scenarios of track damage identification, especially when multiple devices are working together, there is a lack of efficient scheduling mechanisms. Equipment response delays lead to insufficient timeliness of identification under emergency conditions, and single equipment failures will further reduce the robustness of the system. Existing technologies do not effectively distinguish between the differences in demand during normal and emergency operation periods, nor do they fully utilize the time series characteristics and combination frequency patterns of historical equipment identification data, resulting in low resource utilization and rigid response strategies. In addition, the contradiction between the accuracy and real-time nature of track damage identification is prominent. There is an urgent need for a control method that can autonomously optimize equipment scheduling priorities and dynamically adapt to changes in working conditions to improve the operating efficiency and system stability of railway robots in complex environments. Summary of the Invention

[0003] In order to overcome the shortcomings of low equipment dispatching efficiency and insufficient emergency response, the present invention provides a control method and system for an intelligent railway robot.

[0004] The technical implementation scheme of the present invention is: a control method of an intelligent railway robot, comprising the following steps:

[0005] S1: Acquire track identification data of the railway robot; the track identification data includes single-device identification data and multi-device identification data;

[0006] S2: sorting the single device identification data according to the equipment operation turnaround time to obtain a first sorting result; sorting the single device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result;

[0007] S3: Obtain the normal operation time and emergency operation time of the railway robot, and determine a first control method for the railway robot based on the first sorting result; obtain the equipment operation status of the railway robot during the emergency operation time, and determine a second control method for the railway robot based on the second sorting result;

[0008] S4: Determine a final control method for the railway robot based on the first control method and the second control method.

[0009] Preferably, the obtaining of track identification data of the railway robot includes:

[0010] Obtain control data and identification data of each device during the operation of the railway robot;

[0011] Associate the control data with the identification data along the time axis to form N control sequences

[0012] Based on the control sequence, N track identification data are obtained.

[0013] Preferably, the track identification data includes single-device identification data and multi-device identification data, including:

[0014] If the track damage is only independently identified by a single device in the railway robot, the track identification data is defined as single device identification data;

[0015] If the track damage is identified by multiple devices in the railway robot, the track identification data is defined as multi-device identification data.

[0016] Preferably, sorting the single device identification data according to the device operation turnaround time to obtain a first sorting result includes:

[0017] Based on the control sequence, extracting a time difference between the control data and the identification data in the control sequence and defining it as a first time difference;

[0018] At the same time, extracting the time difference between the identification data and the next control data in the control sequence and defining it as a second time difference;

[0019] The single device identification data is sorted from small to large based on the sum of the first time difference and the second time difference to obtain a first sorting result.

[0020] Preferably, sorting the single device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result includes:

[0021] taking a set of collaborative devices for the same damage event in the multi-device identification data as a group;

[0022] Based on historical data, extract the number of occurrences of each of the combinations;

[0023] The multi-device identification data is sorted from high to low according to the number of occurrences to obtain a second sorting result.

[0024] Preferably, the obtaining of the normal operation time and the emergency operation time of the railway robot and determining the first control method of the railway robot based on the first sorting result includes:

[0025] Extracting an average stable time interval of each of the single device identification data based on the first sorting result;

[0026] If the railway robot is in emergency operation time, the calling priority of each device is calculated based on the device selection formula, and the first control method is generated based on the device with the highest calling priority.

[0027] Preferably, if the railway robot is in emergency operation time, the calling priority of each device is calculated based on the device selection formula, and the first control method is generated based on the device with the highest calling priority, including: the device selection formula is as follows,

[0028]

[0029] in, For devices The calling priority, For devices The average stable time interval is the arithmetic mean of the stable time intervals for a single device to complete a track damage identification in historical operations, reflecting the comprehensive response efficiency of the device. The stable time interval is the sum of the first time difference between the issuance of the device control command and the completion of the identification and the second time difference between the completion of the identification and the start of the next command. is the basic stability coefficient, is the emergency reinforcement coefficient, is the emergency factor, , During normal operating hours, During emergency operation period, For equipment The average stable time interval, is the total number of available devices.

[0030] Preferably, the obtaining of the equipment operation status of the railway robot during the emergency operation time and determining the second control method of the railway robot based on the second sorting result includes:

[0031] In an emergency operation, if a missing combination appears in the multi-device identification data, determining a device control method according to the second sorting result;

[0032] The second control method of the railway robot is determined by the multi-device selection formula; the multi-device selection formula is as follows:

[0033]

[0034] in, For combination Emergency call priority, For combination The historical recognition frequency of For combination The average stabilization time is the arithmetic mean of the average stabilization time intervals of all member devices in a multi-device combination, which characterizes the overall response efficiency of the combination. is the frequency weight coefficient, is the time stability weight, and is a faulty device indicator, is the total number of available combinations.

[0035] Preferably, determining the final control method of the railway robot based on the first control method and the second control method includes:

[0036] If the current operation time is normal, the first control method is directly used to determine that a single device performs track damage identification;

[0037] If the current operation is in emergency time and all devices are operating normally, the first control method is used to select the single device with the highest priority to perform rapid identification;

[0038] If the current operation is in emergency operation time and there is equipment failure, the second control method is activated to select the optimal multi-equipment combination to perform damage identification;

[0039] The equipment selection priority calculation result is used as the execution basis to control the railway robot to perform the corresponding recognition task;

[0040] Finally, the control instructions are output to the railway robot actuator to complete the track damage identification operation.

[0041] The control system of the intelligent railway robot includes:

[0042] The data acquisition module is used to obtain the track recognition data of the railway robot, including single-device recognition data and multi-device recognition data;

[0043] a sorting processing module, configured to sort the single-device identification data according to the equipment operation turnaround time to obtain a first sorting result, and to sort the multi-device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result;

[0044] A control decision module is used to determine a first control method according to normal operation time and emergency operation time, and to determine a second control method according to the equipment operation status during the emergency operation time;

[0045] The execution output module is used to determine the final control method based on the first control method and the second control method, and output the control instruction to the railway robot actuator to complete the track damage identification operation.

[0046] Beneficial effects: The present invention generates the first and second sorting results respectively through the time sequence sorting of single-device identification data and the frequency sorting of multi-device combinations, thereby optimizing the equipment scheduling logic; based on the division of normal and emergency operation periods, the priority of a single device is dynamically calculated in combination with the equipment selection formula to ensure that the optimal device is quickly called under emergency conditions; when equipment fails, a high-frequency and high-stability combination is screened through a multi-device selection formula to ensure system redundancy; the final control method adaptively switches the single / multi-device mode according to the operation time and equipment status, outputs instructions to drive the actuator, and realizes full-process autonomous control of track damage identification, greatly improving the operation efficiency and system reliability of the railway robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of the control method of the intelligent railway robot of the present invention;

[0048] Figure 2 This is a structural diagram of the control system of the intelligent railway robot of the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] Example 1: Control method of intelligent railway robot, such as Figure 1 As shown, the following steps are included:

[0051] S1: Acquire track identification data of the railway robot; the track identification data includes single-device identification data and multi-device identification data;

[0052] S2: sorting the single device identification data according to the equipment operation turnaround time to obtain a first sorting result; sorting the single device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result;

[0053] S3: Obtain the normal operation time and emergency operation time of the railway robot, and determine a first control method for the railway robot based on the first sorting result; obtain the equipment operation status of the railway robot during the emergency operation time, and determine a second control method for the railway robot based on the second sorting result;

[0054] S4: Determine a final control method for the railway robot based on the first control method and the second control method.

[0055] Obtain track recognition data for railway robots, including:

[0056] Obtain control data and identification data of each device during the operation of the railway robot;

[0057] Associate the control data with the identification data along the time axis to form N control sequences

[0058] Based on the control sequence, N track identification data are obtained.

[0059] It should be noted that control data: command timestamps and operating parameters are obtained in real time from the robot actuator (such as the robotic arm controller); identification data: damage type, location and detection time are synchronously collected through track scanning equipment (such as laser sensors).

[0060] The control instructions and recognition results of the same operation cycle are associated along the time axis to form a closed-loop sequence of control → recognition; each sequence records a single complete operation, and N sequences constitute the original data set for track recognition.

[0061] For example, control data: {Command: Start crack scan, time: 08:00:00}; identification data: {Damage type: Rail crack, Location: K25+300, Time: 08:00:03}; → Composition control sequence: [Start scan] → [Crack @ K25+300] (takes 3 seconds); 1000 operations cumulatively form 1000 sequences (N=1000).

[0062] Track identification data includes single-device identification data and multi-device identification data, including:

[0063] If the track damage is only independently identified by a single device in the railway robot, the track identification data is defined as single device identification data;

[0064] If the track damage is identified by multiple devices in the railway robot, the track identification data is defined as multi-device identification data.

[0065] It should be noted that based on the difference in the number of damage identification devices, track identification data is divided into two categories: single-device identification data: a single device independently completes damage detection (such as only a laser sensor identifies a crack); multi-device identification data: multiple devices collaboratively verify the same damage (such as a laser sensor and an image recognition system simultaneously confirming a crack); this classification provides a data basis for subsequent single-device priority scheduling and multi-device combination optimization.

[0066] For example (connecting to the crack detection scenario mentioned above), single-device data: only the laser sensor identifies a crack at K25+300 → the sequence is marked as single-device identification data; multi-device data: the same crack is identified by both the laser sensor (08:00:03) and the image recognition system (08:00:05) → a multi-device control sequence is formed: [Laser: Crack @ K25+300] + [Image: Crack @ K25+300].

[0067] Sorting the single device identification data according to the device operation turnaround time to obtain a first sorting result, including:

[0068] Based on the control sequence, extracting a time difference between the control data and the identification data in the control sequence and defining it as a first time difference;

[0069] At the same time, extracting the time difference between the identification data and the next control data in the control sequence and defining it as a second time difference;

[0070] The single device identification data is sorted from small to large based on the sum of the first time difference and the second time difference to obtain a first sorting result.

[0071] It should be noted that sorting optimization is achieved by quantifying equipment operating efficiency indicators: the first time difference is the time from the issuance of a control command to the completion of damage identification (reflecting the equipment's response speed); the second time difference is the interval from the completion of damage identification to the initiation of the next command (reflecting the equipment's idle state); and the stable time interval is defined as the sum of these two time differences. The shorter the interval, the higher the equipment turnover efficiency. Sorting rules: sort by stable time interval from smallest to largest (i.e., prioritize short-interval, high-efficiency equipment), generating the first sorting result.

[0072] The equipment operation turnover time refers to the total time it takes for the equipment to complete an operation cycle, which is equivalent to the stable time interval, including the sum of the response time from the issuance of the control instruction to the completion of the recognition and the idle time from the completion of the recognition to the start of the next instruction.

[0073] Example (Continued Crack Detection Scenario): Laser sensor operation sequence: First time difference: Command issued (08:00:00) → Recognition completed (08:00:03) = 3 seconds; Second time difference: Recognition completed (08:00:03) → Next command (08:00:18) = 15 seconds; Stable interval = 3 + 15 = 18 seconds. Ultrasonic probe sequence: Interval = 8 seconds (response) + 5 seconds (idle) = 13 seconds. Sorting result: The ultrasonic probe (13 seconds) takes precedence over the laser sensor (18 seconds), and the device with the shorter interval receives higher scheduling priority.

[0074] Sorting the single device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result includes:

[0075] taking a set of collaborative devices for the same damage event in the multi-device identification data as a group;

[0076] Based on historical data, extract the number of occurrences of each of the combinations;

[0077] The multi-device identification data is sorted from high to low according to the number of occurrences to obtain a second sorting result.

[0078] It should be noted that for multi-device collaboration scenarios, optimized sorting is achieved through historical combination frequency statistics: combination definition: the identification data of multiple devices in the same damage event are bundled into a collaborative unit (such as the "laser + image" combination); frequency extraction: the number of times the same device combination appears in historical data is counted, and high-frequency combinations indicate that the collaborative model is mature and stable; sorting rules: sort by the number of occurrences from high to low, and give priority to scheduling high-frequency combinations to improve collaborative reliability.

[0079] The collaborative combination historical frequency refers to the number of times a combination of multiple devices collaboratively identify the same damage event in historical data.

[0080] Example (Continued Crack Detection Scenario): Combination A (Laser + Image): 5 cumulative crack detections at locations such as K25+300 and K30+100 → Frequency = 5; Combination B (Laser + Ultrasonic): 2 cumulative crack detections at locations such as K28+500 → Frequency = 2; Second sorting result: Combination A (5 detections) takes precedence over Combination B (2 detections). The high-frequency combination is used for urgent collaborative work. Note: Frequency statistics are based on a historical control sequence library. For example, Combination A is accumulated from five independent sequences (e.g., [Laser: 08:00:03, Image: 08:00:05]).

[0081] Obtaining the normal operation time and the emergency operation time of the railway robot, and determining a first control method of the railway robot based on the first sorting result, including:

[0082] Extracting an average stable time interval of each of the single device identification data based on the first sorting result;

[0083] If the railway robot is in emergency operation time, the calling priority of each device is calculated based on the device selection formula, and the first control method is generated based on the device with the highest calling priority.

[0084] It should be noted that a dynamic scheduling strategy is established based on the results of single device sorting: extract the stable time interval of each device (the sum of the first time difference and the second time difference); calculate the arithmetic mean of the device's historical intervals ( ); In case of emergency operation, the priority of each device is calculated by the device selection formula; the first control method is generated based on the device with the highest priority.

[0085] Example (connecting ultrasound probe and laser sensor data): Ultrasound probe historical interval mean: = 13 seconds (previous example: 13 seconds); historical interval average of the laser sensor: = 18 seconds (previous example 18 seconds); Emergency operation scenario: Detect train approaching ( ), substitute into the equipment selection formula: ,parameter: =1.5 (basic stability), =3.0 (urgent enhancement), output: ≈1.8*10 4 , 18 4.5 →After normalization =0.86; Scheduling decision: The laser sensor receives the highest priority due to its high stability and performs fast recognition.

[0086] If the railway robot is in emergency operation time, the calling priority of each device is calculated based on the device selection formula, and the first control method is generated based on the device with the highest calling priority, including: the device selection formula is as follows,

[0087]

[0088] in, For devices The calling priority, For devices The average stable time interval is the arithmetic mean of the stable time intervals for a single device to complete a track damage identification in historical operations, reflecting the comprehensive response efficiency of the device. The stable time interval is the sum of the first time difference between the issuance of the device control command and the completion of the identification and the second time difference between the completion of the identification and the start of the next command. is the basic stability coefficient, is the emergency reinforcement coefficient, is the emergency factor, , During normal operating hours, During emergency operation period, For devices The average stable time interval, is the total number of available devices.

[0089] It should be noted that the core principle of the equipment selection formula is the dynamic weighted index normalization, which is used to strengthen the scheduling priority of high-stability equipment during emergency operations: Numerator: The average stable time interval of the equipment is the base, the exponent Constitute dynamic weight. The larger the value, the more stable the historical response of the device (for example, the laser sensor interval of 18 seconds is better than the ultrasound probe interval of 13 seconds). (foundation stability factor) and (Emergency Enhancement Coefficient) is a preset parameter, Controls the stability weight of normal operations (default =1.5), In an emergency ( =1) Significantly amplify high Advantages of the equipment (such as = 3.0 increases the index from 1.5 to 4.5). Denominator: for all equipment Sum and normalize to ensure ∈[0,1] and . : Emergency status factor (0: normal period; 1: emergency period), automatically triggered by the system clock matching the preset working period.

[0090] Parameter acquisition method: : Calculate the average equipment operation turnaround time from historical control sequences (e.g. laser sensor: 18 seconds); , : Adjust the preset by actual scene (normal operation =1.5, in case of emergency + =4.5); :According to the time Match preset time period (e.g. 22:00-6:00) and (Such as train approaching warning period).

[0091] Example, scenario: Railway robots need to urgently detect track cracks. Available equipment includes: laser sensors ( = 18 seconds); Ultrasonic probe ( =13 seconds); Parameters: =1.5, =3.0, =1 (emergency state), = 2. Priority calculation: molecular calculation: laser sensor: 18 （1.5+3.0*1） =18 4.5 ≈1.1*10 5 ; Ultrasound probe: 13 4.5 ≈1.8*10 4 Denominator: 1.1*10 5 +1.8*10 4 =1.28*10 5 ; Normalization: =1.1*10 5 / 1.28*10 5 ≈0.86; =0.14; Decision: In emergency, the laser sensor has high stability ( The advantage of the large sensor (large) is exponentially amplified, with 86% priority allocated for rapid identification, while the faster-responding ultrasound probe only has 14%. This means that in emergencies, single-shot speed is sacrificed in favor of long-term device stability to reduce the risk of failure.

[0092] Obtaining the equipment operation status of the railway robot during the emergency operation time, and determining a second control method for the railway robot based on the second sorting result, including:

[0093] In an emergency operation, if a missing combination appears in the multi-device identification data, determining a device control method according to the second sorting result;

[0094] The second control method of the railway robot is determined by the multi-device selection formula; the multi-device selection formula is as follows:

[0095]

[0096] in, For combination Emergency call priority, For combination The historical recognition frequency of For combination The average stabilization time is the arithmetic mean of the average stabilization time intervals of all member devices in a multi-device combination, which characterizes the overall response efficiency of the combination. is the frequency weight coefficient, is the time stability weight, and For faulty equipment indicators, is the total number of available combinations.

[0097] It should be noted that the core principle of the multi-device selection formula is weighted scoring-fault filtering-dynamic normalization, which is used to select high-frequency and high-stability equipment combinations in emergency failures: Numerator: Calculation combination Comprehensive score , integrating historical frequencies (reliability) and mean stabilization time (timeliness), weight 、 Preset (such as =0.7, =0.3). Denominator: Sum the total scores of the available combinations (excluding combinations with faulty devices), (Fault indicator, the value is 1 when the faulty device exists) to achieve filtering, and take max(1,∑) to avoid zero division error. Fault filter item: the end Directly eliminate the combination containing faulty equipment (if the combination If there is a faulty device, =0). Normalization: Final ∈[0,1], the larger the value, the higher the priority.

[0098] Parameter acquisition method: :Combination of statistical historical data The number of occurrences (e.g. combination A appears 5 times); : Calculate the average stabilization time of the devices in the combination (e.g., laser + image combination: (18 + 15) / 2 = 16.5 seconds); : Real-time monitoring of device status (normal = 0, fault = 1); 、 : Preset according to business needs (frequency priority > ).

[0099] Example, scenario: An ultrasonic probe fails during an emergency operation, and a multi-device combination is needed to identify rail cracks. Available combinations: Combination A (laser + image): =5, = 16.5 seconds; Combination B (Laser + Ultrasound): including faulty equipment ( =1); Parameters: =0.7, =0.3, =1, =1 (only combination A is available). Priority calculation: Numerator: combination A score = 0.7×5+0.3×16.5=3.5+4.95=8.45; Denominator: only combination A is valid (combination B is filtered due to a fault), total score = 8.45, take max(1,8.45)=8.45; Fault filtering: combination A has no fault ( =0); Result: =8.45 / 8.45×(1-0)=1.0; Decision: Combination B is excluded because it contains a faulty ultrasound probe ( = 0), combination A is scheduled for collaborative identification with 100% priority. This formula simultaneously implements three optimizations: frequency weighting priority (combination A has been called 5 times in history); troubleshooting combinations (combination B fails); and dynamic normalization to ensure the only available combination responds with full priority.

[0100] Determining a final control method for the railway robot based on the first control method and the second control method includes:

[0101] If the current operation time is normal, the first control method is directly used to determine that a single device performs track damage identification;

[0102] If the current operation is in emergency time and all devices are operating normally, the first control method is used to select the single device with the highest priority to perform rapid identification;

[0103] If the current operation is in emergency operation time and there is equipment failure, the second control method is activated to select the optimal multi-equipment combination to perform damage identification;

[0104] The equipment selection priority calculation result is used as the execution basis to control the railway robot to perform the corresponding recognition task;

[0105] Finally, the control instructions are output to the railway robot actuator to complete the track damage identification operation.

[0106] It should be noted that the final control method dynamically switches strategies based on the operating mode and equipment status: Normal operation: Call the first control method (equipment selection formula), and select a single device with high stability to perform identification; Emergency and no fault: Enable the emergency enhancement mode of the first control method, and dispatch the single device with the highest stability (based on the equipment selection formula, give priority to devices with a large average stable time interval); Emergency fault: Trigger the second control method, and select the optimal available combination based on the multi-device selection formula; Finally, the priority calculation result is used to drive the actuator to complete damage identification and realize full-scenario adaptive control.

[0107] Example (connecting laser sensor / ultrasound probe and combination A data), scenario 1 (normal operation): system time → Press Dispatch: Laser sensor ( =0.59) → Execute command: Start laser single device scanning; Scenario 2 (emergency without fault): And the equipment is normal → Emergency strengthening formula output =0.66 (Laser) → Execute command: Emergency start laser fast identification (Note: Laser Large and stable); Scenario 3 (emergency failure): Ultrasonic probe failure during emergency operation → Second control method output =1.0 (Combination A: Laser + Image) → Execute command: Enable combination A collaborative recognition @K25+300; Execute closed loop: The command is transmitted to the robot controller, driving the robotic arm / sensor to complete the track damage recognition operation.

[0108] Example 2: Based on Example 1, the control system of the intelligent railway robot is as follows: Figure 2 Shown, including:

[0109] The data acquisition module is used to obtain the track recognition data of the railway robot, including single-device recognition data and multi-device recognition data;

[0110] a sorting processing module, configured to sort the single-device identification data according to the equipment operation turnaround time to obtain a first sorting result, and to sort the multi-device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result;

[0111] A control decision module is used to determine a first control method according to normal operation time and emergency operation time, and to determine a second control method according to the equipment operation status during the emergency operation time;

[0112] The execution output module is used to determine the final control method based on the first control method and the second control method, and output the control instruction to the railway robot actuator to complete the track damage identification operation.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A control method for an intelligent railway robot, characterized in that: The following steps are involved: S1: Acquire track identification data of the railway robot; the track identification data includes single-device identification data and multi-device identification data; S2: sorting the single device identification data according to the equipment operation turnaround time to obtain a first sorting result; sorting the single device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result; S3: Obtaining the normal operation time and emergency operation time of the railway robot, and determining a first control method of the railway robot based on the first sorting result; Obtaining equipment operation status of the railway robot during emergency operation time, and determining a second control method for the railway robot based on the second sorting result; S4: Determine a final control method for the railway robot based on the first control method and the second control method.

2. The control method of the intelligent railway robot according to claim 1, characterized in that: The obtaining of the track recognition data of the railway robot includes: Obtain control data and identification data of each device during the operation of the railway robot; Associate the control data with the identification data along the time axis to form N control sequences Based on the control sequence, N track identification data are obtained.

3. The control method of the intelligent railway robot according to claim 1, wherein: The track identification data includes single-device identification data and multi-device identification data, including: If the track damage is only independently identified by a single device in the railway robot, the track identification data is defined as single device identification data; If the track damage is identified by multiple devices in the railway robot, the track identification data is defined as multi-device identification data.

4. The control method of the intelligent railway robot according to claim 2, characterized in that: Sorting the single device identification data according to the device operation turnaround time to obtain a first sorting result includes: Based on the control sequence, extracting a time difference between the control data and the identification data in the control sequence and defining it as a first time difference; At the same time, extracting the time difference between the identification data and the next control data in the control sequence and defining it as a second time difference; The single device identification data is sorted from small to large based on the sum of the first time difference and the second time difference to obtain a first sorting result.

5. The control method of the intelligent railway robot according to claim 2, wherein: Sorting the single device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result includes: taking a set of collaborative devices for the same damage event in the multi-device identification data as a group; Based on historical data, extract the number of occurrences of each of the combinations; The multi-device identification data is sorted from high to low according to the number of occurrences to obtain a second sorting result.

6. The control method of the intelligent railway robot according to claim 1, characterized in that: The obtaining of the normal operation time and the emergency operation time of the railway robot and determining the first control method of the railway robot based on the first sorting result includes: Extracting an average stable time interval of each of the single device identification data based on the first sorting result; If the railway robot is in emergency operation time, the calling priority of each device is calculated based on the device selection formula, and the first control method is generated based on the device with the highest calling priority.

7. The control method of the intelligent railway robot according to claim 6, characterized in that: If the railway robot is in emergency operation time, the calling priority of each device is calculated based on the device selection formula, and the first control method is generated based on the device with the highest calling priority, including: the device selection formula is as follows, in, For devices The calling priority, For devices The average stable time interval is the arithmetic mean of the stable time intervals for a single device to complete a track damage identification in historical operations, reflecting the comprehensive response efficiency of the device. The stable time interval is the sum of the first time difference between the issuance of the device control command and the completion of the identification and the second time difference between the completion of the identification and the start of the next command. is the basic stability coefficient, is the emergency reinforcement coefficient, is the emergency factor, , During normal operating hours, During emergency operation period, For devices The average stable time interval, is the total number of available devices.

8. The control method of the intelligent railway robot according to claim 1, wherein: The obtaining of the equipment operation status of the railway robot during the emergency operation time and determining the second control method of the railway robot based on the second sorting result includes: In an emergency operation, if a missing combination appears in the multi-device identification data, determining a device control method according to the second sorting result; The second control method of the railway robot is determined by the multi-device selection formula; the multi-device selection formula is as follows: in, For combination Emergency call priority, For combination The historical recognition frequency of For combination The average stabilization time is the arithmetic mean of the average stabilization time intervals of all member devices in a multi-device combination, which characterizes the overall response efficiency of the combination. is the frequency weight coefficient, is the time stability weight, and is a faulty device indicator, is the total number of available combinations.

9. The control method of the intelligent railway robot according to claim 1, wherein: The determining of a final control method of the railway robot based on the first control method and the second control method includes: If the current operation time is normal, the first control method is directly used to determine that a single device performs track damage identification; If the current operation is in emergency time and all devices are operating normally, the first control method is used to select the single device with the highest priority to perform rapid identification; If the current operation is in emergency operation time and there is equipment failure, the second control method is activated to select the optimal multi-equipment combination to perform damage identification; The equipment selection priority calculation result is used as the execution basis to control the railway robot to perform the corresponding recognition task; Finally, the control instructions are output to the railway robot actuator to complete the track damage identification operation.

10. A control system for an intelligent railway robot, for implementing the control method for an intelligent railway robot according to any one of claims 1 to 9, characterized in that: include: The data acquisition module is used to obtain the track recognition data of the railway robot, including single-device recognition data and multi-device recognition data; a sorting processing module, configured to sort the single-device identification data according to the equipment operation turnaround time to obtain a first sorting result, and to sort the multi-device identification data according to the historical frequency of the collaborative combination to obtain a second sorting result; A control decision module is used to determine a first control method according to normal operation time and emergency operation time, and to determine a second control method according to the equipment operation status during the emergency operation time; The execution output module is used to determine the final control method based on the first control method and the second control method, and output the control instruction to the railway robot actuator to complete the track damage identification operation.

Images